North from Mexico: the Spanish-speaking people of the United States [Third edition] 9781440836824, 9781440849855, 9781440836831, 1440836833

Cover page; Halftitle page; Title page; Copyright page; Dedication; Contents; Foreword; Introduction to the Third Editio

1,179 68 3MB

English Pages [474] Year 2016

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

North from Mexico: the Spanish-speaking people of the United States [Third edition]
 9781440836824, 9781440849855, 9781440836831, 1440836833

Table of contents :
Cover page
Halftitle page
Title page
Copyright page
Dedication
Contents
Foreword
Introduction to the Third Edition by Alma M. García
Introduction
1 In Spanish Saddlebags
1. The Spanish Prologue
2. Footnote to the Prologue
3. The Play-in-Prose
4. The Climate of Spain
2 The Fantasy Heritage
1. The Man on the White Horse
2. The Birth of a Legend
3. De Anza Doesn't Live Here Any More
3 The Fan of Settlement
1. "Sunshine, Silence, and Adobe"
2. The Forgotten Link
3. Lands of the Spanish-Speaking
4. Mexico Is Not Europe
5. The Border of the Borderlands. 4 Heart of the Borderlands1. A Lost World
2. The People
3. The Flowering of New Mexico
4. After the Conquest
5 The Broken Border
1. Pimeria Alta
2. The Tejanos
3. The Californios
4. Lost Provinces
6 "Not Counting Mexicans"
1. Los Diablos Tejanos
2. Alas! The Alamo
3. The Mexican-American War
4. Slaves and Peons
5. "Red Robber of the Rio Grande"
6. "The Dead-Line of Sheriffs"
7. The Salt War
8. Open Season on Mexicans
7 Gringos and Greasers
1. The Bloodless Conquest
2. The Ghost of Father Martínez
3. Disturbance in Socorro
4. The Don System
5. The Buffer State. 6. The Lynching of Juanita7. Birth of a Stereotype
8 The Heritage of theSouthwest
1. Mr. Marshall's Chispa
2. Comstock's Mistake
3. The Vermilion Cave
4. Anglo-Saxon Law and Order
5. Apaches and Copper
6. Homage to the Churro
7. Los Pastores
8. From Gregorio de Villalobos
9. "Ten Gallon Hats"
10. Cortez Had a Brand
11. A Drop of Water
9 The Borderlands Are Invaded
1. Spanish Trails, American Rails
2. Life in a Boxcar
3. Cotton in Texas
4. Cotton Moves West
5. Vitamins and Mexicans
6. Coyotes and Man-Snatchers
7. Los Betabeleros
8. In Midwest Industries. 9. The Balance Sheet10 The Second Defeat
1. The Myth of Docility
2. The Honeymoon Is Over
3. The Gallup Incident
4. In the Copper Mines
5. La Niña de Cabora
6. The Forty Blonde Babies
7. The Battle of Cananea
8. "The Wearers of the Red"
11 "The Mexican Problem"
1. The Structure of the Problem
2. The Buffer Group
3. The Conflict in Cultures
4. The Pattern of Employment
5. The Colonia Complex
6. The Northern Settlements
7. Que Maravilla!
12 The Pattern of Violence
1. The Case of Sleepy Lagoon
2. Captain Ayres: Anthropologist
3. Plotting a Riot
4. The Origin of Pachuquismo. 13 Blood on the Pavements1. The Taxicab Brigade
2. Operation "Dixie"
3. When The Devil Is Sick . . .
4. The Strange Case of the Silk Panties
5. The Politics of Prejudice
14 The War Years
1. Joe Martinez and Company
2. A Tear for José Davilla
3. Across the Border
4. Los Braceros
5. The Counterpoint of Migration
6. Good Neighbors and Band Music
15 After a Hundred Years
1. A Beginning Is Made
2. Grassroots Democracy
3. The Westminster Case
4. "Utilizable Cultural Residues"
5. From De Anza to Juan López
16 "One and Together"
1. By Any Other Name
2. Words That Fit.

Citation preview

North from Mexico

North from Mexico The Spanish-Speaking People of the United States Third Edition by Carey McWilliams second edition updated by Matt. S. Meier third edition updated by Alma M. García Foreword by Mario T. García

Copyright © 1948 by Carey McWilliams. Copyright renewed 1975 by Carey McWilliams. Updated material copyright © 1990 by Matt S. Meier. Updated material copyright © 2016 by Alma M. García. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, except for the inclusion of brief quotations in a review, without prior permission in writing from the publisher. Library of Congress Cataloging-in-Publication Data Names: McWilliams, Carey, 1905–1980. | Meier, Matt S. | García, Alma M. Title: North from Mexico : the Spanish-speaking people of the United States /   by Carey McWilliams ; second edition updated by Matt. S. Meier ; third edition   updated by Alma M. García; foreword by Mario García. Description: Third edition. | Santa Barbara, California : Praeger, an imprint of ABC-CLIO, LLC,   2016. | “This book is also available on the World Wide Web as an eBook”—Title page verso. Identifiers: LCCN 2015046909 | ISBN 9781440836824 (hardback : acid-free paper) |   ISBN 9781440849855 (paperback : acid-free paper) | ISBN 9781440836831 (electronic) Subjects: LCSH: Mexican Americans—History. | Mexicans—United States—History. |   Immigrants—United States—History. | Southwest, New—History. | United States—   Ethnic relations—History. | United States—Emigration and immigration—History. |   Mexico—Emigration and immigration—History. | BISAC: SOCIAL SCIENCE /   Ethnic Studies / Hispanic American Studies. Classification: LCC E184.M5 M394 2016 | DDC 973/.046872—dc23 LC record available at http://lccn.loc.gov/2015046909 ISBN: 978-1-4408-3682-4 (cloth)     978-1-4408-4985-5 (paper) EISBN: 978-1-4408-3683-1 20 19 18 17 16  1 2 3 4 5 This book is also available on the World Wide Web as an eBook. Visit www.abc-clio.com for details. Praeger An Imprint of ABC-CLIO, LLC ABC-CLIO, LLC 130 Cremona Drive, P.O. Box 1911 Santa Barbara, California 93116-1911 This book is printed on acid-free paper Manufactured in the United States of America

For Iris. . . .

Contents

Foreword by Mario T. García

xiii

Introduction to the Third Edition by Alma M. García

xvii

Introduction xix   1 In Spanish Saddlebags 1. The Spanish Prologue—2. Footnote to the Prologue—3. The Play-in-Prose—4. The Climate of Spain

1

  2 The Fantasy Heritage 1. The Man on the White Horse—2. The Birth of a Legend—3. De Anza Doesn’t Live Here Any More

15

  3 The Fan of Settlement 1. “Sunshine, Silence, and Adobe”—2. The Forgotten Link—3. Lands of the Spanish-Speaking—4. Mexico Is Not Europe—5. The Border of the Borderlands

27

  4 Heart of the Borderlands 1. A Lost World—2. The People—3. The Flowering of New Mexico—4. After the Conquest

41

  5 The Broken Border 1. Pimeria Alta—2. The Tejanos—3. The Californios—4. Lost Provinces

57

  6 “Not Counting Mexicans” 1. Los Diablos Tejanos—2. Alas! The Alamo—3. The MexicanAmerican War—4. Slaves and Peons—5. “Red Robber of the Rio

71

viii  Contents

Grande”—6. “The Dead-Line of Sheriffs”—7. The Salt War—8. Open Season on Mexicans   7 Gringos and Greasers 1. The Bloodless Conquest—2. The Ghost of Father Martínez—3. Disturbance in Socorro—4. The Don System—5. The Buffer State—6. The Lynching of Juanita—7. Birth of a Stereotype

87

  8 The Heritage of the Southwest 1. Mr. Marshall’s Chispa—2. Comstock’s Mistake—3. The Vermilion Cave—4. Anglo-Saxon Law and Order—5. Apaches and Copper—6. Homage to the Churro—7. Los Pastores—8. From Gregorio de Villalobos—9. “Ten Gallon Hats”—10. Cortez Had a Brand—11. A Drop of Water

103

  9 The Borderlands Are Invaded 1. Spanish Trails, American Rails—2. Life in a Boxcar—3. Cotton in Texas—4. Cotton Moves West—5. Vitamins and Mexicans—6. Coyotes and Man-Snatchers—7. Los Betabeleros—8. In Midwest Industries—9. The Balance Sheet

127

10 The Second Defeat 1. The Myth of Docility—2. The Honeymoon Is Over—3. The Gallup Incident—4. In the Copper Mines—5. La Niña de Cabora—6. The Forty Blonde Babies—7. The Battle of Cananea—8. “The Wearers of the Red”

149

11 “The Mexican Problem” 1. The Structure of the Problem—2. The Buffer Group—3. The Conflict in Cultures—4. The Pattern of Employment—5. The Colonia Complex—6. The Northern Settlements—7. Que Maravilla!

163

12 The Pattern of Violence 1. The Case of Sleepy Lagoon—2. Captain Ayres: Anthropologist—3. Plotting a Riot—4. The Origin of Pachuquismo

181

13 Blood on the Pavements 1. The Taxicab Brigade—2. Operation “Dixie”—3. When the Devil Is Sick . . .—4. The Strange Case of the Silk Panties—5. The Politics of Prejudice

195

Contents   ix

14 The War Years 1. Joe Martinez and Company—2. A Tear for José Davilla—3. Across the Border—4. Los Braceros—5. The Counterpoint of Migration—6. Good Neighbors and Band Music

209

15 After a Hundred Years 1. A Beginning Is Made—2. Grassroots Democracy—3. The Westminster Case—4. “Utilizable Cultural Residues”—5. From De Anza to Juan López

223

16 “One and Together” 1. By Any Other Name—2. Words That Fit—3. Neighbors in Isolation—4. Who Is Being Stubborn?—5. The Indelible Imprint—6. “The Sun Has Exploded”

235

17 Chicano Leadership and Organization by Matt S. Meier

249

1. Effects of World War II—2. Demographic Changes—3. Organizational Development—4. Community Service Organization—5. American G.I. Forum—6. California MAPA—7. A National Organization?—8. Leadership in the Protest Movement—9. Man of the Migrants: César Chávez—10. Champion and Crusader: Rodolfo “Corky” Gonzales—11. King Tiger: Reies López Tijerina—12. A Student Leader: José Angel Gutiérrez—13. Summary 18 Politics, Education, and Culture by Matt S. Meier 1. Chicano Powerlessness—2. Government Response—3. The Chicano Student Movement—4. Student Movement Results—5. Other Chicano Organizations—6. National Organizations—7. Chicanas—8. Chicana Liberation—9. Chicana Organizations—10. Chicanas in Business and Government—11. Chicano Culture: Alive and Well—12. Literature—13. Journals and Magazines—14. Chicano Music and Art—15. Spanish Language Use—16. Bilingualism—17. The Movimiento: Radical or Reformist?—18. The Movimiento: Success or Failure?—19. Undocumented Immigrants

263

x  Contents

19 North from Mexico by Matt S. Meier

287

1. North from Mexico: Three Waves—2. Undocumented Entrance—3. Interim Bracero Hiring—4. The El Paso and Tivoli Incidents—5. The Second Bracero Program—6. Operation “Wetback”—7. U.S. Immigration Policy—8. The Quota System—9. Refugee Legislation—10. Hart-Cellars Act—11. Permanent Visa Mexican Immigration—12. Green Carders—13. Commuters—14. Undocumenteds—15. Simpson-Mazzoli: Simpson-Rodino—16. Results of the Simpson-Rodino Law 20 A Demographic Profile of the Mexican-Born Population in the United States by Alma M. García

311

1. The Nexus of Education, Occupation, and Poverty—2. New Destination States: Emergent Mexican Immigrant Communities—3. Finding Religious Space in the New South—4. Unauthorized Immigrants Living in the United States—5. Looking Toward the Future 21 Still Coming North from Mexico: Immigration Constraints and Contestations by Alma M. García

329

1. Maquiladoras: A Plan for Economic Development—2. Demonization of Mexican Immigrants: The Context for the Immigration Act of 1990—3. The Immigration Act of 1990—4. California’s Proposition 187 of 1994—5. Unauthorized Immigration: Persistent Struggles—6. Immigrants, Dual Nationality, and Voting in Mexican Elections—7. The Immigrant Spring of 2006: Mass Demonstrations—8. Gendered Journeys: Mexican Immigrant Women and Transnationalism—9. The Border: Mexican Women Confront Violence 22 Mexican Immigrants in a Changing Society by Alma M. García 1. Children of Mexican Immigrants: The Second Generation—2. Undocumented Students: Court Cases and Legislation—3. The DREAMers: Struggles and Mass Mobilization—4. The 2008

361

Contents   xi

Presidential Race and Latinos—5. Arizona’s Senate Bill 1070: Criminalization of Immigrants—6. Latinos, the 2012 Presidential Race, and Beyond—7. Immigration and the Deferred Action for Childhood Arrivals (DACA) and Its Impact on the 2012 Presidential Election—8. Latinos, the Issues, and the 2012 Presidential Race—9. Assessing DACA—10. Obama’s Executive Order—11. Social Changes and the Future of Mexican Immigrants and Their U.S.-born Children

Epilogue by Alma M. García 393 Acknowledgments 397 Acknowledgments to the New Edition by Alma M. Garcia 399 Notes on Sources 401 Additional Notes on Sources 403 by Matt S. Meier Chapter Notes 407 Index 429

Foreword Mario T. García

I received my MA in history at the University of Texas at El Paso (UTEP) in 1968. I tried to get a community college job in different parts of the country but to no avail. One day in early spring of 1969, while looking at the job listings at the student center, I saw a notice that some recruiters from San Jose State College (later San Jose State University) were coming to campus and were interested in interviewing students with an MA for possible teaching positions on their campus. I don’t think there was a reference to Chicano candidates, but I signed up and had an interview. Little did I know that Chicano students at San Jose State, as part of the emerging Chicano movement, had demonstrated for Chicano studies classes and Chicano professors. To my surprise, about a month later, I received a letter offering me an appointment as a lecturer in history to commence that fall. I would teach four introduction classes to U.S. history each semester. Not having any other job possibilities, I eagerly accepted, even though it meant leaving my hometown and my family. I looked forward to this new opportunity. However, a few weeks later, the chair of the History Department wrote to me again and asked if I would like to do a Chicano history course. In fact, I would replace two of the U.S. history classes that fall with two Chicano history classes. I was intrigued about taking this on and agreed to do so. The Chicano movement, influenced by the struggle of César Chávez and migrant farmworkers plus urban unrest led by Chicano students, had not really reached El Paso. I knew something about all of this, including the land-grant movement in nearby northern New Mexico led by Reies López Tijerina, but was not very well versed in the movement. More importantly, although I had studied the history of Mexico and U.S. Western history, I didn’t know much about what constituted Chicano history. Still, I had no choice but to look forward to the summer when I would try to piece together enough material to teach one of the first Chicano history courses in the country.

xiv   Foreword

But this dramatically changed later that spring when I met Professor Ramón Eduardo Ruiz, from Smith College in New England, who had been invited to give a lecture at UTEP. I went to his lecture on the U.S.-Mexico War and the history faculty went out of their way to introduce me to Ruiz, a MexicanAmerican born in San Diego. Meeting Ruiz and spending some time with him was a huge break. I would later work on my PhD with Ruiz as my mentor at the University of California at San Diego, but on this occasion, when I told him that I was going to teach two Chicano history classes that fall and that I didn’t know where I would find materials, he said to me, “Read Carey McWilliams’s North from Mexico.” I had never heard of McWilliams or his book. Taking Ruiz’s advice, I learned that McWilliams had first published his book in 1948 and that it had just been republished in a new hardback edition by Greenwood Press. I ordered a copy of the book, which I still have. This was a game changer! McWilliams, to my surprise, had written the first history of MexicanAmericans in the United States, or what he called the Spanish-speaking people of the country. His coverage went from the Spanish colonial period of the Southwest to just after World War II. Here were my lectures! Here was my contextualization and periodization of how to teach a Chicano history course. It was not comprehensive and didn’t cover the later period of the Chicano movement, but it was a gold mine as far as I was concerned. That summer of 1969 I spent all of my time taking copious notes from North from Mexico and trying to organize the material into specific lectures. McWilliams gave me the Spanish borderlands, the U.S.-Mexico War, the postMexican War period, the beginning of mass immigration from Mexico in the early twentieth century, the effects of the Great Depression and MexicanAmerican labor upheavals, the Zoot Suit Riots, Mexican-American participation in World War II, and the immediate postwar conditions. This became the organization for my lectures. I had never encountered this history before. I never had a Chicano history class at UTEP because there were none until the early 1970s. For my generation of Chicano historians, McWilliams was essential. It became our bible. It was a lifesaver, or at least a professional lifesaver. There was literally nothing else we could use to teach the initial Chicano history courses. North from Mexico is a classic, and McWilliams is the godfather of Chicano history. Contemporary Chicano historiography begins with McWilliams. For my graduate course on Chicano historiography, I have the students first read North from Mexico. The text not only influenced those of us who had the responsibility to teach the early Chicano history courses but also influenced the early themes of Chicano historiography. What themes? While it does not commence Chicano history with the indigenous populations in Mexico and the Southwest, it does stress the importance of the Spanish colonial period as being complex and heterogeneous, including the mixing of Spanish and Indians

Foreword   xv

in the Spanish borderlands or El Norte. Recognizing the results of mestizaje that produced mixed peoples and mixed or hybrid cultures, McWilliams set down the foundation of Chicano history. His discussion of the U.S.-Mexico War (1846–1848) as a war of Anglo-American manifest destiny, expansion, and conquest would resonate with the Chicano movement, and early Chicano historians took up this theme and expanded it to introduce the theory of “internal colonialism” or the idea that Chicanos first began their U.S. experience as a conquered people and hence represented a colonized group within the “belly of the beast.” The uprooting and racist treatment of the first generation of Mexican-Americans, or what I call the “conquered generation,” in the Chicano history that McWilliams treats also introduces what later scholars of race call “racialization” or the political and economic roots of racism. This racialization is further applied to the first large wave of Mexican immigrants in the early twentieth century, and indeed, it is institutionalized through “Mexican jobs,” “Mexican wages,” “Mexican barrios,” and the infamous or bad “Mexican schools.” McWilliams doesn’t use the term “racialization” but later Chicano historians would tease it out of his text. Chicano historians would be further affected by McWilliams’s discussion of the Mexican Revolution of 1910 and its impact on these early Mexican immigrants. The Mexican immigrant strikes, especially in the fields and mines of the Southwest in the 1920s and 1930s, motivated Chicano historians who sought a radical historical heritage and the history of working-class Mexicans in the United States. In his autobiographical treatment of the 1943 Zoot Suit Riots in Los Angeles, McWilliams helped Chicano historians and Chicano Movement activists rediscover the pachucos and zoot-suiters of that era as countercultural warriors, as did many in the Chicano Movement. These are only some, although perhaps the key, influences North from Mexico has had on early Chicano historiography, and it is still important today. The fact is that this classic text is still very relevant, as my graduate students attest, and it is also a prophetic text. McWilliams prophesized the emergence of Mexican-American political power. He lived to see the Chicano movement of the late 1960s and early 1970s but did not live long enough to see today’s unprecedented Latino political power, which has recently helped to elect U.S. presidents. But he was also prophetic in his treatment of such themes as Mexican immigration, labor and community struggles, and the continued racialization of Chicanos and Latinos, which are still realities today. We can understand the roots of these manifestations by reading McWilliams. This is why it has been so important to make North from Mexico available to new generations of readers of Chicano history. The republication of the Greenwood Press 1968 edition was fundamental, as noted. The reissuing of a new edition in 1990 by Praeger and skillfully updated by Professor Matt S. Meier of Santa Clara University, including the addition of new chapters to

xvi   Foreword

cover the periods after the 1948 publication was noteworthy and a great contribution by Professor Meier. In that same tradition, Professor Alma M. García, also of Santa Clara University, has expertly arranged the newest edition of this classic text with a new update and important analysis, which is equally welcomed and of great importance in further introducing North from Mexico to the millennial generation. I am honored to write this foreword to a text from an incredible writer and historian, Carey McWilliams, who pioneered the history of what is now the largest minority in the United States and whose history is essential for all Americans to know about. It is also an honor because of the significant influence McWilliams had on my own development as a Chicano historian. I thank Professor García and ABC-CLIO Press for this opportunity to say my own gracias to Carey McWilliams. Mario T. García is Professor of Chicano Studies and History at the University of California, Santa Barbara. He is the author of numerous books on Chicano history, such as Desert Immigrants: The Mexicans of El Paso, 1880–1920, Mexican Americans: Leadership, Ideology & Identity, 1930–1960; Memories of Chicano History: The Life and Narrative of Bert Corona; Católicos: Resistance and Affirmation in Chicano Catholic History; Blowout! Sal Castro and the Chicano Struggle for Educational Justice; The Chicano Generation: Testimonios of the Movement; and The Latino Generation: Voices of the New America.

Introduction to the Third Edition by Alma M. García In the introduction to the first edition of his North from Mexico: The SpanishSpeaking People of the United States, Carey McWilliams explains why he selected the first part of the book’s title. McWilliams wanted it to “refer to a process, a movement.”1 He strove to craft a narrative of the history of Mexican immigrants to the United States as one that would capture the dynamic movement of a people settling in a new country. North from Mexico analyzed the journey of Mexican immigrants to a “brave new world” with all its struggles and triumphs. In his earlier publication, Factories in the Field: The Story of Migratory Farm Labor in California, McWilliams saw himself as a writer who would uncover the “hidden history” of Mexicans in the fields, the factories and all the other sites of conflicts experienced by Mexican immigrants. Similarly, in his updated second edition to North From Mexico, Matt Meier, himself a pioneer of Chicano history, followed in McWilliams’s tradition by continuing the story of Mexican immigrants and their descendants. I am humbled to have been asked to write the update for the third edition of this classic work. In keeping with the intellectual spirit of McWilliams and Meier, my update of North From Mexico traces sociohistorical and political movement of Mexican immigrants and their descendants by focusing on key events and topics. Twenty-five years have passed since Meier’s update and, as such, it would be impossible to provide a comprehensive analysis of Mexican immigration from 1986 to 2016. The overarching goal of this update is to present a panoramic history that highlights key developments needed to understand the lives of Mexican immigrants in the United States, ones that continue to be characterized by the struggles and triumphs of an immigrant group and their U.S.-born children; this group has always formed an integral part of U.S. history. My update consists of three chapters, each of which focuses on key areas in the study of Mexican immigrants. The rest of Carey McWilliams’ classic work, including the 1990 update by Matt Meier, has been left as is. The foreign-born population of Mexicans living in the United States has experienced exponential

xviii  Introduction to the Third Edition by Alma M. García

growth over the period since Meier’s update. Such growth has contributed to increased diversity in U.S. society. Nevertheless, some upward mobility among Mexican immigrants and later generations has been eclipsed by ongoing inequalities in access to higher education, limited access to occupations outside the service sector, and stagnant incomes. Over the years since the last edition of North From Mexico, border violence against Mexican immigrant women has increased, and this violence has now taken an anti-lesbian tenor. A series of state initiatives and congressional legislation, such as California’s 1994 Proposition 187 and Arizona’s 2010 Senate Bill 1070, has reflected widespread anti-immigrant sentiment. In response, Mexicans joined in collective protest against these efforts. The mass mobilization and social protests that developed during the “immigrant spring of 2006” represented a direct and sustained contestation of the 2005 Border Protection, Antiterrorism and Illegal Immigration and Control Act (H.R. 4437). Similarly, collective actions of protest erupted as unauthorized Mexican immigrant youth, who had entered the United States with their undocumented parents, came together to engage in demonstrations. In addition, these unauthorized youth created various types of organizations to address the specific and immediate concerns as individuals who had lived most of their lives in the United States—not Mexico—but were forced to live in the shadows due to their marginalized status as noncitizens. A gradual increase in voter participation occurred due to the growth of naturalized Mexicans and their U.S.-born children and later generations of Mexican-Americans who formed an eligible pool of voters. Mexicans and other Latino groups began to be seen as an increasingly powerful political group; the Latino vote surged in importance in the presidential elections of 2008 and 2012 and promises to do so in the upcoming election of 2016. Mexican immigrants will continue to move “North from Mexico” and establish families with U.S.-born children. I believe Carey McWilliams and Matt Meier would agree with me that Mexicans living in the United States—both authorized and unauthorized—will continue to confront both expected and unexpected ruptures in the fabric of American society. But they will struggle, survive, and triumph as they move to make the United States live up it its narrative of the American Dream where immigrants are indeed welcomed arrivals. In 1951, Oscar Handlin, in his classic history of immigrants, The Uprooted: The Epic Story of the Great Migrations that Made American People, made the now oft-quoted assertion that “immigrants were American History.”2 It should be remembered that three years earlier, in 1948, Carey McWilliams, in his classic North from Mexico: The Spanish-Speaking People of the United States, already viewed the immigrant saga as one representing this main trope of American society. McWilliams emphasized that the immigrant story was one of “struggles and conflicts.” It is my hope that my update to this classic work will do justice to the memory of Carey McWilliams by continuing the unfolding story of the thousands upon thousands who came from Mexico to the United States.

Introduction

Since North from Mexico was first published in 1950, there has been a new burst of interest in Mexican-Americans, which, in large part, has come about as a result of activities and developments for which they themselves are responsible. Historically, the Spanish-speaking have often complained that little is known about them (which is true) and that their problems have received little attention by the larger American public (which is also true). Dr. George Sanchez, a distinguished spokesman for the Spanish-speaking, once referred to them as “an orphan group, the least known, the least sponsored, and the least vocal large minority group in the nation.” In the same vein, Representative Edward R. Roybal of California, himself of Mexican-American descent, has said that “the Mexican population of the Southwest . . . is little known on the East Coast and not much better understood in the Southwest itself.” A Mexican-American was quoted in Newsweek as saying, “We’re the best kept secret in America.” But this is certainly no longer the case. Suddenly the nation has discovered Mexican-Americans. Witness the feature stories about them in The Wall Street Journal (May 3, 1966), Newsweek (May 23, 1966), Time (April 28, 1967), U.S. News & World Report (June 6, 1966), and other publications. In this introduction to the reprint edition, I have sought to summarize certain important developments since 1950 which account for this new burst of interest and concern. Without attempting to fix an arbitrary date, it can be said that the Spanishspeaking began to develop a new political awareness and self-consciousness as a minority in the wake of World War II and, more noticeably, since the early 1950s, when the Negro civil rights movement began to emerge. The growing political maturity of Mexican-Americans, which became evident during World War II, was given a powerful impetus in the 1960 presidential campaign in which Viva Kennedy clubs sprang up throughout the Southwest, with the aid and encouragement of the late President Kennedy. This was the first time that the Spanish-speaking vote had figured prominently in a presidential election. It was this vote, enlarged by an active registration campaign, that probably saved Texas for the Kennedy-Johnson ticket. As a mark of recognition, President

xx  Introduction

Kennedy appointed Raymond Telles, who had served as mayor of El Paso, as U.S. ambassador to Costa Rica. Other developments, at about the same time, helped set the stage for new efforts at self-advancement among the Spanish-speaking. The legislation under which Mexican farm labor had been imported by agreement between that country and the United States was finally permitted to expire in December 1964. During the years of the labor importation, or bracero, program (1942– 1964), resident Mexicans had carried on an active agitation against it on the ground that the imported workers constituted a form of unfair labor competition. Once the bracero program was terminated and the “wetback” influx had been checked, it became possible, in theory at least, to organize Mexican farmworkers. Previously, the annual influx of imported farmworkers and “wetbacks” had so thoroughly demoralized the labor market that any thought of organization was unrealistic. The fact that Mexican immigration was placed on a quota basis in 1965 also helped to stabilize the labor market. But there are still many “green card holders,” those who live in Mexico but hold permits which make it possible for them to commute to jobs on the American side of the border. On September 8, 1965, the eight-month-long “grape pickers strike” began in Delano, California, under the brilliant leadership of César Chávez, who, incidentally, had received some training in organizational techniques when he was associated with the Community Service Organization. In no small measure, the success of the strike was due to the manner in which Chávez converted it into a “strike of families” by basing the union structure on the strong Mexican family structure. The strike quickly became a national news story, extensively covered by all the media, and succeeded in winning the support of many church and civic organizations. The culmination of the strike was a long march from Delano to Sacramento—the state capital. The story of the strike, and the march, is told in an excellent pamphlet issued by the Farm Workers Press, entitled Basta! La Historia de Nuestra Lucha or Enough! The Tale of Our Struggle. A passage from the text, illustrated with superb photographs, reads as follows: We are conscious of the historical significance of our Pilgrimage. It is clearly evident that our path travels through a valley well known to all Mexican farmworkers. We know all of these towns of Delano, Fresno, Madera, Modesto, Stockton and Sacramento, because along this very same road, in this very same valley the Mexican race has sacrificed itself for the last hundred years.

At about the same time that the Delano grape pickers were marching into Sacramento, the Equal Rights Opportunity Commission decided to hold a public hearing in Albuquerque, New Mexico, on March 28, 1966. At this hearing, the entire Mexican-American delegation of fifty or sixty persons walked

Introduction   xxi

out in protest against the disinterested, condescending way in which the hearing was being conducted. This walkout has assumed great symbolic importance in the eyes of the “new breed” of Mexican-American leaders. To them it symbolizes the “coming of age,” politically and socially, of the Spanish-speaking of the Southwest. They have referred to it as a “second El Grito,” a reference to the famous Grito de Dolores—“the cry of Dolores”—with which Father Miguel Hidalgo y Costilla launched the revolt against Spain in Dolores, Mexico, on September 16, 1810. The Los Angeles delegates who had walked out of the Albuquerque hearing were honored, as heroes, at a banquet in Los Angeles on April 28, 1966, at which they wore tiny huaraches as proud symbols of their participation in the walkout. The walkout had wide repercussions throughout Spanish-speaking communities in the Southwest. Like the strike of the grape pickers at Delano, it was a sign that Mexicans had had “enough.” Then Mexican farmworkers in Texas, under the leadership of Eugene Nelson, who had been identified earlier with the Delano activities (see The Nation, June 5, 1967, and also the article by Doug Adair, The Nation, December 11, 1967), called a strike on June 15, 1966, which, like the California strike, culminated in a long march from the Lower Rio Grande Valley to Corpus Christi and then on to the state capital. This dramatic 400-mile march terminated on September 4, 1966, when forty weary and sun-blackened Mexican farmworkers marched into Austin shouting “Huelga!” or “Strike,” and drew cheers from 8,000 enthusiastic supporters. “La Marcha,” in Texas, is said to have awakened the conscience of the one and a half million Mexicans in that state as nothing has done in their history. What these recent developments mean—the strikes in California and Texas and the walkout at Albuquerque—is that the Mexican has resumed his neverentirely-abandoned struggle to achieve dignity, respect, and equality—or, in a word, full citizenship. To no small extent, these activities have been stimulated by, and patterned after, the kinds of marches and demonstrations that Negroes and their allies in the civil rights movement have organized. Politically the Spanish-speaking vote, if it could ever be properly organized, would be of national significance. Accurate estimates are not possible, but there are probably six million Mexicans in the United States today and perhaps a total of nine or ten million Spanish-speaking, which would include 1.5 million Puerto Ricans in New York, a large Cuban contingent in Florida, quite a few Latin Americans, and, of course, some immigrants from Spain. There has been little contact or interchange between the other Spanish-speaking and the Mexicans of the Southwest, but such contacts are now being established. In estimating the future significance of the Spanish-speaking vote, it should be kept in mind that the young age brackets are heavily represented. Marvin Alisky, for example, estimates that by 1975 the Spanish-speaking may well total twenty million (The Reporter, February 9, 1967). Even today the advances

xxii  Introduction

which Mexicans have made politically are quite impressive. Senator Joseph R. Montoya, of New Mexico, and Representatives Edward R. Roybal (30th Congressional District, California), Henry Gonzales (20th Congressional District, Texas), and Eligion De La Garza (15th Congressional District, Texas) serve in Congress. Four Mexican-Americans serve in the Arizona state legislature, one in Colorado, thirty-three in New Mexico, and ten in Texas. But not one Mexican serves in the California state legislature, although there are nearly two million Mexicans in the state—an indication of the extent to which the Spanish-speaking are still unrepresented. Mexicans also serve today on city councils and boards of education, and here and there, one will be found serving as a judge (Judge Cal Córdova of Maricopa County, Arizona) or county commissioner (Albert A. Pena Jr., in Texas). Mexicans are making political gains all the time, and the White House no longer needs to be reminded that the Spanish-speaking have attained a new political maturity. By way of making amends for the Albuquerque hearing, President Johnson appointed Vicente Ximenes as the first American of Mexican descent to serve on the Equal Employment Opportunity Commission and, in White House ceremonies on June 9, 1967, released the report of a Cabinet committee entitled: “The Mexican American—A New Focus on Opportunity.” On this same occasion, the President mentioned some of his recent Spanish-speaking appointments: Raul H. Castro, of Arizona, as Ambassador to El Salvador; Benigno C. Hernandez, of New Mexico, as Ambassador to Paraguay; followed by a long listing of MexicanAmericans recently appointed to various federal positions. This growing political influence of the Spanish-speaking reflects new educational and economic gains and the rise of a new middle class. Among Mexican males in the Southwest today, some 29,229 are engaged in professional, technical, and similar work, either as employees or as self-employed persons: 3,761 as teachers in elementary and secondary schools, 404 in colleges and universities, 3,044 as engineers, 2,357 as designers and draftsmen, 2,140 as auditors and accountants, 1,866 as technicians, 1,291 as clergymen, 1,405 as musicians and music teachers, and, of course, quite a number as doctors, lawyers, and social workers. Not many Mexicans have yet achieved recognition as writers and artists, but there is a wealth of talent in the group that will some day—and fairly soon—burst forth on the national scene. Occupational differentiation is still restricted, but the spread of employment is increasing. To measure the distance that must still be traveled, consider these facts: out of approximately 25,000 students at Berkeley (University of California), there are only 78 Mexican students—this is 1967—and at the University of California at Los Angeles, with an estimated enrollment of 26,000, there are only 70 Mexican-American students, about three in every thousand. This in a community that has close to a million Mexican residents! In 1966 there

Introduction   xxiii

were more Negroes registered in state colleges in California than there were Mexicans, although there are more Mexicans living in the state than there are Negroes. More impressive than statistics, though, is the quality of the new leadership that is emerging. In the past, leadership was in short supply. Too many Mexican-American leaders were of the “Uncle Tom” or “Tío Tomás” variety— the type who tried to improve their own lot by selling out the interests of the rank-and-file Mexicans they were supposed to represent. The Mexicans of the border have a phrase which describes this type: “El es un zero a la izquierda”— “He is a left-handed zero”—that is, a pompous flunky who is trying to act like a big shot. But these Uncle Toms are being rapidly succeeded today by young, honest, intelligent, progressive, well-educated types. Literally hundreds of these young men, in the twenty-five to thirty-five age category, have emerged as leaders. Many of them are college educated, with impressive war records, and they hold all manner of jobs—as social workers, in the antipoverty program, and in city, county, and state employment. They are articulate, bilingual, and thoroughly familiar with the background and the problems of the Spanish-speaking. At Berkeley, for example, the young Mexican-American graduate students have formed a most interesting organization known as Quinto Sol, which has issued some excellent studies about the problems of the Spanish-speaking. And Ralph Guzman, son of a Mexican migrant worker, is today Assistant Director of the Mexican-American Study Project at the University of California at Los Angeles, which was set up in February 1964, on the basis of a large grant received from the Ford Foundation. It has issued a series of studies about Mexican-Americans and has succeeded in focusing a great deal of attention on the problems of the Spanish-speaking. The new generation of Mexican “activists” has formed some impressive action organizations: MAPA, the Mexican-American Political Action organization; PUMA, Political Unity for Mexican-Americans; and PASO, Political Association of Spanish-Speaking Organizations. One may conclude, therefore, and with confidence, that the Spanishspeaking are rapidly achieving a new political maturity. One does not need to worry over much at the moment about the “slowness” with which the Mexican is being “assimilated”; he is moving forward with a new vigor and confidence and with a new, strong sense of purpose and direction. —Carey McWilliams New York, 1968

1

In Spanish Saddlebags

The Hispanic heritage of the Southwest has two parts: the Spanish and the Mexican-Indian. Originally one heritage, unified in time, they have long since been polarized. Carefully distinguished from the Mexican-Indian, the Spanish heritage is now enshrined throughout the Southwest. It has become the sacred or templar tradition of which the Mexican-Indian inheritance is the secular or profane counterpart. In a somewhat similar fashion, the Spanish part of the tradition has also been divided into two parts. The romantic-heroic side has been accepted and enshrined; the prosaic or mundane phase has been ignored and discredited. While there are many explanations for this dichotomy in the tradition only one or two phases of the matter need be noted here. The eastern seaboard colonies were, of course, largely British in origin. Britain and Spain were mortal enemies. Hence the American colonists were predisposed to accept the Anglo-Saxon version of all things “Latin.” That Spaniards could be prosaic realists was, of course, an idea that few Anglo-American historians were prepared to concede. Then, too, a considerable part of the territory of the United States was acquired by conquest from Mexico shortly after its liberation from Spain. In view of the antecedent hostility between Britain and Spain, it was quite natural that this conquest should be rationalized as a mere by-product of the innate superiority of the Anglo-Saxon. Nevertheless a niche had to be found for the Spanish part of the tradition—after all Spain had discovered America; so the romantic and heroic aspects of the tradition were artificially divorced from its prosaic accomplishments and preserved like a rusty suit of armor in a museum. This tendency toward selective interpretation was abetted by a highly paradoxical aspect of Spanish influence in the New World. Although the Spaniards succeeded in transplanting their language, their religion, and many of their institutions in the Americas, they did so largely through the instrumentality of other groups. Spanish culture was superposed and inflicted on native peoples in the Americas; the Anglo-Saxon culture was carried over and preserved by

2  North from Mexico

immigrants. Not more than three hundred thousand Spaniards came to the Americas in the three colonial centuries and many of these, of course, came only for short periods and later returned. In the end the Spanish withdrew—from Florida, from Louisiana, and finally from Mexico itself. A retreating people may leave monuments and ruins but the reality of their experiences vanishes with the people, leaving only a faint memory, an easily erasable record. In such a sequence, it is only natural that the monuments and ruins should later acquire a piquant antiquarian interest and that the figures of the earlier scene should assume heroic proportions in the imagination of a later period. In the territory that now makes up the United States, Spaniards have always been a negligible ethnic element. It is doubtful if more than fifty thousand Spanish-born persons have resided in the United States at anyone period from 1820 to the present time. Spanish immigration, for this entire period, has probably not been in excess of 175,000. Thus things Spanish have been neglected or misunderstood in the absence of those familiar with the tradition and capable of interpreting the past in realistic terms. American schoolchildren have always known about Lafayette and of the aid which France furnished the colonies in the Revolutionary War; but few of them have ever been told that Spain also aided the colonies, that Spanish ports were open for the sale of prize-ships captured by American men-of-war, or that the Spanish governor of Louisiana furnished supplies of crucial importance to the American forces. Then, too, the splendor and romance of the Spanish story have tended to obscure its prosaic realities. The Latins were the great discoverers and explorers: Columbus, Magellan, Balboa, Cortez, Coronado, De Soto; and many of them were grandees and nobles. Somehow they invited the romantic apotheosis, the heroic summation. Lifting the curtain on the New World, they enacted the prologue to the drama of its settlement. Prologues, by the nature of their function, are heroic, splendidly phrased, lofty in sentiment; but never really a part of the play. In the latter-day versions, the prologue has been retained but the play itself has been rewritten and embellished. One need never worry about the heroic side of the Spanish tradition being forgotten or neglected; the Spanish explorations in the Southwest will live forever, gleaming like fine Toledo blades in the history of the region.

1. The Spanish Prologue From their base in the Indies—“the nursery of Spanish culture in the Western Hemisphere”—the Spaniards by 1525 had explored the entire shore line from Cape Breton to Cape Horn. Ponce de León had perished in Florida in quest of the fountain of youth. Lucas Vásquez de Ayllón had been lured to his death by the tall tales of Francisco Chichorana, a captured Indian, who had told the Spaniards of a race of Indians in the Carolinas with white skins and brown

In Spanish Saddlebags   3

hair hanging from their heels, of domesticated deer, of pearls and precious stones, and of a giant named Datha who had a tail like a crocodile . . . . After weeks of marching through the Florida everglades, fording rivers and wallowing through swamps up to their armpits, the survivors of Panfilo de Narvaez’s ill-fated expedition of 1528 finally returned to the shore, where their ships had been left at anchor, only to find that the ships had vanished. Beating their crossbows into saws and their stirrups and spurs into nails, they made five boats of palmetto bark, saplings, and horsehair and set sail from Florida. Part of the expedition was lost at sea but one of the boats, containing Cabeza de Vaca, was shipwrecked on an island near Galveston. Most of the survivors perished during the first winter among the Indians on the island and along the shore. To escape death, the resourceful De Vaca learned all that he could from the Indians: their language, their habits, and their lore. For nearly six years, he lived as a slave and medicine man among the various Indian tribes. Compelled to dig for edible roots along the shore, his fingers became so worn that “did a straw but touch them they would bleed.” Journeying inland to exchange sea-snails, shells, and trinkets for food, De Vaca ever returned to the island, where Oviedo, one of his companions, was still confined. Eventually the two of them made their way to the mainland, naked, halfstarved, armed only with bow and arrows and conch-shell knives. Deserted by the timid Oviedo, De Vaca finally came on Dorantes and Castillo, two members of the expedition, and with them the Christianized Moor, Estevánico, sometimes called Estevan, “the Black Mexican.” Traveling with Indian tribes, the four men journeyed west of the Colorado “on the prickly pear plains.” For a time they lived with the Arbadaos Indians who occupied the sand belt between the Nueces and the Rio Grande. Wherever they journeyed, it was De Vaca’s fame as a medicine man that won food and protection for them. Convinced that direct passage to Mexico was impossible, the party started westward, crossed the Rio Grande west of the Pecos and followed it to a point near El Paso. Over all this distance, hordes of Indians trave­led with them. “The Indians,” wrote De Vaca, “ever accompanied us until they delivered us to others.” From El Paso the party traversed the plains of Chihuahua and crossed the Sierra Madre Mountains to a town on the Rio Yaqui—somewhere near Sahuaripa, Sonora—where they were given six hundred “hearts of deer” and five arrows tipped with emeralds. Descending the Yaqui, they came upon a Spanish slave-hunting expedition on the Gulf of California and finally reached Mexico City on July 24, 1536. Excited by the stories of De Vaca, the Viceroy Mendoza sent a preliminary expedition northward from Culiacán in 1539, consisting of Fray Marcos, the by-now-famous Estevan, and a small party of Indian servants and interpreters. No character in the Spanish prologue is quite as interesting as Estevan, the Arab Negro from Azemur, who, in the course of his wanderings with De Vaca,

4  North from Mexico

had acquired a knowledge of six Indian dialects and had learned the arts of the medicine man. “He really explored,” writes Harvey Fergusson, “more of the Southwest and perhaps learned more about its people than any other man of his day.” Being a slave, he knew how to dominate men. Once a mark of bondage, his black skin made him a strange and supernatural being to the Indians. Marching on foot, unarmed, without maps or charts, Fray Marcos and Estevan journeyed northward through Sonora. As they proceeded, Estevan marched ahead, several days in advance, planting crosses in the desert to mark the way. Up the Sonora Valley they went, through the valley of the San Pedro in Arizona, to the Zuñi villages of western New Mexico where, at long last, Estevan finally sighted in the distance Cibola, the first of the fabled seven cities of gold. Among the polygamous Pima and Opata Indians, Estevan had been received like a strange and wonderful god. Bedecked with quantities of turquoise and feathers, carrying a gourd with tinkling bells, preceded by a large gray hunting dog, and followed by an entourage of enchanted Indian women, he sighted and perhaps entered the first of the Zuñi villages. From the Pima villages to Cibola, the number of attractive and brightly adorned Indian women who accompanied him had increased with every mile. It was probably this attraction he had for Indian women that brought about his death. For the Zuñi Indians were a sober, industrious tribe who greatly prized and honored their women. A year later the Zuñi chieftains told Coronado that they had murdered Estevan “because he had assaulted their women, whom the Indians love better than themselves.” A strange figure of legend, this Estevan, who, born in Africa, wandered over most of the Southwest, the first European—really an African— to find the seven cities of gold. Upon Fray Marcos’s return to Mexico, the great expedition led by Francisco Vásquez de Coronado was organized in 1540. Journeying northward into the San Pedro Valley, Coronado skirted the Santa Catalina Mountains, crossed the Gila River, and proceeded to the Zuñi River and beyond to Hawikuh, one of the seven cities of gold, “a little, crowded village, looking as if it had been crumpled all together.” Eastward the expedition came to the famous pueblo of Acoma, built on a flat-topped mountain three hundred and fifty feet in the air. It was at the village of Tiguex, above Albuquerque, that Coronado first heard, from an Indian whom the Spaniards named El Turco, of Golden Quivira on the Great Plains. Now that the seven cities of gold had turned out to be mud villages, Quivira became the goal of the expedition. Before leaving Tiguex, however, Coronado laid the foundation for most of the subsequent troubles of Spain in the borderlands by burning two hundred Indians at the stake—an “incident” which the Indians never forgot and for which they never forgave the Spaniards. In search of the Golden Quivira, Coronado continued his march eastward, entered the Great Plains of Texas, crossed the Arkansas River into Kansas, and

In Spanish Saddlebags   5

finally reached the first of the Quivira villages near the vicinity of Great Bend. Annoyed to find that Golden Quivira was merely another collection of mud villages, Coronado had El Turco garroted, set up a cross to mark the spot, and began the long march back to Mexico. When he finally presented himself before Mendoza, writes Bolton, he had brought with him nothing more precious than the gold-plated armor with which he had set out two years before. A member of the Coronado expedition, Fray Juan de Padilla, returned alone to the Quivira villages where he was murdered by the Indians—the first Christian martyr on American soil. It is said that his body is buried at Isleta, the pueblo below Albuquerque. “According to legend,” writes Ruth Laughlin, “Padre Padilla rises in his coffin, hollowed out of a cottonwood log, every twenty years. Some say that his emaciated body is as dry as mummy and his brown gown crumbling . . . but when his coffin bursts the mud floor before the altar, it is the blessed omen of a good year.” On June 27, 1542—while Coronado was returning from the buffalo plains to Mexico—Juan Rodríguez Cabrillo set sail up the west coast of Mexico to find the fabled island of California, peopled by black women, “on the right hand of the Indies, very close to the Terrestrial Paradise.” On the twenty-fifth of September, Cabrillo discovered the Bay of San Diego, where his ships rode at anchor off Point Loma for three days. Further north he came upon Santa Catalina and San Clemente Islands, entered the Bay of Santa Monica, or the Bay of Smokes, as he called it, and a few weeks later went ashore at San Buenaventura. With Viscaíno’s discovery of the Bay of Monterey in 1602, one might well say that the prologue was concluded. Between 1528 and 1602, a handful of Spaniards had explored the borderlands: from Galveston to San Diego; from Sonora to Santa Fe; from the west coast of Mexico to Monterey. If myth set them in motion, it was the Indians who lured them still further from their bases with tales of gold and silver, always seeking to draw them out of the Southwest. Where they had expected to find cities of gold, they found mud villages and uninhabited desert wastes. Something of this initial disappointment must have influenced their subsequent policy. For they neglected California from 1542 until the arrival of Fray Junípero Serra at San Diego in 1769; Texas was ignored for a century; and forty years elapsed before the settlement of New Mexico was undertaken. The colonization of New Mexico—“the first white colony in the transMississippi West”—was a by-product of the discovery of rich silver deposits at Zacatecas in 1548. It was a Zacatecas millionaire, Juan de Oñate, one of the four richest men in Mexico, who set forth in 1598 with eighty-three carretas, seven thousand head of stock, and four hundred soldiers to colonize New Mexico. Cabeza de Vaca had entered the borderlands from the southeast; Fray Marcos and Coronado from the southwest by way of Sonora; but De Oñate moved directly northward to El Paso and then up the Rio Grande to a point

6  North from Mexico

near Santa Fe. Poaching on the Pueblo Indians, the colonists managed to establish a series of settlements and by 1630 had founded some twenty-five missions. These initial settlements were extinguished in 1680, when the Indians revolted and drove the Spaniards from New Mexico in three days, killing four hundred of the settlers. Twelve years later, Diego de Vargas reconquered the province, made peace with the Pueblo Indians, and laid the foundations for settlements which survive to this day. In Northern Sonora and Southern Arizona—Pimeria Alta—the work of settlement was begun in 1687 by the Jesuit, Father Eusebio Francisco Kino. Italian by birth, Bavarian by education, Spanish by consecration, Father Kino founded the mission of San Xavier del Bac—the most impressive monument that Spain left in the Southwest—on April 28, 1700. In twenty-four years of service on the border, Father Kino, “the padre on horseback,” made over fifty important journeys of exploration and charted the Colorado from the mouth of the Gila to the Gulf of California. Based on these journeys, his map of the region was not improved upon for over a century. It was from Tubac, in Arizona, that Juan Bautista de Anza set forth on his famous march across the California desert to San Gabriel in 1775. Returning to Tubac the next year, he led a second expedition from that point to the missions of Southern California and from there to Monterey and on to San Francisco. While De Anza was exploring the Bay of San Francisco, seeking a site for the presidio, the American colonists on the eastern seaboard, three thousand miles away, were celebrating the signing of the Declaration of Independence. This second expedition of De Anza’s—“the longest overland migration of a colony in North American history before the settlement of Oregon”—has been rightly called the Anabasis of California history. Probably nothing in Xenophon’s recital of the march of the ten thousand to the sea exceeds it in courage and endurance. First and last, the Spaniards established twenty-one missions in California, spread like beads on a necklace at about one day’s march apart, along the rim of the seacoast from San Diego to San Francisco; founded four presidial towns— San Diego, Santa Barbara, Monterey, and San Francisco; and two pueblos—San Jose (1777) and Los Angeles (1781). None of these settlements was more than a day’s ride from the sea. While the Spanish penetrated the Central Valley, they established no colonies there (the Indians of the interior were quite unlike the docile coastal tribes). Although the Spanish founded some twenty-five missions in Texas, their principal and ultimately their only settlements between the Sabine and the Rio Grande were San Antonio (1718); Goliad or La Bahia; and Nacogdoches. Thus the Spanish settlements in the borderlands really consisted of a firmly rooted colony in New Mexico; an easily held and fairly prosperous chain of missions in coastal California; and a number of feebly garrisoned, constantly imperiled settlements in Texas and Arizona.

In Spanish Saddlebags   7

While a dozen or more settlements were founded in Florida, largely as a protective flank for the silver of Mexico, the fate of these settlements was sealed when the British occupied Charleston in 1670. Reverting to Spanish rule in 1769, the Florida settlements were again outflanked when the United States negotiated the Louisiana Purchase. In 1821 the colonies were transferred to the United States, coincident with the liberation of Mexico from Spanish rule, and the Spaniards withdrew from the eastern seaboard.

2. Footnote to the Prologue On April 16, 1605, Juan de Oñate chiselled a notation on Inscription Rock, located midway on the trail from Zuñi to Acoma. “Paso por aquí” it read; “Passed by this way.” One of the earliest historical records within the present confines of the United States, the notation epitomizes the heroic side of the Spanish adventure in the borderlands. For many years the Spanish names on Inscription Rock were of no more interest to the Anglo-Americans than the monument itself. But toward the end of the nineteenth century, America discovered the Southwest and tourist hordes beat a trail to the monument to stare at the Indian hieroglyphics and the odd-sounding Spanish “names on a rock.” No sooner had the tourists begun to encircle Inscription Rock than a school of romantic American historians proceeded to elevate everything the Spanish had done in the Southwest to the level of high heroism and intense passion. In one sense the Spanish explorers fully merited this posthumous laudation for they were brave and courageous men—tough, hard-bitten, and indomitable. But the saga that has grown out of their explorations is so inhumanly grandiose as to create a most misleading impression of their real accomplishments. Reading the admirable narratives which the Spanish left, one is today impressed with the prosaic, matter-of-fact manner in which they recounted their adventures. They were not nearly so impressed with their own heroism as were the American historians of a later date. What they consistently emphasized was their good luck, or as they put it, the grace of God. That the scale of their achievement has somehow become askew can be easily demonstrated. Reading De Anza’s report on the province of Sonora in 1777, one is struck by the small stage, the tiny settings, the miniature proportions of the drama. De Anza was not deploying “armies”: he had exactly 502 soldiers at his disposal in all Pimeria Alta. Nor was he opposed by “armies” of Indians. In a great expanse of largely uninhabited territory, a handful of Spaniards—tiny pinpricks on the desert—were conducting a not-too-gory duel with a few widely scattered Indian tribes. At the heyday of Spanish expansion, one could have placed all the inhabitants of the borderlands, including the Indians, in a corner of the present-day County of Los Angeles. The Spanish

8  North from Mexico

prologue was a drama-in-miniature; the quality was heroic but the actors were puppet-like figures against an enormous backdrop. Actually the famous marches and entradas were not quite as heroic or as difficult as they have been made out to be. Possessing an invincible superiority of weapons, the Spanish could march wherever they wanted to march; there was nothing to stop them. Trained and seasoned campaigners, they knew all the tricks of survival in an arid environment. Journeys that appear to us in retrospect as feats of incredible endurance were merely another day’s march to them. Mexicans and Indians accompanied them wherever they went, often trudging on foot while they rode on horseback. During both the period of exploration and of settlement, Spanish casualties in the borderlands were negligible. While they were contained, within narrow limits, by warlike nomadic tribes, they fought few pitched battles. Had it not been for the Indians, neither their explorations nor their settlements would have been possible. Indian guides took them from one water hole to the next, from one tribe to another; and wherever they settled they lived for years off the bounty of the Indian villagers. The key to Cabeza de Vaca’s survival is simply that he was compelled to live with Indians and did so. Wherever they went, he went; whatever they did, he did. The Spanish prologue should not, of course, be disparaged; but it cannot be understood or appreciated until it is first reduced to scale. In the fervid rhetoric of the historians, every move of these “soldiers of the Cross” was “magnificent,” “heroic,” and “epochal.” Their achievements have been so inflated that they have lost much of their meaning and relevance. Once the record of achievement is looked at, not through a mist of romantic preoccupations, but as the Spaniards themselves saw their position and their achievements, then the facts fall into place and the scale assumes proper proportions. For the heroic side of the tradition, the part that is remembered and honored, was not the whole of the Spanish story in the Southwest. There was a play, a real drama, as well as a prologue. The prologue was written in heroic couplets; the play in prose. We know the prologue; it is the play that needs to be re-enacted.

3. The Play-in-Prose From the conquest of Mexico to the end of Spanish rule, a corporal’s guard of Spanish soldiers was able to march at will throughout Mexico, Central and South America, and the borderlands. A few hundred Spanish soldiers, officials, and priests seized, held, and ruled the province of California; for years on end the military might of the Spanish empire was represented in New Mexico by several companies of soldiers crudely armed. At the close of the period, there were not more than fifteen thousand Spaniards in the Americas—less than one-third of one percent of the population of New Spain. “Upon this slender basis,” writes Ernest Gruening, “the conquering Hispanic minority superposed

In Spanish Saddlebags   9

its political structure, religion, and language on the natives.” What was the secret of this astonishing success? The answer is twofold: horses and weapons. The Spaniards were superb horsemen: “the centaurs of the mesa.” That they were horsemen enabled them to travel great distances and to wear armor. With neither armor nor horses, the Indians were simply no match for the Spaniards. Hence Cortez, with a few hundred men, could fight his way into Mexico City and hold it against unthinkable odds. “Superiority of weapons,” writes F. S. Curtis Jr., “made such exploits a possibility. . . . Against stone-tipped arrows and lances, obsidian daggers, stone-headed clubs, and the propelling force of the human arm alone, the Spanish opposed steel-tipped arrows and lances, steel swords and daggers, and the propelling force of gunpowder.” As long as the Spanish possessed this superiority in arms and held exclusive possession of the horses which made their use possible, they could enact the role of conquistadores; and, for a time, they took every precaution to prevent the Indians from gaining possession of either. The Spaniards themselves were well aware of the secret of their success. “The arrival which has been of greatest importance to the Spaniards of all those brought to the Indies,” wrote the Jesuit, Bernabe Cobo, “is the horse; because with its aid they have been able to make so many and such notable conquests, and have discovered so many regions and have spread so far in so short a time through so many and extensive lands.” When the Spanish began the conquest of the Americas, they had the best horses in Europe. Even before the Moors had brought them the famous “horses of the desert,” they had developed fine breeds of their own. The American horse came almost entirely from the Spanish stock of New Mexico and Florida. One reads, for example, that Patrick Henry was in the habit of sending to the Pawnee country “for the best and most pure Spanish breed.” The Apaches began to steal horses from the Spaniards as early as 1600; in fact, the Apache menace was as old as the first settlements in the borderlands.1 Contrary to Spanish policy elsewhere in the Americas, the New Mexico rancheros had been authorized in 1621 to use captive Indians as herders and to allow them to ride horses—a fateful mistake. To the nomadic Indians of the borderlands, the coming of the horse was as important as the discovery of steam to the white man. “Mounted,” writes Donald E. Worcester, “the Apaches presented a problem unlike any with which the Spaniards had previously been plagued. Whereas it was fairly simple to surround a pueblo and force the inhabitants to surrender, the Apaches had no homes or towns to be defended, and no large armies to be defeated.” Spanish colonization in the Southwest proceeded, from the outset, under the cloud of Apache terror. The fifth century Europeans felt no greater fear of the Huns of Attila than the fear which the Apaches inspired in the Spaniards and the Pueblo Indians. Later the Indians began to barter captives for horses and to

10  North from Mexico

exchange horses, which they stole from the Spanish, for guns and powder which they got from the French. While horses could carry the Spaniards into the Southwest, their system of colonization did not travel so well. Based on the presidio, the mission, and the hacienda—the conquistadores to conquer; the priests to convert; and the encomenderos to exploit—the system worked satisfactorily in the Central Mesa region of Mexico where sedentary Indians could be exploited. In the borderlands, however, the system quickly broke down, although it worked fairly well in New Mexico: the one province with a large population of sedentary Indians. Defeated in battle by the Spanish, the Pueblo Indians returned to their villages and tilled acres and accepted Spanish dominance; in effect, they were driven back by the nomadic tribes who functioned as policemen for the Spanish. But the nomads, having nothing to defend, could never be subdued. Although few in number, the heavily armed Spaniards could conquer wide territories but they could never consolidate their conquests nor could they win a decisive battle with the nomads. As a consequence, the Spanish were compelled to do what the Pueblo Indians had done centuries before: they concentrated their forces in a few well-chosen and easily defended positions. To these positions, however, they were pinned like butterflies on a screen. The coastal Indians in California were sedentary but they did not live in villages; hence they had to be rounded up and brought to the missions. While some of the missions prospered, the Indians certainly did not; on the contrary, they died like flies. In Arizona and Texas, the Spaniards tried in vain to subdue the Apaches and the Comanches. Unworkable in the borderlands, the encomienda system was abandoned about 1720. Fifty years later the mission system had been virtually abandoned in Texas and Arizona and the secularization of the missions in California brought an end to the period of mission rule in 1834. “As the Spanish arms and civilization,” writes Dr. George P. Garrison, “came in closer contact with the wild and fierce Apaches and Comanches of the north their progress became increasingly difficult and uncertain.” Before many years had passed, it was brought to a kind of halt or stalemate. While their explorations “were wide enough for a schoolboy’s dream, their grasp was too weak for permanence.” In locating settlements in the borderlands, the Spanish had pursued their time-honored policy of settling in areas occupied by sedentary Indians. Quite apart from the fact that their colonization scheme was premised upon the use of Indian labor, a sound instinct told them that where Indian settlements existed human life could be supported. Prior to their appearance in the Southwest, however, a great drama had been enacted of which they knew nothing and concerning which not too much is known today. It would seem, however, that the Pueblo Indians had for many years been fighting a losing battle against their hereditary enemies, the nomadic tribes. Driven out of the river bottoms

In Spanish Saddlebags   11

and valleys, they had finally sought shelter in the nooks and crevices of the mountainous portions of the Southwest. Here, in cliff dwellings, terraced adobes, and mountain villages, they were able to survive although in constant peril. As a result of protracted defensive warfare, their culture had begun to disintegrate and showed a marked decline in vigor at about the time the Spaniards arrived. In fact, it is altogether probable that the Spaniards rescued and to a degree revitalized the culture of the Pueblo Indians. Unaware of this drama, the Spaniards passed by the rich gold, silver, and copper deposits and marched through the fertile river bottoms of the south to plant their colonies in the narrow valleys of the Rio Arriba in northern New Mexico. It is important to note that the Spanish never occupied or settled more than a small portion of the province. Despite the great distances and the largely uninhabited terrain of New Mexico, the colonists were, paradoxically, more crowded than their contemporaries in the English colonies on the eastern seaboard. For the arid nature of the environment and the location of the Indian villages confined their settlements to the narrow strips of fertile land in the valleys of the north. Not knowing their problems, unaware of their predicament, the Anglo-Americans later jeered at their “mistakes” and derided their “failures.” It was this failure to understand the nature of the situation which the Spanish faced that so largely accounts for the dismissal of their prosaic accomplishments as of slight importance.

4. The Climate of Spain Despite the insurmountable obstacles which they encountered, the Spanish left an imprint on the borderlands which, as Bolton once said, “is still deep and clear.” The names of three states in the region are Spanish in origin: California, Nevada, and Colorado. “Scores of rivers and mountains and hundreds of towns and cities,” to quote Bolton again, “still bear the names of saints dear to the Spanish pioneers.” From Los Angeles to San Antonio, the Spanish language is spoken today by upwards of two million people. Thousands of Indians living in the region speak Spanish in preference to English and profess the Catholic faith. The imprint of Spain is to be found in the land systems of the region; in the law of waters, of minerals, and of community property; and in many institutions now firmly planted in the Southwest. “It is to them and their followers and descendants,” writes Dr. Frank C. Lockwood, “few though they were, and opposed as they were by harsh nature and hostile savages, that our Anglo-Saxon pioneers owed from the first a degree of exemption from such extremes of ignorance and crudity as most American pioneer settlers have experienced.” A major factor in the prosaic success of the Spanish in the Southwest consisted in the similarity of the climate and environment to that of Spain. Spain is the only European country with an arid or semi-arid environment. Over a

12  North from Mexico

period of thirty generations, the Spanish had acquired a profound knowledge of the nature and limitations of such an environment. The Spaniards knew and understood irrigated farming, for irrigation is necessary throughout the whole of the Spanish peninsula as it is throughout the Southwest. They had solved the problem of building without timber by the use of adobe, brick, and stone, a practice which proved of great value in the Southwest. In almost every respect, their institutions and culture were adapted to the environment which they found in the borderlands.“The scale was higher,” writes Dr. Walter Prescott Webb, “but the difference was more of degree than of character or quality.” Everything that the Spanish brought to the Southwest was either driven or carried in saddlebags or carretas. But what they did bring was of enormous importance (I am speaking now merely of things; the question of institutions and cultural practices will be considered later). They brought the first cattle, horses, goats, pigs, cats, and barnyard fowls to the Southwest. At great effort they brought from Europe the first hoes, spades, grinding stones, clamps, plows, files, and pliers used in the region. The first wheels that turned on American soil were Spanish in origin. “There was no product contributed to the agriculture of America by the English,” writes Harvey Bernstein, “which the Spanish had not planted earlier.”2 The list is a long one, indeed, including peaches, figs, oranges, apples, grapes, apricots, limes, pomegranates, pears, olives, and lemons. Over 260,000 orange trees were planted in the older mission groves of California by 1880. The vineyards that developed so rapidly in California after 1848 were planted in or near the original vineyards which the Spanish had laid out. Up to 1860, there was only one grape grown in California—the Mission grape; but in that year the state’s thirty million vines were producing upwards of seven million gallons of wine. The Spanish introduced raisin culture to California. The first wheat seeds brought to California came from Spain and were, on their arrival in 1770, already well-adapted to climatic and soil conditions not unlike those which existed in the state. The first wheat planted in Colorado was Spanish in origin and was known as “Sonora wheat.” The grandfather of Don Amado Chaves brought the first alfalfa seed to New Mexico tied in a manta. The Spanish brought the important seed of the alfilaria or filaree to California as well as the first flax seeds. In 1806 the Spaniards introduced a cotton seed from Mexico that produced exceptionally large, wide, open bolls. Introduced by way of New Orleans, this seed spread rapidly throughout the Deep South. While the area farmed by the Franciscans in California was not large— perhaps not more than 10,000 acres—the mission “gardens” provided a preview of the state’s agriculture. “The fruits and nuts known to have been grown by the padres,” writes Dr. Frank Adams, “included almost all those now produced in California, and some that have not succeeded commercially. There were pears, peaches, apples, almonds, plums, quinces, pomegranates, oranges,

In Spanish Saddlebags   13

lemons, citrons, limes, dates, cherries, plantains, walnuts, grapes, olives, figs, strawberries and raspberries.” Not only did the Franciscans demonstrate that all these crops could be produced in California, but some of the fruits now grown in the state have come directly from the mission gardens, notably the Mission grape, Mission fig, and Mission olive.3 From the Spaniards, the Indians of the Southwest learned to hammer silver and copper; to work iron; and to use plows and hoes. From the Spanish the Navajo took over much of their present-day pastoral culture, including a knowledge of how to card and to weave wool. To this day, the Indians use Spanish terms to designate the colors used in their blankets: morada subido, rosa baja, oro, amarillo, tostado, grano. The Spanish introduced sugar cane to Louisiana and built the first sugar refinery in New Orleans in 1791. Spanish gardeners, farm laborers, blacksmiths, millwrights, and artisans brought a variety of skills to the Southwest. In fact it would take pages merely to list the things which the Spanish brought to the region that have long since been incorporated in its culture. Savants have written learned papers on what the introduction of the cat meant to the people of the New World.4 What the Spaniards contributed to the Southwest, in addition to their language, religion, law, and institutions, were the seeds of things which were later of enormous importance. They were the trailblazers and seed planters. In most cases, the key to the success with which these seeds were transplanted is to be found in the similarity of the Southwest environment to that of Spain. The horses, sheep, cattle, and goats that the Spanish brought to the region, as well as the plants that they imported, were alike well adapted to the environment; and so were many of their cultural practices and institutions. That many of these seeds failed to reach full development and maturity under Spanish rule is due to a variety of factors: the intractable character of the plains and desert Indians; the dead weight of tradition; the highly centralized character of Spanish administration; the senseless restrictions imposed on initiative and innovation; the cheapness of Indian labor; and the feudal, caste-like social system which the Spanish also brought to the Southwest. That the Spanish failed to make the most of their opportunities, however, cannot detract from the importance of their prosaic accomplishments. Still another key to the amazing success with which Spanish cultural influences were planted in the Southwest is to be found in the isolation of the borderlands. Isolated in time and space, the settlers were compelled to plant things firmly in the soil. What survived may later have appeared to be scrawny, crude, and misshapen; but it was unquestionably tough, well adapted, and indigenous. While the form or model was often Spanish, the ultimate adaptation showed unmistakable Mexican and Indian influences. If the Spanish were the carriers of seeds and plows, Mexicans and Indians were the planters and plow hands. Beyond all doubt the culture of the Southwest, in 1848, was a trinity: a whole

14  North from Mexico

consisting of three intricately interwoven, interpenetrated, thoroughly fused elements. To attempt to unravel any single strand from this pattern and label it “Spanish” is, therefore, to do a serious injustice to the Mexicans and Indians through whom, and only through whom, Spanish cultural influences survived in the region. “I do not agree with all this talk about Coronado,” Pablo Abeyta, the governor of the Isleta Pueblo, has said; “I don’t know what they mean by Spanish culture. . . . The Spaniards got lost on the ocean and accidentally ran across the country.” Whether by accident or otherwise, however, the prologue to the settlement of the borderlands is indubitably Spanish in origin.5

2

The Fantasy Heritage

Long, long ago the borderlands were settled by Spanish grandees and caballeros, a gentle people, accustomed to the luxurious softness of fine clothes, to well-trained servants, to all the amenities of civilized European living. Inured to suffering, kindly mission padres overcame the hostility of Indians by their saintly example and the force of a spiritual ideal, much in the manner of a gentle spring rain driving the harsh winds of winter from the skies. Life was incomparably easy and indolent in those days. There was none of the rough struggle for existence that beset the Puritans in New England. The climate was so mild, the soil so fertile, that Indians merely cast seeds on the ground, letting them fall where chance deposited them, and relaxed in the shade of the nearest tree while a provident and kindly nature took over. Occasionally one of the field hands would interrupt his siesta long enough to open one eye and lazily watch the corn stalks shooting up in the golden light. . . . In the evenings one or the other of the patios would witness the gathering of the Spanish dons from the ranchos. Here in the coolness of the evening air they would talk of the day’s events, sipping gentle wines that revived memories of castles in Spain. While the men were thus pleasantly engaged, the women would continue their never-ending routine of tasks that kept the large households functioning smoothly. For the young people, it was a life of unrivalled enjoyment; racing their horses over the green-rolling hills and mustard fields of Southern California; dancing the contradanzas and jotas to the click of castanets. In the evening, the young ranchero strolled beneath the window of his love’s boudoir. As the moon rose high over the Sierra Madres, he would sing the old love songs of Spain. . . . All in all, this life of Spain-away-from-Spain in the borderlands was very romantic, idyllic, very beautiful. . . . Indeed, it’s really a shame that it never existed. Never existed? How can this be said when so much of the public life of Los Angeles is based on the assumption that it did? Why do churches in Los Angeles never hold bazaars? Why are they always called fiestas? Why is a quarter acre and twenty chickens called a rancho? Why does a leading newspaper

16  North from Mexico

gossip columnist adopt the nom de plume of “La Duenna”? Why does the largest women’s club, composed exclusively of Anglo-American women, hold an annual “gala Spanish fiesta program” in which the ladies appear in “full Spanish costume” to admire Señor Raoul de Ramirez’ presentation of The Bells of San Gabriel?1 And, lastly, why do so many restaurants, dance halls, swimming pools, and theaters exclude persons of Mexican descent? Los Angeles is merely one of many cities in the borderlands which has fed itself on a false mythology for so long that it has become a well-fattened paradox. For example, the city boasts of the Spanish origin of its first settlers. Here are their names: Pablo Rodríguez, José Variegas, José Moreno, Felix Villavicencio, José de Lara, Antonio Mesa, Basilio Rosas, Alejandro Rosas, Antonio Navarro, and Manuel Camero. All “Spanish” names, all good “Spaniards” except—Pablo Rodríguez who was an Indian; José Variegas, first alcalde of the pueblo, also an Indian; José Moreno, a mulatto; Felix Villavicencio, a Spaniard married to an Indian; José de Lara, also married to an Indian; Antonio Mesa, who was a Negro; Basilio Rosas, an Indian married to a mulatto; Alejandro Rosas, an Indian married to an Indian; Antonio Navarro, a mestizo with a mulatto wife; and Manuel Camero, a mulatto. The twelfth settler is merely listed as “a Chino” and was probably of Chinese descent. Thus of the original settlers of Our City the Queen of the Angels, their wives included, two were Spaniards; one mestizo; two were Negroes; eight were mulattoes; and nine were Indians. None of this would really matter except that the churches in Los Angeles hold fiestas rather than bazaars and that Mexicans are still not accepted as a part of the community. When one examines how deeply this fantasy heritage has permeated the social and cultural life of the borderlands, the dichotomy begins to assume the proportions of a schizophrenic mania.

1. The Man on the White Horse “Three hundred years,” writes Tom Cameron in the Los Angeles Times of August 29, 1947, “vanished in an instant here in Santa Barbara today as the city and more than 100,000 guests plunged into a three-day round of pageants, parades, street dancing and impromptu entertainment. It is La Fiesta. Santa Barbara is a particularly bewitching señorita today. With glowing copa de oro flowers entwined in her raven tresses and with her gayest mantilla swirling above her tight-bodied, ruffled Spanish colonial gown, she is hostess to honored guests from near and far. It is a time when Santa Barbara gazes over her bare shoulders [sic] to a romantic, colorful era of leisurely uncomplicated living . . . .” With one thousand beautiful “gaily caparisoned” Palomino horses prancing and curveting along State Street—renamed for three days “Calle Estado”—the history of the region is dramatized in costly and elaborate floats. This year,

The Fantasy Heritage   17

1947, the Kiwanis Club enters a float in honor of Juan Rodríguez Cabrillo; Rotary honors Sir Francis Drake; the Exchange Club pays homage to Sebastián Viscaíno. “A traditional wedding party of 1818 escorted by caballeros, canters along. It represents the wedding of Anita de la Guerra and Capt. Alfred Robinson.” Following the charros, riders from San Gabriel and the Spanish grape carts drawn by donkeys with flower girls astride, come the Long Beach mounted police, the Del Rey Palomino Club, Los Rancheros Visitadores (headed by J. J. Mitchell of Juan y Lolita Rancho), and of course the Los Angeles sheriff ’s posse headed by Eugene Biscailuz, the sheriff, himself an “early Californian.” The celebration comes to a finale with the presentation in the Santa Barbara Bowl of a pageant written by Charles E. Pressley entitled Romantic California— and very well titled it is. “Spanish” food is served; “Spanish” music is played; “Spanish” costumes are worn. For this is the heritage, a fantasy heritage, in which the arbiters of the day are “Spaniards.” The Mexicans—those who are proud to be called Mexican— have a name for these “Spaniards.” They call them “Californios” or “Californianos” or, more often, “renegados.” These are the people after whom streets are named in Los Angeles: Pico, Sepúlveda, Figueroa. It is they who are used by the AngloAmerican community to reconcile its fantasy heritage with the contemporary scene. By a definition provided by the Californios themselves, one who achieves success in the borderlands is “Spanish”; one who doesn’t is “Mexican.” This fantasy heritage makes for the most obvious ironies. Cinco de Mayo is one of the Mexican national holidays which Los Angeles, now a Good Neighbor, has begun to observe. It is celebrated by parades, fiestas, and barbecues; speeches by the mayor and the Mexican consul constitute the principal order of the day. Invariably the parade winds its way through Olvera Street and the Plaza—sections of the old Mexican town now kept in a state of partial repair for the tourist trade—to the City Hall. Leading the parade through the streets, riding majestically on a white horse, is a prominent “Mexican” actor. Strangely enough, this actor, a Californio three hundred and sixty-four days of the year, becomes a “Mexicano” on Cinco de Mayo. Elegantly attired in a ranchero costume, he sits proudly astride his silver-mounted saddle and jingles his silver spurs as he rides along. The moment he comes into sight, the crowds begin to applaud for he is well known to them through the unvarying stereotypic Mexican roles which he plays in the films. Moreover, they have seen him in exactly this same role, at the head of this or some similar parade, for fifteen years. Of late the applause is pretty thin and it may be that the audience is becoming a little weary of the old routine. A union organizer of Mexican descent once remarked to me: “If I see that white horse once more, I’m going to spit in its eye.” Following the man on the white horse will be other horsemen, few of them with any pretensions of Mexican descent but all similarly attired, mounted on

18  North from Mexico

splendid Palominos, horses worth their weight in gold, decorated with their weight in silver trappings. At one time there were men in Mexico who dressed in nearly this fashion. The full irony of the situation dawns when one realizes that the men who lead the parade are dressed like the same class whose downfall is being celebrated. The irony would be no greater if the Angelenos put on the brilliant red uniforms of British grenadiers when they paraded on the Fourth of July. For on Cinco de Mayo blood was shed to rid Mexico of grandee landowners who threatened to suck it dry. Here, in Los Angeles, the men who lead the parade symbolically represent the grandees while the Mexicans line the pavements. These Californios are in no small part responsible for the fact that the Mexican population of Los Angeles—the largest minority in the city—is so completely deprived of meaningful civic representation. Since it is impolitic for any Los Angeles official to ignore the Mexican vote completely, care is taken that the roster of civic committees shall always include at least one name which is obviously Spanish or Mexican. If a quick glance is taken of the list of names appearing on the civic committees devoted to housing, juvenile delinquency, racial, and welfare problems, these same names constantly reappear. It has only been of recent years that the Californios have been elevated to this anomalous and largely factitious status. There was a time when they scarcely existed in the eyes of the Anglo-Americans. When the Native Sons of the Golden West were asked, in the early 1900s, to submit a list of “the men who had grown up with Los Angeles,” for a civic memorial, they included only Anglo-American names. When the first “pioneer society” was formed in Los Angeles in 1896, not a single Mexican or Spanish name appeared on the membership roster and the bylaws expressly provided that “persons born in this state are not eligible to membership.” Ignored throughout this early period, the Californios promptly acquired a new and spurious status the moment it became necessary to use them to maintain the subordination of Mexican immigrants in the general scheme of things. Today the typical Californio occupies, in most communities, a social position that might best be compared with that of the widow of a Confederate general in a small southern town. On all ceremonial occasions, the “native Californians” are trotted forth, in their faded finery, and exhibited as “worthy representatives of all that is finest in our Latin-American heritage.” In appointing Californios to civic committees, most officials realize that they have achieved the dual purpose, first, of having a Mexican name on the roster for the sake of appearances, and, second, that the persons chosen will invariably act in the same manner as Anglo-Americans of equal social status. Thus, the dichotomy which exists throughout the borderlands between what is “Spanish” and what is “Mexican” is a functional, not an ornamental, arrangement. Its function is to deprive the Mexicans of their heritage and to keep them in their place.

The Fantasy Heritage   19

In community after community, the Anglo-Americans genuflect once a year before the relics of the Spanish past. Just as Tucson has its annual La Fiesta de los Vaqueros so nearly every city in the borderlands now has its annual Spanish Fiesta. It is during La Fiesta in Santa Barbara that the annual ride of the Rancheros Visitadores occurs. This particular revival is based on a practice of former years, when the rancheros made the rounds of the ranchos to pay a visit to each in turn. “In May, 1930,” to quote from the Santa Barbara Guide (WPA), “some sixty-five riders assembled for the first cavalcade. Golden Palominos and proud Arabian thoroughbreds, carrying silver mounted tack, brushed stirrups with shaggy mustangs from the range. Emerging from the heavy gray mist of a reluctant day, they cantered with casual grace down the old familiar trails of Santa Ynez to converge on Santa Barbara. . . . Here amid the tolling of the bells, the tinkling of trappings, and the whinnying of horses, the brown-robed friars blessed them and bade them ‘Vayan con Dios.’ . . .” This was the first ride of the Rancheros Visitadores whose president, today, is Señor J. J. Mitchell. Since this auspicious beginning, the affair has steadily increased in pomp and circumstance. Nowadays it is invariably reported in the Southern California society columns as a major social event of the year. A careful scrutiny of the names of these fancily dressed visitadores—these gaily costumed Rotarians—reveals that Leo Carrillo, “the man on the white horse,” is about the only rider whose name carries a faint echo of the past that is being celebrated so ostentatiously. Numerous institutions have been founded in the borderlands to keep the fantasy heritage alive. First performed at Mission San Gabriel on April 29, 1912, John Steven McGroarty’s “Mission Play” was presented at over 2,600 performances and was seen, according to its modest author, by over 2,500,000 people. The Padua Institute, located at the base of the Sierra Madre Mountains near Claremont, is another institution which works hard to keep the fantasy heritage alive. Here, in a beautiful setting, the lady from Des Moines can have lunch, see a Spanish or Mexican folk play, hear Mexican music, and purchase a “Mexican” gift from the Studio Gift Shop. The Padua Institute is dedicated to “keeping alive the romantic life and music of Old Mexico and Early California.” Olvera Street, in the old Plaza section of Los Angeles, is still another attempt to institutionalize the false legend. Harmless in many ways, these attempts to prettify the legend contrast most harshly with the actual behavior of the community toward persons of Mexican descent. To the younger generation of Mexicans, the fantasy heritage, and the institutions which keep it alive, are resented as still additional affronts to their dignity and sense of pride. Try as they will, the Anglo-Americans cannot quite enter into the spirit of La Fiesta. Compliments are exchanged between the mayor and the consulgeneral and the usual remarks are made about Benito Juárez and Abraham

20  North from Mexico

Lincoln; but, somehow, the emptiness of the occasion echoes in the platitudes spoken. This meretricious quality is always apparent in the gauche efforts of the press to whip up some semblance of enthusiasm. “Vivas and olas filled the air. . . . Los Angeles yesterday donned the festive regalia of her Mexican heritage . . . Cinco de Mayo Festival On, Si, Si”—are excerpts from the Los Angeles Times of May 6, 1947. On the Sixteenth of September 1947, a Miss Frances Anderson was selected as the reigning señorita in one Southern California town; while, in another, a Miss Virginia Thomas was selected. Both towns have a large Mexican population. In an editorial commending a program to teach Spanish in the lower grades, the Los Angeles Times (August 29, 1944) in a fervor of españolismo wrote: “we have missed learning the homey, friendly gossip of the little people who have big hearts even if lean purses. We have missed much, señores . . . Viva Mexico! Viva el Español!”

2. The Birth of a Legend Throughout the Southwest today the most striking aspect of Anglo-Hispano relations consists in this amazing dichotomy between the Spanish and the Mexican-Indian heritage. There is scarcely a public building constructed since the turn of the century, whether it be a library, a post office, or a courthouse, without murals depicting scenes in which Cabrillo, Serra, De Oñate, and Coronado played a part. Nowadays Juan Bautista de Anza could travel over the trail that he blazed from Tucson to San Jose and spend every night in a De Anza hotel. But there is scarcely a single community in the region in which the living side of this tradition has not been consciously repudiated. “In spite of their willingness to borrow local color from neighboring Mexico,” reads the Arizona Guide (WPA), “the Anglo-Saxons of Arizona have usually made a conscious effort to avoid the adoption of the more fundamental traditions and characteristics of the Mexican people. In general the ArizonaMexicans have been segregated from the more fortunate Arizonians, both as strangers belonging to an alien race of conquered Indians, and as persons whose enforced status in the lowest economic levels make them seem less admirable than other people. They have consequently retained a firmer hold on their native customs and folklore than have other groups of foreigners less discriminated against.” This same statement could be made of any state in the region. “When one sees the great sums spent to reconstruct the Spanish missions and other buildings of the Latin-American occupation,” writes Jovita Gonzales de Mireles, “one cannot help but wonder at the inconsistency of things in general. If Anglo-Americans accept their art and culture, why have they not accepted the people?” One reason, of course, is that the discovery of Spain-in-America has been of comparatively recent origin. Harry Carr, a veteran Los Angeles

The Fantasy Heritage   21

newspaperman, once remarked that when he came to live in the city as a youngster in the eighties, schoolchildren were taught nothing about the epochal adventures of Coronado, De Oñate, and De Anza. Once the Spanish past was resurrected, this early neglect was greatly overcompensated. Discovered as a tourist promotion in the 1880s, the Spanish mission background in Southern California was inflated to mythical proportions. Originating in Los Angeles, the “landmarks” movement spread throughout the Southwest. Today community after community is busily resurrecting its “Spanish” ruins and, in a number of cases, master-plans have been adopted—as in St. Augustine in 1936; Monterey in 1939; and San Antonio in 1938—to rebuild whole communities along lines consonant with the original Spanish conception.2 A second factor has to do with the amazingly heterogeneous character of the Spanish-speaking minority. “Biologically,” writes Dr. George Sanchez, “they range over all the possible combinations of, first, their heterogeneous Spanish antecedents and, then, of the mestizaje resulting from the crossing of Spaniards and various indigenous peoples of Mexico and the Southwest. Historically, they are both old and new to this region—some came with Oñate in 1598, others with missionaries of the eighteenth century; some were a part of the gold rush of ‘49, others came to build railroads a few decades later; many came as contract-labor during World War I. Culturally, reflecting their varied biological and historical backgrounds, they are many peoples—the californios, the hispanos, the mexico-tejanos, and numerous other cultural personalities produced by the range of their antecedents and their environments, by their occupations, by their culture-contacts. These people, of whom only a minority are citizens of Mexico, are most often referred to as ‘Mexicans.’ Their mother tongue, their vernacular, is usually Spanish—though every conceivable variation of that tongue obtains, in terms of all phases of both quantity and quality. In fact, for some the home-language is English; for others a part-English, part-Spanish vernacular is the rule. These Spanish-Mexican Americans of the Southwest, then, defy categorical classification as a group and no term or phrase adequately describes them.” The native-born Spanish-speaking elements resent any attempt to designate them in a manner that implies a “nonwhite” racial origin. Being called “Mexican” is resented, not on the basis of nationality, but on the assumption of racial difference. Because of the Anglo-American’s attitude toward race, the first reaction of the New Mexican, as Dr. Arthur L. Campa has pointed out, “is to disassociate himself from anything that carries a Mexican implication.” To do this, he must insist on his difference in origin. Thus he is of “pure Spanish blood,” a direct descendant “of the Spanish conquerors,” etc. Carried to its logical conclusion, this line of reasoning results in the deductions (a) that the New Mexican is not “Mexican”; and (b) that he has no Indian blood. “Being American citizens, the next step is to combine the concept of race with that of nationality and the hyphenated Spanish-American is the result. Such a term

22  North from Mexico

serves a triple purpose: it lifts from the New Mexican the opprobrium of being a Mexican; it makes him a member of the ‘white’ race, and expresses his American citizenship.” But the difficulty with “Spanish-American,” as Dr. Campa adds, is that, while it suits the New Mexican in the abstract, there is little in his appearance and origin that upholds the distinction he is trying so hard to make.3 The difference between New Mexicans and Mexicans being regional distinctions occurring within a similar culture, the substitution of the name “Spanish” does not change the substance of traits that are undisputedly Mexican. “The ‘Spanish’ suppers,” writes Dr. Campa, “given by clubs and church societies are in reality Mexican dishes to which no truly Spanish palate is accustomed. The ‘Spanish’ songs sung by school children and by radio performers in New Mexico are as Mexican as tortillas de maiz, chicharrones de puerco, chile con carne, and the sopaipillas at Christmas time.” To the Anglo-Americans of the borderlands, with their racial preoccupations, it is second nature to refer to the Spanish-speaking group as “Mexican”; whereas the Californios, the Tejanos, and the New Mexicans insist that they are “Spanish” or “Spanish-American.” The trouble with all the terminology, as Dr. Campa puts it, is that it is based on logic and excludes the human factor. “The whole thing is characterized by anomalies which attempt to justify prejudices and defense mechanisms.” Certainly the attitude of the Californios, the Tejanos, and the New Mexicans has been a factor in the cultivation of an absurd dichotomy between things Spanish and things Mexican.

3. De Anza Doesn’t Live Here Any More Still another reason for the persistence of the fantasy heritage has been the negligible amount of immigration from Spain. The number of Spaniards in the United States, in or out of the borderlands, has always been so small that they have never been a factor in group competition. Hence it has always been possible to praise things Spanish without having to accept an embarrassingly large Spanish element. Despite the poverty of its population, Spain has always rigorously discouraged emigration. The number of Spanish-born in the United States was 22,108 in 1910; 49,535 in 1930; 109,407 in 1940 (half of whom were temporarily resident in this country)—0.3 percent of the population. Furthermore, it is a safe assumption that most of the Spanish born have always resided outside the Southwest. In Florida and Louisiana a few colonies survived after the Spanish withdrew. For example, St. Bernard’s Parish in Louisiana, named after Bernardo de Galvez, once governor of the province under Spanish rule, is still made up of the descendants of some 1,500 Canary Islanders—“the Isleños” as they are called—who settled there in 1770.

The Fantasy Heritage   23

In the Far West one can find a few colonies of Basque immigrants who came to this country by way of South America. Today some seven thousand Basques reside in the Boise, Idaho, area. They are, for the most part, descendants of immigrants who moved eastward from Jordan Valley in Oregon as the West Coast flocks began to be driven eastward for pasturage. Basques have been coming to California, in small numbers, for a hundred years. One reads that Yudarte, a Basque herdsman, once grazed hundreds of sheep in San Francisco between what is now Van Ness Avenue and the Presidio. Basque names like Duque, Echeverre, Mindiano, Hermasoillo, and Indiano are not uncommon in California. The Spanish Basques are to be found north of San Francisco, along the Sacramento River, and down the coast range; while the French Basques are concentrated around Fresno and Bakersfield. The Hotel Español and the Hotel De España in San Francisco have long been recognized as unofficial headquarters for the Basque sheepmen or boscos. There are, also, a few colonies of Spanish-speaking Puerto Ricans in California, mostly in the East Bay area. Most of these immigrants came to California from Hawaii following the expiration of the labor-contracts under which they had been imported to the islands. Mingled with the Puerto Ricans are a few thousand Spaniards who managed to elude Spanish immigration inspectors by going to Gibraltar where they signed labor-contracts to work in the canefields of Hawaii. For the most part, however, Puerto Ricans are concentrated in the City of New York. Today it is estimated that Little Spain or Spanish Harlem has a population of about 350,000 Puerto Ricans. Most of the recent immigrants have come to New York in “bucket seat” planes at a fare of $72 from San Juan.4 To the number of Puerto Ricans one might add some 47,699 Filipinos, many of whom, of course, are Spanish-speaking. However none of these elements—Spanish immigrants, Spanish Basques, Puerto Ricans or Spanish-speaking Filipinos—figure at all in the Spanish-Mexican scheme of things in the Southwest. In Florida there are perhaps 30,000 Spanish-speaking people: around 15,000 Cubans; 8,500 Spanish-born; and small settlements of several thousand each of Minorcans and Spaniards in St. Augustine and a few other towns. With the exception of those in St. Augustine, most of these people are fairly recent immigrants. The movement of Cubans to Florida began in 1868, when Vicente Martínez Ybor, a Spaniard who owned a large cigar factory in Havana, opened a factory in Key West. He was followed to Florida by other factory owners who sought to evade the import duty on cigars. The struggle for independence in Cuba was, also, an important factor in stimulating this exodus. With the cigarowners came thousands of cigar-makers from Havana, Bejucal, San Antonio de los Baños, Guines, Santiago de las Vegas, and other small towns near Havana.5 When the cigar-makers first arrived, Tampa was “a wilderness settlement comprising some four blocks of houses.” As the cigar-making industry expanded, several thousand Spanish cigar-makers joined the colony.

24  North from Mexico

The Spanish-speaking colonies in Florida were, of course, “the cradle of Cuban Independence.” At one time the headquarters of the Cuban revolutionary movement was located in Key West. It was here that José Martí, “the George Washington of Cuba,” founded the Partido Revolucionario Cubano in 1892. Originating in Florida, sympathy for Cuban independence spread throughout the nation. Just as there were Americans who took part in the movement which finally won independence for Cuba in 1902, so there were a few Cubans who fought with the colonists in the American Revolutionary War. Today small, scattered colonies of Cubans are to be found outside Florida, chiefly in New York, Philadelphia, New Orleans, and in Atlanta. “Cuba,” writes Manuel Pedro Gonzales, “more than any other Spanish American country, has contributed to the evolution of cultural relations between the two races. From 1823 to the present time she had been a kind of connecting link between the two cultures; she has played the double role of interpreter and propagandist of both cultures. For more than a century her most distinguished poets, writers, historians, and philosophers have endeavored to disseminate knowledge in the United States about Spanish America and her intellectual life; at the same time these men, who were far more familiar with the culture of the United States than their colleagues south of the Rio Grande, have tried to interpret and reveal to them through translations and critical studies the literary and scientific wealth of this country.”6 Cuban cigar-makers founded the first trade-unions in the South. One of these organizations, Los Caballeros del Trabajo was a branch of the Knights of Labor; and, from 1886 to 1901, the dominant Tampa group was known as La Resistencia. In 1900, and again in 1910, the Spanish-speaking cigar-makers of Florida conducted militant strikes which were suppressed with great violence. They founded the cigar industry in Florida which in 1908 was valued at $17,175,000; some 10,500 employees of the industry were then receiving a weekly wage of $200,000. Important as the Florida colonies have been they have had no relationship with Spanish-speaking settlements in the borderlands. And it is a rare case, indeed, when a Spanish immigrant has established any sort of contact with the Spanish-speaking people of the Southwest. Dr. Joaquin Ortega, of the University of New Mexico, born in Ronda, Spain, is one of the few individuals in whose career the two traditions are linked. From 1820 to 1940, the total Spanish immigration to the United States was about 175,000, three-fourths of which came after 1900. During the first World War, a large number of Spaniards worked in the shipyards but returned to Spain after the war. Most of the Spanish immigrant colonies are located outside the borderlands: in West Virginia; Philadelphia; Cleveland; Newark; New York; and Tampa. There is, therefore, simply no relation between Spanish immigration and the Spanish-speaking minority in the Southwest.

The Fantasy Heritage   25

Quite apart from these factors there has long existed, as I have previously pointed out, a determination to subordinate the Spanish-speaking minority in the Southwest. One of the techniques used to effect this subordination has been to drive a wedge between the native-born and the foreign-born and to cultivate the former at the expense of the latter. To some extent, elements of the native-born have encouraged this strategy by seeking to differentiate themselves from the immigrants. By emphasizing the Spanish part of the tradition and consciously repudiating the Mexican-Indian side, it has been possible to rob the Spanish-speaking minority of a heritage which is rightfully theirs, rather in the same manner that Negroes have been robbed of their heritage. The constant operation of this strategy has made it difficult for the Spanish-speaking people to organize and it has retarded their advancement. One of the first conditions to an improvement in Anglo-Hispano relations in the Southwest, therefore, is, as Dr. A. W. Bork has suggested, to give back to Indio-Hispano citizens the heritage of racial pride of which we have robbed them and to teach Anglo-Americans to respect and honor this heritage.7 The first step in this direction is to get rid of the fantasy heritage, the latter-day version of the Spanish prologue, which has so perniciously beclouded relations between Anglos and Hispanos in the borderlands. Once this veil of fantasy has been lifted, it should be possible for both groups to recognize the reality of cultural fusion in the Southwest. This reality is to be sought, first of all, in the nature of the region in which the great bulk of the Spanish-speaking people have always resided.

3

The Fan of Settlement

Starting in California, run a line from Santa Barbara along the base of the Tehachapi Mountains, around the rim of the San Bernardino Range, and then across the desert to the border of Arizona; draw the line through the center of Arizona; embrace all of New Mexico, including the San Luis Valley and portions of southeastern Colorado; then move down the eastern boundary of New Mexico to El Paso. From El Paso, pull the line diagonally to San Antonio and from San Antonio to Corpus Christi on the gulf. Between this line—the northern border of the borderlands—and the present Mexican-American border, reside upwards of eighty-five percent of the Spanish-speaking people in the United States. Stated another way and with less exactness, the Spanishspeaking reside in a belt of territory about 150 miles in width, paralleling the border, and extending from Los Angeles to the Gulf of Mexico. This territory is the fan of Spanish-Mexican influence “north from Mexico” which spreads across the borderlands with the tip of the fan resting in New Mexico.

1. “Sunshine, Silence, and Adobe” As one can see by glancing at a map, the fan of Spanish-speaking settlement embraces a subregion of the American Southwest. Topographically, it is a fairly distinct region-within-a-region: the basin and range country which extends from Los Angeles to the southern tip of the Great Plains, including the Lower Rio Grande Valley in Texas. By no means uniform in terrain, climate, and vegetation, it does have certain distinctive regional characteristics. The surface is made up of mountain ranges and basins; the climate is subhumid, semiarid, or arid; and the land itself is largely treeless. Many varieties of cactus are to be found throughout this region of glaring sunlight and intense heat. The amount of rainfall is slight and the rate of evaporation is extremely high. The principal rivers are the Colorado, the Gila, and the Rio Grande, none of which is navigable. Water is of prime importance throughout the area. The mountains tend to be bare and rugged, harshly sculptured, violently silhouetted, seamed with

28  North from Mexico

deep arroyos. Embracing the old Spanish borderlands, or Spanish highlands as they were sometimes called, this is the region, as Charles Fletcher Lummis once said, of “sunshine, silence, and adobe.” Historically, it is much the older part of the Southwest, being roughly coterminous with the original area of Spanish settlement. In settling the borderlands, the Spanish moved north from Mexico up the V-shaped central plateau: from the Mesa Central to the Mesa del Norte to the borderlands. The Florida and Louisiana settlements were detached from New Spain, based on parent colonies in the West Indies; the borderland settlements were integrally a part of Mexico. Between St. Augustine and Santa Fe, the Spanish made little impression within the present boundaries of the United States; but the borderlands carry the indelible imprint of Spain and Mexico to the present day. “Institutionally,” write Messrs. Richardson and Rister,1* “the basin and range country is a land of peculiar interest. Its northern boundary constitutes an institutional fault line that divides the Anglo-American Southwest, on the north, from the country to the south, where Spanish and north European institutions are mixed. In the more favored localities the natural environment afforded by the basin and range country is similar to that in Spain and Mexico over which the Spaniards passed in their northward advance.” Settling the region when the Anglo-Americans were hardly aware of its existence, the Spanish planted their institutions so firmly “that the trace of the Spaniard and his Mexican successor can never be beaten out of the land.” Although geographically similar to Mexico, the borderlands were separated from the heart of Spanish settlement in Mexico by a belt of desert which served to isolate the borderland settlements in time and space. Just as the Pueblo Indians were isolated from Indians of a similarly high culture in central Mexico, so the Spanish settlers in the borderlands were isolated from the older and more populous Spanish settlements in Mexico. The borderlands were thus true borderlands in the dual sense that they extended as a wide belt between AngloAmericans, on the north, and the Spanish Americans on the south and in the further sense that they were separated from their parent colonies in Mexico by the great wastes of desert in northern Sonora, Chihuahua, Coahuila, Nuevo Leon, and Tamaulipas. The ban of desert terrain across northern Mexico always served as a more or less effective barrier to the extension of Spanish influence and control. There were no barriers of this sort, however, to the east or to the north; in fact the weight of geographical factors tilted the borderlands more toward the AngloAmerican than toward the Spanish-American sphere. With the opening of the

* From The Greater Southwest by Rupert N. Richardson and Carl C. Rister, The Arthur H. Clark Company.

The Fan of Settlement   29

Santa Fe Trail in 1820, residents of New Mexico could trade with St. Louis more easily than with Chihuahua. Similarly the Spanish settlers in California could trade with Boston merchants, via the clipper ships, more easily than they could with points in Mexico by the overland routes. While the weight of geographical factors ultimately severed the borderlands from Mexico, the fact that they were similar to and yet detached from Mexico is largely responsible for the persistence of Spanish cultural influences in the region. The similarity of environment accounts for the fact that Spanish institutions were readily adaptable to conditions in the borderlands and proved to be of lasting value. By the same token, however, the southwest was a new world for the Anglo-Americans, a world quite unlike that in which their institutions had been evolved. Learning to live in the borderlands was a novel experience for the Anglo-Americans and one that required a basic modification in their institutions and a slow adaptation of their imported cultural practices. In the ensuing competition, Spanish cultural traits showed a remarkable tenacity since they were better adapted to the environment and more firmly rooted in time. Not feeling at home in a semiarid environment, the Anglo-Americans were at first somewhat reluctant to settle in the Southwest. Had the region been more accessible and familiar, more fertile and inviting, every vestige of Spanish influence might have been quickly obliterated after 1848. But the isolation and aridity of the borderlands kept the Anglo-Americans out for a time and later limited their number in relation to the Spanish-speaking. Coincidental with the annexation of the Southwest, gold was discovered in California and the lure of gold pulled the tide of Anglo-American immigration through and beyond the borderlands. The discovery of gold in turn accounts for the fact that railroads did not penetrate the Southwest until the 1880s, so that the isolation of the region survived for twenty or thirty years after through rail service had been extended to the Pacific. The time-lag created by these factors naturally favored the survival of Spanish and Mexican influences. Later, when the tide of Anglo-American immigration began, the arid nature of the environment imposed definite limitations on the number, the spacing, and the character of the settlements which they established. Thus, in some areas, Spanish-speaking people continued to outnumber English-speaking for many years. The Spanish-Mexican settlements, moreover, were located in more or less compact clusters, as a consequence, again, of the nature of the environment. The cluster-like pattern of these settlements made, of course, for cultural cohesiveness and permanence of influence.

2. The Forgotten Link Under the terms of the Treaty of Guadalupe Hidalgo, executed on February 2, 1848, Mexico ceded to the United States a vast territory, including California,

30  North from Mexico

Arizona, New Mexico, and other large fragments, and also approved the prior annexation of Texas. The lands which Mexico ceded to the United States were greater in extent than Germany and France combined and represented onehalf of the territory which Mexico possessed in 1821. All citizens of Mexico residing within the ceded domain were to become citizens of the United States if they failed to leave the territory within one year after ratification of the treaty. Only a few thousand Mexican nationals, perhaps not more than 1,500 or 2,000, took advantage of this provision; the rest became citizens-by-default. The treaty also provided specific guarantees for the property and political rights of the “native” population and attempted to safeguard their cultural autonomy, that is, they were given the right to retain their language, religion, and culture. No provisions were made, however, for the integration of the native peoples as a group, as a society. While the treaty contained a promise of early statehood, the promise was not redeemed, in the case of New Mexico and Arizona, until some sixty-four years after the treaty was ratified. The reluctance to grant statehood to New Mexico and Arizona, moreover, should be compared with the alacrity with which California and Nevada were admitted to the union. Obviously the reluctance to admit Arizona and New Mexico was based on “skepticism as to the advisability of granting full civil rights to a people largely illiterate and of an alien culture.” The fact that statehood was so long delayed served to retard the “assimilation” of the native population and encouraged the survival of Spanish-Mexican cultural influences. There was still another factor, however, which made for the survival of these same influences. When the Treaty of Guadalupe Hidalgo was signed, approximately 75,000 Spanish-speaking people were living in the Southwest: around 7,500 in California; a thousand or so in Arizona; 60,000 in New Mexico; and perhaps 5,000 in Texas. The great majority of these people were of mixed Spanish-Indian blood. But there were then living in the Southwest some other people who also passed under American sovereignty with the ratification of the treaty, namely, about 180,000 Indians, not including some 72,000 Indians in California. These Indians constitute the forgotten link in Anglo-Hispano cultural relations in the Southwest. In 1848 the Indians of the Southwest, apart from the California Indians, could be divided into two broad categories: Pueblo and non-Pueblo; mesadwellers and desert-and-plains Indians. The Pueblo Indians were, of course, one of the few sedentary tribes of the New World. At the time the Spanish arrived, they were divided into four groups: the Pueblos along the Rio Grande; the Zuni in western New Mexico; the Hopi in northwestern Arizona; and the Pima in southern Arizona. These four branches of the Pueblo tribe lived at peace with the Spaniards for many years, becoming closely identified with them in blood, language, and religion. In fact the Spanish settlements were superposed on the Pueblo Indian villages and never extended beyond the

The Fan of Settlement   31

ambit of these villages. Although the four Pueblo groups differed in custom, language, and mythology, they had some traits in common: they lived in villages of stone or adobe dwellings; they were monogamistic; and they had practiced irrigated farming in the Southwest for centuries. The nomadic Indians of the Southwest—the “eastern” and “western” Apaches, the Utes, the Navajo, and the Comanches—were of an entirely different breed. For untold centuries, they had preyed upon the Pueblo Indians. For many years these nomadic and warlike Indians surrounded, bedeviled, and lived off the Spanish settlements just as they had always exploited the Pueblo villages. Never vanquished by the Spanish, they were the real masters of the Southwest in 1848; in fact the Spanish-speaking were everywhere on the defensive at the time of the Anglo-American conquest and actually occupied less territory than at any previous period. From the earliest date, the nomadic Indians had prevented the Spanish from either expanding or consolidating their settlements in the borderlands. The Apaches were in full possession of the desert territory which separated the California and the Arizona settlements. Their destruction of the Spanish outpost at the junction of the Gila and Colorado rivers in 1780 broke the link between California and Arizona. So complete was this severance that the earlyday Mexican residents of Tucson scarcely knew that such a country as California existed. The Apaches likewise held undisputed control of the territory separating the settlements in southern Arizona from those in New Mexico. As a matter of fact it was not until about 1880 that the United States finally succeeded in wresting control of this territory from the Apaches. Nor were the Spanish ever successful in attempting to open a through-line of communication between the New Mexico and the Texas settlements. Surrounded on three sides by nomadic Indians, the New Mexico colonists had great difficulty in maintaining the route between Santa Fe and Chihuahua. As late as 1848, the trail from San Antonio to Santa Fe ran south to Durango; from Durango to El Paso; and from El Paso to Santa Fe. The resistance of the nomadic Indians retarded Anglo-American settlement of the Southwest for several decades and thereby threw a mantle of protection around Spanish cultural influences and institutions. From 1848 to 1887, the Anglo-Americans were so preoccupied with the Indians that they had little time left to devote to the settlement of the region or the exploitation of its resources. For many years the federal government was forced to maintain a series of military posts throughout the Southwest which provided important local markets for the native New Mexicans and gave them the first real protection against Apache raids that they had ever known. Both under Spanish and Mexican rule, the borderland settlements were separated from each other and from parent settlements in Mexico as much by Indian hostility as by the nature of the terrain. On the other hand, it was the nature of the terrain, as much as

32  North from Mexico

anything else, which enabled the nomadic Indians to hold out for so many years against Spaniards, Mexicans, and Anglo-Americans.

3. Lands of the Spanish-Speaking The most graphic way to envisage Spanish settlement north of the Rio Grande is to imagine a fan thrust north from Mexico with its tip resting on Santa Fe. Gradually the fan unfolds—eastward to Texas, westward to California—with the ribs of the fan extending northward from the base in Mexico. Long a part of the social structure of the region immediately north of the present border, that fan-of-influence still rests on the land. Indigenous and indestructible, it is a basic aspect of the cultural landscape. Today an undetermined number of Spanish-speaking people reside in the old Spanish borderlands. The first attempt to estimate the size of this population was made in the 1930 census. This census included in the category of “Mexican” all persons “born in Mexico or having parents born in Mexico who are not definitely white, Negro, Indian, or Japanese.” Premised upon the assumption that the Mexican minority should be regarded as “a race,” this category was obviously misleading and inexact. It overlooked the fact that portions of the Southwest had been a Spanish cultural province for several hundred years. For example, the census returned a figure of 59,340 “Mexicans” for New Mexico; but Dr. George Sanchez demonstrated, through a school census, that the figure should have been 202,709—an underenumeration of 144,389. Similarly the census returned the number of Mexicans in Los Angeles as 97,116, when 385,000 would have been a more accurate estimate. Nevertheless an analysis of the 1930 census does provide a substantially accurate picture of the distribution of “Mexicans” in the United States. The total figure given by the census was 1,422,533—the third largest “racial” group in the nation. Nine-tenths of the Mexican population was found in five states: Texas, California, Arizona, New Mexico, and Colorado. Texas had 683,681 or 48.1 percent of all Mexicans living in the United States; California 368,013 or 25.9 percent; Arizona, 114,173 or 8 percent; New Mexico, 59,340 or 4.2 percent (a grossly inaccurate figure); and Colorado had 57,676 or 4.1 percent. Illinois returned 28,906—virtually all in Chicago; and Michigan 13,336—half of this number residing in Detroit. A language enumeration made in 1940, based on Spanish as the mother tongue, indicated that there were 1,861,400 Spanishspeaking people in the United States and that, of this number, 1,570,740 resided in Texas, California, Arizona, New Mexico, and Colorado. There is good reason to believe that the concentration of Mexicans in the Southwest is even more pronounced today than it was in 1930, since the Middle Western states furnished a relatively higher percentage of Mexican repatriados during the depression than did the states of the Southwest.

The Fan of Settlement   33

Inaccurate in other respects, the 1930 census may also be taken as a fairly reliable immigration census. In this census the Mexican population was divided into the following categories: first generation foreign-born immigrants 616,998 (43.4 percent of the total); native-born of foreign-born or mixed parentage, 541,197 (38 percent of the total)—in effect the second generation; and 264,338 or 18.6 percent native-born of native-born parents. The first category is substantially accurate; the second is reasonably accurate; but the third is quite misleading. For in 1930 there were probably more Spanishspeaking people in this category in New Mexico than the census returned for the nation. Of this category—“native Mexicans of native parentage,”—95.6 percent were concentrated in the Southwest. The numbers so reported, and the percentage which they represented of the Mexican total for each state, were as follows: Texas, 146,806 or 21.5 percent; Colorado, 32,956 or 51.1 percent; California, 29,138 or 7.9 percent; New Mexico, 25,586 or 43.1 percent; and Arizona, 18,955 or 16.6 percent. These figures are only accurate in the sense that they indicate that the “native-born of native-born parents” are most heavily concentrated in New Mexico, Colorado, and Texas. The importance of the Spanish-speaking group in the Southwest may be shown in the proportion that this group constitutes of the total population in the key states: 26.2 percent of the population in Arizona; 14 percent of the population of New Mexico (this is the census figure; the actual percentage for New Mexico would be much closer to 40 percent); 11.7 percent in Texas; 7 percent in California; and 6 percent in Colorado. Ignoring the 1930 census, one can estimate the total Spanish-American population of the Southwestern states somewhat as follows: Texas, 1,000,000; New Mexico, 250,000; Arizona, 120,000; California, 500,000; and Colorado around 90,000. These estimates are based on school censuses, local social studies, and similar sources. That two million is a conservative estimate may be shown by the fact that the National Resources Committee in 1938 placed the total for the United States at close to three million. Not only is the Mexican population overwhelmingly concentrated in the Southwest, but it is highly concentrated within the belt of territory which I have previously described. This is merely another way of saying, of course, that the bulk of the Spanish-speaking people are concentrated within the boundaries of the old Spanish borderlands. In each of twenty-four counties extending from Santa Cruz in Arizona to Willacy in Texas, more than fifty percent of the population is of Mexican origin. There is a tier of counties, predominantly Mexican, in southern Texas that is three-fourths as large in area as that of the New England states combined. Roughly speaking, there are ten counties in southern Texas in which so-called “Latin-Americans” constitute seventy percent or more of the youngsters of school age; twenty-five counties in which they constitute from fifty to seventy-five percent of the scholastics; and

34  North from Mexico

nineteen counties in which the percentage ranges from twenty-five to fifty percent. In New Mexico the Spanish-speaking element is highly concentrated in the Upper Valley of the Rio Grande—the Rio Arriba section where the first colonies were planted. In each of fifteen out of thirty-one counties in the state, the Spanish-speaking comprise fifty percent of the population and in each of seven counties they make up eighty percent or more of the population. In Colorado, the native-born Spanish-speaking element is concentrated in the San Luis Valley and the southeastern corner of the state, while the immigrant element is to be found in Denver and the northern sugar-beet counties. In Arizona the bulk of the Mexicans reside in the southern tier of counties along the border. Today as yesterday most of the Spanish-speaking residents of California are to be found in the southern counties. In 1920 these counties had seventy-eight percent of the Mexican population and this proportion would be higher today than in 1920. Under Spanish rule, the missions in Southern California were more prosperous than those north of the Tehachapi Mountains and had a larger Indian-Mexican population. By the time of the American conquest, two political parties had developed in the state: the arribeños or norteños (the upper or northern people) and the abajeños or sureños (the lower or southern people). This same north-south division became more pronounced under American rule, for the discovery of gold brought thousands of immigrants into the northern counties but left the predominantly Spanish-speaking counties in the south in relative isolation for many years. Thirty years after the discovery of gold, Los Angeles was still a small Mexican town in which Spanish was spoken almost universally, with all official documents, including city ordinances, being published both in Spanish and in English. Contrary to popular belief, the Mexican population in the Southwest is today predominantly an urban population. In 1930, for example, 723,428 “Mexicans”—fifty-one percent of the total in the United States according to this census—resided in urban areas. The Mexican population in the United States is thus much more highly urbanized than the population of Mexico. In areas outside the Southwest, the Mexican population is overwhelmingly urban: eighty-five percent of the seventy thousand-odd Mexicans in the Midwest reside in cities. Within the Southwest, the three largest Mexican urban centers are Los Angeles, around 385,000 (for the county area); El Paso, 58,291 (fifty-seven percent of the population); and San Antonio, 82,373 or thirty-six percent of the population. Next to Mexico City, Los Angeles has the largest Mexican urban population of any city in the world. Social as well as physiographic factors have influenced the concentration of Spanish-speaking people in the old borderlands. The most pronounced discrimination against Mexicans has always existed, not in the predominantly

The Fan of Settlement   35

Spanish-speaking or bilingual counties along the border, but in the areas further removed from the border. The existence of this invisible but highly effective social wall has kept both old-resident and immigrant Mexicans within the familiar, and, to some extent, more congenial social atmosphere of the old borderlands.

4. Mexico Is Not Europe Since the tendency of Spanish-speaking people to concentrate in the Southwest appears to be permanent and cumulative, it is reasonable to assume that the small islands of Mexican settlement outside the region will soon be absorbed in the general population. Studies of these communities have shown that they are rapidly losing their distinctive Mexican characteristics. But the situation is quite different in the Southwest. Here the Spanish-Mexican influence cannot be “beaten out of the land.” For the Mexican minority—actually a majority in some areas—occupies a unique relation to the land, the culture, and the institutions of the region. Like the Indians, the Mexicans were “here first.” It is most misleading, therefore, to assume that they occupy a relation to the majority element which is like that, say, of Poles in Detroit or Italians in New York. Most Americans have been taught to think of immigration as a process by which Europeans picked up bag and baggage and came “to these shores.” From the immigrant’s point of view, the Atlantic crossing was of the utmost psychological and sociological importance; it was a severance, a crossing, an abrupt transition. But Mexican immigrants have seldom ventured beyond the fan of Spanish influence in the borderlands. They have been drawn to the borderlands by a feeling of continuity, of gradual transition, of movement within the confines, the protective mantle, of a familiar environment. The river which many of them have forded to enter the United States has a Spanish name. Most of the cities and counties in which they have settled, and even the streets on which they live, have Spanish names. One can travel from Chihuahua to Santa Fe with scarcely any feeling of abrupt change in the physical environment. Migration from Mexico is deeply rooted in the past. It follows trails which are among the most ancient on the North American continent. Psychologically and culturally, Mexicans have never emigrated to the Southwest: they have returned. In many cases, they have returned for the second, third, fourth, or fifth time. It is altogether possible that there are immigrants from Sonora now living in Los Angeles whose grandparents or great-grandparents once lived in the old pueblo or who, as Sonora miners, made the long trek to the goldfields. “The old Mexican centers and the old routes,” as Semple noted, “have still the power to attract.” In migrating to the borderlands, Mexicans have not founded immigrant colonies so much as they have “moved in with their relatives.” In fact their

36  North from Mexico

in-laws and relatives are scattered all along the route. One can travel from Sinaloa or Sonora to Los Angeles, or San Antonio, speaking Spanish the entire distance, moving continuously within the fan-spread of Spanish-Mexican culture, and living throughout the journey among Spanish-speaking people. Anglo-Americans emigrate to Mexico; but no Mexican is really an “immigrant” in the Southwest. The key to this distinction is to be found in the nature of the “border” which separates Mexico from the United States—one of the most unrealistic borders to be found in the Western Hemisphere.

5. The Border of the Borderlands From Brownsville to San Diego, the present boundary between Mexico and the United States is approximately 2,000 miles in length. From Brownsville the border extends up the Rio Grande to El Paso; proceeds westward from El Paso across some of the most rugged and desolate terrain in America; skirts the Gulf of California; and then extends westward to the Pacific. Not too much was known about this long stretch of territory when the Treaty of Guadalupe Hidalgo was signed in 1848. The negotiators did fairly well in fixing the line from Brownsville to El Paso—they simply followed the Rio Grande. But at El Paso they got lost, for a subsequent survey revealed that the line described in the treaty had been made with reference to a map which proved to be inaccurate. Once this discovery was made, Texas threatened New Mexico; New Mexico threatened Texas; both threatened Mexico; and Mexico sent troops to the border. For a moment or more it looked as though the Mexican-American War of 1846 would be resumed. To escape from this impasse, the United States sent James Gadsden to Mexico City to negotiate a new acquisition. By the use of high-pressure methods already painfully familiar to the Mexicans, Gadsden managed to secure for us another bite of Mexican territory—45,532 square miles in size. The original line—that defined by the treaty—would have followed the Gila River in Arizona. By the time Gadsden arrived in Mexico, however, we had become interested in the possibility of a rail line along the southern route to the coast. Hence the line that Gadsden finally accepted ran south of the Gila and just managed—by pure accident—to include in the United States, by a mile or more, some of the most important copper-mining properties in the world. As finally fixed, the border was a border of the borderlands rather than a national boundary based on economic and ethnic factors. Economically, the line made very little sense. By including Guaymas in the Mexican zone— the Mexican negotiators tried their best to force Guaymas on Mr. Gadsden— the interior basin was deprived of what might have become an extremely important seaport and the naturally advantageous and long-established

The Fan of Settlement   37

commerce between Sonora and Arizona was disrupted. This circumstance has given rise to various ambitious schemes, originating in Arizona, for the purchase of Lower California. The present border also left the whole question of water rights—a matter of vital importance—suspended in midair. From 1848 to the present time, Mexico and the United States have been at loggerheads over the water rights of the Colorado and Rio Grande rivers. From the point of view of the borderlands, the boundary should have been pushed further south so as to have included within the United States Lower California, the port of Guaymas, and all lands lying within the watershed of the Colorado and Rio Grande river systems. “The Rio Grande,” wrote Ellen Churchill Semple in American History and Its Geographic Conditions (1903), “is anything but a satisfactory boundary between the United States and Mexico. Dry for many months of the year, it bears no semblance of a barrier . . . . The political frontier line which is run along a river is an artificial one, for every drainage system forms an unbroken whole.” It should be noted, also, that the Mexican population living north of the border is, even today, somewhat larger than the Mexican population in the border states of northern Mexico. At various periods, this circumstance has filled the Anglo-Americans with fear and apprehension and has made for troubled relations between Mexico and the United States. As between Mexico and the United States, the movement of population has been consistently northward: into the northern Mexican border states and from there into the borderlands. The reverse movement, that is, of Anglo-Americans into Mexico, has always been negligible. Ethnographically, therefore, the border might well have followed the northern line of the old borderlands from Santa Barbara eastward, as traced in the first paragraph of this chapter. Due to the isolation of the region, the border for many years was merely a line on a map. The Border Patrol of the Immigration Service was not established until 1924. As late as 1911, a few Mounted Watchmen, as they were called, patrolled this two-thousand-mile-stretch of territory most of which was desolate and sparsely populated. Prior to 1924 the border could be crossed, in either direction, at almost any point from Brownsville to San Diego, with the greatest of ease. While the border was at least visible along the Rio Grande, it was by no means permanent. For the river has constantly changed its channel; islands formerly in Mexico have passed over to the American side; and new islands have been formed. From El Paso to Brownsville, the Rio Grande does not separate people: it draws them together. Along the river, as along the entire border, the towns are twins, and Siamese twins, in some cases, for many of them have interconnecting communications. Below El Paso, Ysleta is linked with Sargossa, San Elizario with Loma Colorado, Del Rio with Villa Acuña, Eagle Pass with Reimosa, Brownsville with Matamoros. El Paso and Juárez are, of course, essentially one

38  North from Mexico

community. Throughout south Texas, back from the border, most of the towns are twins: an American town and a Mexican town being joined together. Speaking of the residents of El Paso and Juárez, Walter Prescott Webb has said that “the river instead of separating them, rescued them from the desert and bound them together; all depended upon it both for domestic purposes and for irrigation of their meager crops; across its muddy channel the Mexicans intermarried, celebrated the same festivals, observed the same religious rites, rejoiced in the same feast days, and shared their sympathies, passions, and prejudices.” Westward from El Paso, the desert mines along the border draw people together much as the Rio Grande does from El Paso to Brownsville. Here, again, the towns are twins, sometimes having a similar name: Douglas, Arizona, and Agua Prieta, Mexico; Naco, Arizona, and Naco, Sonora; Nogales, Arizona, and Nogales, Mexico; Calexico, USA, and Mexicali, Mexico. Prior to the time when the government required sixty feet of “free space” along certain sections of the border, the bordertowns not only overlapped but the mythical line ran through particular stores, buildings, and saloons. Many stories are told, for example, of the old Brickwood store in Nogales, the counter of which was right on the line. For years the proprietor evaded the tax systems of both countries by stepping to one side or the other of the counter in making sales. Wanted by Mexican authorities lined up outside an adobe house in Nogales, an Arizona cowboy once escaped by sawing a hole in the rear wall and stepping out on American soil. Unreal in every sense, the Mexican-American border has greatly influenced relations between the two countries and has profoundly affected AngloHispano relations throughout the Southwest. In summarizing the issues which have disturbed Mexican-American relations through the years, J. Fred Rippy has given top priority to the unreality of the border. Between Mexico and the United States, he writes, “there have been no natural barriers, the two nations being separated by an imaginary line, a barbed-wire fence, an easily forded river, an undergrowth of mesquite or chaparral. Citizens of both nations have passed back and forth with little difficulty or interruption, or have settled in neighboring states amidst natural surroundings which have not repelled them by their unfamiliar aspects. Bandits, filibusters, and Indians have raided freely back and forth. Smugglers have often plied their trade with ease and security. Robbers of stock sometimes have been able to operate on a large and profitable scale. Political insurgents and refugees have often sought and found safety across the international line.”2 In short, the old Spanish borderlands were well named. The dictionary definition of “borderland” is the land lying along the frontier of two adjacent countries; but, in this case, the borderland runs along only one side of the frontier and that side is in the United States. I have often speculated on what form relations between the United States and Mexico might have assumed if

The Fan of Settlement   39

there had been no Spanish-speaking people or settlements within the borderlands. Had this been the case, the chances are that more formality and less friction might have prevailed. But, in the long run, the distance between the two countries might have been greater; the sense of separation more pronounced. While relatives who live in widely separated communities may quarrel less than those who live on the same street or in the same house, it is doubtful if much real affection or understanding ever develops between them. Anglos and Hispanos have fought and quarreled along the border as only close relatives can quarrel; but they have not faced each other across a fixed boundary with the sullen and undying enmity that the Germans and the French have faced each other across the Rhine. Borderlands unite as well as separate; they make for fusion rather than total acceptance or rejection. Differences tend to shade off in such a complex manner that soon various combinations of the two major types have appeared and with the emergence of these intermediate types the two antithetical elements have been inextricably bound together. There is no stronger bond between Mexico and the United States today than the living and organic union of the two cultures which exists in the borderlands. The process by which this union has been effected can never be reversed for it is a product of the similarity, the oneness, of the environment.

4

Heart of the Borderlands

“The most cohesive Hispanic population in the United States,” writes Dr. Joaquin Ortega, “the one most faithful to a long and uninterrupted tradition of identification with the soil is to be found in New Mexico.” The New Mexico settlements are, of course, among the oldest in the United States. Founded in 1609, Santa Fe is the oldest capital in the nation and, next to Florida, New Mexico is the oldest “state.” Over a period of a hundred years, from 1846 to 1946, the population of Santa Fe only increased from 6,000 to 20,325—old roots, slow growth. Like the dwarf evergreens on the surrounding hills, however, these roots have acquired a remarkable strength and sturdiness.

1. A Lost World In isolation, a people identified itself with its environment. dr. george sanchez

Isolation is the key to the New Mexico cultural complex. “The deepest penetration of civilized man in North America,” New Mexico was a lonely outpost of Spanish settlement for three hundred years—isolated from Mexico, California, Texas, and Arizona; isolated by deserts, mountain ranges, and hostile Indian tribes. It would be difficult, in fact, to imagine an isolation more nearly complete than that which encompassed New Mexico from 1598 to 1820. For its isolation was multiple and compound: geographic isolation bred social and cultural isolation; isolated in space, New Mexico was also in time. Primitive means of transportation and the lack of navigable streams extended distances a thousandfold. It took the New Mexicans five months to make the 1,200-mile round-trip, along the Turquoise Trail, from Santa Fe to Chihuahua. On the west, the north, and the east, the settlements were hemmed in by warlike nomadic tribes whose presence in these areas isolated New Mexico more effectively than distance or the lack of natural communications.

42  North from Mexico

To appreciate the degree of isolation, it should be noted that the Spaniards settled only a small portion of New Mexico. The early settlements were all in the Upper Sonoran life-zone in which agricultural possibilities were narrowly limited but where the Spanish did find a limited source of wood, fairly good forage, and a dependable water supply. Ever-mindful of the fact that fully half the men who came to the New World died of hunger in the first thirty years, the Spaniards in New Mexico followed the age-old device of conquerors: they made the conquered work for them and support them. Hence they settled near the Pueblo Indian villages. Down to the summer of 1608, they had failed to make a crop and were entirely dependent upon the resources of the Indian villages. Later, when small outposts were established along the streams of the Upper Rio Grande Valley, the new settlements were often as shut off from each other and from the major towns as the towns were from the nearest settlements in Mexico. The north-central part of New Mexico is rugged and mountainous and communication between the villages has always been difficult. Many of the villages in this area are severely isolated to this day. In the Upper Rio Puerco Valley, it still takes the villagers of Guadalupe two hours, under favorable conditions, to make the twelve-mile trip to Cabezon. Since the irrigable lands were few and far between, the settlements were noncontiguous and small in size. Hidden in mountain valleys, many of these settlements were tiny worlds-tothemselves. The coming of the railroad had little effect on the mountain villages some of which can only be reached today by barely passable roads. At the present time, eighty percent of the Spanish-speaking people of New Mexico live in villages of less than one thousand population. Automobiles, radios, and telephones are still rare items in many villages and studies made as late as 1940 have shown that eighty-five percent of the villages in the more inaccessible areas receive virtually no mail. Since the imported cultural pattern heavily stressed traditional values, it is not surprising that a homogeneous culture should have developed in an isolation so severe or that this culture should have shown such a sturdy resistance to change. Protected from outside or alien influences, the culture became highly integrated and relatively static. “Custom ruled with a blinding force” and innovation was suspect. Over the years the people themselves became genetically homogeneous. In many respects, as John Russell has pointed out, the “people are alike physically and in behavior. Born in the same region, they came from parents alike in race, and in physical type. Although there has been considerable intermarriage with the Indian tribes at present the descendants seem to have taken on characteristics which differentiate them from the Indians. They learn the same language and have similar cultural forms. Thus, they think alike, talk alike, differing only as individuals, but more alike in their social behavior than different.”

Heart of the Borderlands   43

With many of the villages being settled by a few families—in some cases by a single family—there naturally has been much marriage among blood relatives. Intermarriage and social isolation have made for an extremely cohesive family unit. To be born into a community is to inherit an identification with it that is never forgotten. With satisfactory companionship being found within the enlarged family group, individuals were seldom attracted outside the village in which they were born. Being largely self-sufficient, there was little specialization of function or division of labor. Much of the work was performed on a communal basis and the villages were generally of such a size that little outside help was necessary in the performance of routine tasks. Thus most contacts were limited to a single, uniform culture type. Many of the mountain villages have never known anything but poverty so that poverty itself has become an isolating factor in their lives. “Poverty,” as Dr. Paul Walter has said, “is part of their cultural inheritance.” The religion of the villager has also been a factor making for social isolation since it has always been a central, unifying, cohesive force in their culture. Jealous of their loyalties, it has deeply penetrated every aspect of their existence and has been a powerful shield against intrusive alien influences. The color of its pageantry, the mystery of its rituals, and the dramatic character of its ceremonies have always been potent attractions to lonely settlers in a forgotten world. Under the circumstances, its value as pure entertainment has been, perhaps, the principal explanation of its survival and dominant influence. In most instances, also, the priests were the only educated or learned men in the province, a circumstance that naturally gave added weight to their edicts and pronouncements. The practices of the Inquisition were not unknown in New Mexico. “Independence of thought or action,” writes Dr. Carolyn Zeleny, “brought heavy punishments.” In the isolation of New Mexico, the power of the Church was greatly magnified, for there was no escape from its rulings. The Church was also an enormous economic burden on the people. Its fees, tithes, and other exactions were truly exorbitant in a society in which money was always scarce. Superior both to the military and to the administrative bureaucracy, the Church was the dominant institutional influence in the lives of the people. The skeins of its influence were woven, from the earliest date, into the social fabric of the province. The institutions of the family and the Church were, in turn, closely interlocked with the patron-peon relationship. Based on tradition and authority, each institution supplemented the other and made for social isolation. By tradition, leadership rested in the priest, the patron, and the head of the family. Being institutional rather than personal, leadership tended to be noncompetitive. In the mountain villages today, the viejo or “old man” and the vieja or “old woman” are consulted on all matters of importance and their decision is usually final.

44  North from Mexico

Similarly the patron and the local jefe político still retain much of their traditional authority. At nearly every point of contact, the insulated character of the culture re-enforced institutional authority. “Peonage,” wrote Josiah Gregg, “acts with terrible severity upon the unfortunate poor, whose condition is but little better, if not worse indeed than that of the slaves of the south.” And peonage could no more be avoided than the power of the Church, for the nomadic Indians were the most efficient constabulary that despotic authority could desire. No prison was ever guarded more effectively than the Utes, the Apaches, and the Comanches “guarded” New Mexico. “Indian warfare,” writes R. E. Twitchell, “sapped the very life-blood of the intrepid settlers for more than two hundred and fifty years.” Generations were born with a mortal fear of Indians and lived and died with this fear uppermost in their minds. The hatred of the Apaches for the New Mexicans and of the latter for the former reached a point of ferocity beyond which it is impossible to go. The Indians were still making devastating raids on the New Mexico settlements in the 1860s and in 1879 alone over a hundred expeditions were sent out against the Apaches. Between 1861 and 1870, the federal government spent $40,000,000 and lost one thousand lives in its duel with these same Apaches. Contemptuously referring to the New Mexicans as their “shepherds,” the nomadic Indians harried the settlements on all sides; prevented the establishment of communications with other outposts in the borderlands; restricted the area of settlement; drained off the energy and wealth of the settlers and robbed them of incentives as well as goods and livestock. Rarely attacking a hacienda, the Navajo relentlessly pursued the poor New Mexican villagers and drove the peons “under the shadow of the great houses.” Indeed, most of the other factors mentioned in this section pale into insignificance when compared with the influence of Indians in the production of an inbred, isolated, homogeneous culture in New Mexico.

2. The People Today there are approximately 250,000 Spanish-speaking people in New Mexico most of whom are native-born of native-born parents, grandparents, and great-grandparents. With little latter-day immigration from Mexico, the population is indigenous to the region. The number of Spaniards was never large and for many years they were outnumbered by the Pueblo Indians in the ratio of ten to one. Dr. Zeleny puts the number of Spaniards at 2,400 in 1680 and around 3,479 in 1750. In fact it was not until about 1800 that the SpanishMexican group achieved a numerical ascendancy over the Pueblo Indians. “In every frontier Spanish colony,” writes Bolton, “the soldiery was to a large extent made up of castes—meztisos, coyotes, and mulattoes—and New Mexico was no exception to the rule.” Tlascalan Indians accompanied De Oñate as

Heart of the Borderlands   45

well as a number of Mexico-Indians who served as muleteers, packers, camp attendants, herdsmen, and drivers. While the Spanish soldiers and colonists were supposed to live apart from the Indian pueblos, the rule was never enforced. According to Gregg, an entire Indian village often abandoned its seclusion and became identified with the conquerors. Most of the hybrid population, however, developed out of the traffic in Indian slaves. “Indian women,” to quote Bolton, “were required for household service, with resulting scandals.” The practice of taking Indian captives, moreover, had a long history in New Mexico. As late as 1866, an Indian agent complained of the “pernicious system of slavery.” Over the years, therefore, the amount of Spanish blood declined in direct relation to the increase in population. By 1822 the population had increased to forty-two thousand but only a small portion of this total could be regarded as Spanish in ancestry. In the preceding quarter-century, the Pueblo Indian population had remained nearly stationary while the number of mixed-bloods had rapidly increased. It is quite obvious, therefore, that the present Spanishspeaking population is of a very mixed racial origin with the Spanish strain being the least important element in the mixture. For example, Gregg estimated the number of creoles or Spaniards at one thousand in 1846 by comparison with sixty thousand mixed-breeds. In the Rio Arriba section the Indian strain was more pronounced, of course, than in the Rio Abajo, where the large landed estates were located. From an early date the population of New Mexico was divided into two major classes: ricos and pobres, the rich and the poor. To some extent, the division marked a caste as well as a class differentiation, for the ricos were “lighter,” more “Spanish,” than the pobres. Holding the reins of social, economic, and political power—the beneficiaries of the large land grants—the ricos were a law unto themselves. Constituting one-fiftieth of the population, they owned all that was worth owning and were autocrats in every sense of the word. Theirs were the great estates and the vast herds of sheep in the Rio Abajo section. The soldiers, artisans, and peasant farmers were allotted small family and community grants in the Rio Arriba and worked, often as peons, on the large estates. Always quite rigid, the barriers between the two groups became more effective as the lower classes mixed their blood with the Indians. From the founding of the colony until 1846, writes Haniel Long, the ricos grew richer, the pobres poorer, and “the priests lazier and more avaricious.” The Navajo preyed on the rich, who in turn preyed on the poor, “and the poor could prey on nobody.” Harried by Indians, the poor were driven to the rich for protection; protection implied peonage; and from peonage there was no escape. In one sense, however, even the ricos lived in a state of bondage. For them everything was cheap and nothing quite so cheap as labor. They had hardly more incentive for progress and activity than the poor. In such a society there

46  North from Mexico

was really little need for literacy and most of the population, of course, was illiterate. There were no schools or newspapers and few books. The government was highly personalized in character and absolutist, so that little competition for place or power could take place. Don Pedro Pino was the first native-born New Mexican to visit Spain and his visit took place in 1810. While a few of the ricos sent their sons and daughters to St. Louis and other Missouri Valley points for their education, the ricos were certainly not an educated class. There was no tradition of self-government or, as Blackmar said, “no rights which arose out of the situation.” An appointed governor appointed the provincial officials, and much the same arrangement prevailed under territorial rule. The New Mexicans took no part in the movement by which Mexico achieved its independence and only once, in 1837, was the province stirred by a revolution. Throughout the period of Mexican rule, the province was corruptly and despotically “bossed” by one man, Manuel Armijo. In this intellectually airless world, the ricos developed a life of their own. Sociability, within the class, was a necessity born of isolation. The families, large in size and ruled in patriarchal style, saw much of one another. Built as fortresses against the Navajo, the great low houses of the ricos, shaded by cottonwood trees, were scattered along the highway near the river within easy traveling distance. There was little furniture in the homes, for a lack of tools and cabinetmakers had forced the colonists to adjust to a rather primitive existence. Chairs, bedsteads, and tables were rare items and, in eating, the ricos squatted on the beaten-earth floors “a l’Indienne.” Walls three and four feet in thickness enclosed a courtyard or patio, behind which was still another square used as a quarter for the slaves and peons and a place for carts and wagons. “Each great house,” writes Harvey Fergusson, “reproduced the isolation which beset the colony as a whole.” The New Mexicans knew little of the mechanical arts and much of what they made was rough and unfinished in character. Sawmills were unknown; there was little mining; and agricultural implements were largely limited to the hoe and wooden plow. Nor were they an agricultural people. Such agriculture as they developed rested solidly on Pueblo Indian foundations, with the principal crops being such Indian staples as corn and beans. “Of fruits,” writes Twitchell, “there were practically none.” Wheat was cut by hand with a scythe and threshed by turning livestock loose in a compound. In the mountain villages, grain was ground between large pumice stones run by primitive water wheels, some of which are still in use. Essentially a pastoral people, the New Mexicans showed great skill and enterprise in the management of their herds. Like their counterparts in California, the ricos were superb horsemen. Of commerce there was very little, for the economy of the province was a cambalache or barter economy. To the annual fair at Taos, the New Mexicans brought horses, mules, knives, hatchets, and trinkets which they traded to the

Heart of the Borderlands   47

Indians for captive children and the skins of deer and buffalo. In the Chihuahua markets, a handsome price was paid, in barter terms, for the few luxury items which could be imported. Once partially free of colonial restrictions, the New Mexicans did conduct a considerable commerce with the Indians on the eastern plains and deep in the Utah basin. The two annual fairs; a trip to the salt mines; the buffalo-hunting expeditions in the fall; and the Indian campaigns made up the principal items in the routinized existence of the ricos. As a class, the ricos have largely vanished. According to Erna Fergusson, there are not more than a dozen of the old families left. Not one of the great houses has been preserved. In disappearing, the ricos even failed to leave much in the way of memoirs, records, letters, or memorials: “the only form of life in the Southwest,” writes Haniel Long, “that has left no ruins.” The mission system, as it was known in California, never existed in New Mexico. “There were no mission estates,” wrote Bancroft; “no temporalities managed by padres.” Each pueblo had a church, where the padres preached, taught, and said mass; but the Indians were left in their villages with the padres being essentially parish priests or curates. Prior to Bishop Lamy’s arrival in 1851, the New Mexico priests were indescribably corrupt, lazy, and avaricious. Most of them had numerous wives and a vast collection of children whose paternity was hardly a secret and not much of a scandal. While they did try to throw a mantle of protection around the Pueblo Indians, they lacked the enterprise of the Franciscans in the heyday of the mission system in California. The real settlers of New Mexico were the villagers, whose descendants still inhabit such fabulous mountain villages as Cundiyo, Cordova, Truchas, Trampas, Chamisal, and Penasco. Most of these villages were established as part of a series of outer defenses against the Indians. Here the soldier-settler was given a small grant in payment for his services in manning a lonely outpost against the Indians. Many of the mountain villages were based on community or family grants. Often the irrigable lands were granted in individual ownership with the grazing lands and wooded portions being held in common. In addition to tending their small acres, the villagers herded sheep, hauled wood, campaigned against the Indians, and drove the large herds and the carts to the market fairs. They milled their own grain; wove their own cloth; made their own tools, cooking utensils, household goods, and furnishings; and built their own adobe homes. Probably as much of their inheritance is Indian as Spanish: Spanish their language—a baroque sixteenth century Spanish—and their religion; but Indian their knowledge of hunting, farming, and the ways of the land. Their attitude toward land tenure, for example, is quite similar to that of the Indians. These are the real peasants, the paisanos, the men of the country. And it is through them that Spanish-Mexican influences have survived in New Mexico.

48  North from Mexico

No new currents of life moved in this remote colony of Spain for nearly three hundred years. Education had little meaning in a society in which there was literally nothing to learn. Competition and change, initiative and innovation were, for similar reasons, mostly nonexistent. The life of any today was the same as the most remote yesterday that anyone could remember; and tasks were performed as they had always been performed. Aside from the Indian influence, the society was, in Haniel Long’s phrase, “a huge room in the Southwest hermetically sealed so far as any vital touch went with the life they had left behind in Mexico and Spain.” Over the mystery of the slow growth of New Mexico, writes Ross Calvin, “there really is very little mystery after all”: poverty and isolation explain whatever was mysterious.

3. The Flowering of New Mexico Such terms as “isolation” and “poverty,” however, must be qualified as applied to the Spanish settlements in New Mexico. Poverty is always relative to historical and social circumstances. If the New Mexicans were poor, their poverty originally had little meaning and was hardly “poverty” as we understand the term today. For if money was scarce, there was little to purchase. Isolated the settlements never were in any complete sense, for the Pueblo, the Navajo, and the other Indians colored the pattern of the culture from the earliest date. As the settlements became more firmly established, more of the energy of the people began to flow into ornament, design, and decoration. The best that they achieved may have been unsophisticated and naive, and somewhat primitive, but it was certainly not lacking in grace and fine feeling. In the relative isolation of the province, a folk culture developed which, in the period from the opening of the Santa Fe Trail to 1846, achieved a brilliant, if brief, flowering. In any number of crafts, the New Mexicans achieved real artistic distinction. In the so-called “Rio Grande” silverwork, fashioned into dinner services, candlesticks, household ornaments, and picture frames; in the elaborate wrought-iron designs; in the tinwork with which the pobres sought to imitate the silver ornaments of the ricos; in the wood-carving on the old cupboards or trasteros, chests, chairs, doorways, and the lintels, corbels, and posts; in all types of elegant needlework; and in the beautiful weaving which centered in the village of Chimayo, the New Mexicans made a contribution to the colonial crafts as interesting and as distinctive as that which developed on the eastern seaboard. Filigree work, in silver and gold, was another highly developed craft. Images of saints were painted upon small panels of wood (santos de retablos) and carved from wood (bultos or vueltos). Some of the santeros, or saintmakers, were indeed extraordinary workmen. Considering the limited number and crude character of the tools available, much of this craftwork was amazingly fine in design and execution. What emerged in the way of arts and crafts

Heart of the Borderlands   49

was, as Frank Applegate has pointed out, Indian in feeling, Spanish in plan. Creating from memory, working with crude tools, improvising materials, the New Mexicans were consistently under a heavy debt to the Indians who, in turn, borrowed freely from them. Rediscovered after 1918, the SpanishColonial arts and crafts have since enjoyed a great vogue. It is significant that saint-making and weaving—two of the most highly developed arts in New Mexico—reached their peak in the period from 1830 to 1835. This would seem to indicate that the stimulus of trade and the new wealth that came with trade had momentarily opened up new markets for the arts and crafts. Since 1890, writes Ruth Laughlin, there has been a decreasing demand for wooden santos, although a few of the santeros still practice their craft. Spinning, dyeing, and needlework are lost arts; furnituremaking largely ceased with the appearance of the Anglo-Americans; while the main arts to survive are those of weaving and working in tin, iron, gold, and silver.1 Since the Indian influence was much stronger in New Mexico than elsewhere in the borderlands what emerged there in the way of architectural forms is both more interesting and more usable than what currently passes for “Spanish” or “Spanish-Colonial” in Texas, Arizona, and California. To the use of adobe, the Spanish brought the practice of molded and sun-dried bricks to replace the puddling system which the Indians had previously used. To the adobe house, also, the Spaniards added doors, windows, stairs, fireplaces, and flues. Rescued from a threatened oblivion by the Santa Fe renaissance which began about 1910, this Pueblo-Spanish house, as improved and adapted by Anglo-Americans, is a most attractive form, simple, sturdy, well-adapted to the environment. In California nearly everything that passes architecturally by the name of “Spanish” stems from the rash of Spanish-Colonial stucco-and-red-tile construction that swept the region after the San Diego World Fair of 1910. It is primarily the Indian influence, the influence that made for adaptability, that is lacking in this ostentatious pseudo-Spanish architecture. While the Spanish missions of the Southwest have great historic interest and some slight merit as structures, they quickly became a deplorable architectural influence throughout the region. “More architectural crimes,” wrote Irving Gill, the California architect, “have been committed in the name of the missions than in any other unless it be the Grecian Temple.” Nevertheless in domestic architecture, and to some extent in public buildings, the Southwest has benefited from the adaptation of Indian-Spanish forms and usages. The folk tradition of the New Mexicans is, of course, rich and varied. Dr. Arthur L. Campa has made amazing collections of the folk songs and poetry— the décimas, corridos, and versos—of the Spanish-speaking and Dr. Aurelio M. Espiñosa has accumulated a library of the richest folk materials—folk tales, riddles, proverbs, myths, and children’s games. Some of the folklore current in

50  North from Mexico

New Mexico has long since been forgotten in Spain. Much of this traditional Spanish folklore was brought to New Mexico by settlers who had first lived for some years in Mexico. This circumstance is said to account for the fact that, in all the New Mexico folk materials, there is scarcely a mention of a king or queen or royal family. Generally speaking, the oral side of the tradition has been losing ground rapidly in New Mexico since 1910 and much of the authentic folk quality has been lost.2 In many respects, the arts, crafts, and folk culture of the New Mexicans is strikingly similar, in type, to the folk culture of the “mountain people” of the Southern states. Both represent curious survivals in the New World of Old World cultural traits and archaic modes of speech and expression. “A Castilian of the year 1525,” writes Mencken, “would understand a New Mexican far more readily than he would a Spaniard,” just as an Englishman of 1630 would understand a Kentucky mountaineer more easily than he would a resident of Louisville. Like all survivals of this kind, the folk culture of New Mexico seems doomed to extinction. Only a few of the crafts remain alive; tradition, custom, and folk expression are rapidly losing their indigenous quality; and archaicisms in speech are vanishing. In central New Mexico, where the AngloAmerican influence is most pronounced, Dr. Campa notes that “the whole manner of living is fast becoming Americanized, in some cases to the point where Spanish is no longer spoken in the home.” Where the language vanishes, the traditions and customs are soon forgotten. On the other hand, in southern New Mexico, near the border, Spanish has the vitality of a language that is living and active. This is the area of true bilingualism in New Mexico. For in northern New Mexico, English is spoken with a heavy accent just as in central New Mexico, Spanish is spoken with an English accent. What appears to be happening in New Mexico, particularly in the southern part of the state, is that encrusted Spanish traits, long preserved in isolation, are being replaced with traits more clearly Mexican in character. In effect the flowering of the older culture ceased when its social and economic underpinnings began to disintegrate under the impact of the Anglo-American invasion.

4. After the Conquest Motivated by a lust for conquest, Spanish colonization in the Americas was imposed from above. By comparison with the English, a large part of the Spanish expeditions were made up of the nobility and gentry, elements that loomed large in Spanish life. Since Spain had very little in the way of a middle class, middle-class elements were hardly represented at all in the expeditions that came to the New World. Few lower-class Spaniards settled in Mexico, South America, or in the borderlands. The nobles came to win fame and to

Heart of the Borderlands   51

replenish their fortunes, not to settle; and no succeeding waves of immigration fortified or revivified the culture which their conquests had succeeded in imposing on the native peoples. The absence of a Spanish-speaking middle class was one of the factors that complicated Anglo-Hispano relations in the borderlands. A large part of the Anglo-American influx to the borderlands after 1846 was made up of middleclass elements in the sense that they were neither very rich nor extremely poor. In the borderlands, these elements did not find their Hispanic “opposite numbers.” What they found, in Harvey Fergusson’s phrase, was “a small feudal aristocracy and an illiterate half-savage proletariat.” The absence of local selfgovernment and the presence of a population that was seven-eighths illiterate in 1850, predisposed the Anglo-Americans to form an extremely negative opi­­ nion of the Mexican lower classes who constituted nine-tenths of the population. If a larger middle-class element had existed, the adjustment between the two cultures might have been facilitated and the amount of intermarriage might have been greater. The subordinate status of the pobres in relation to the ricos, and their poverty, served to set them apart in a category that, in Anglo-American eyes, was roughly comparable to that of the Indians. By comparison with the pobres, the ricos made the transition to American rule with comparative ease. Some had traveled in the states and spoke the English language. Others were related to the newcomers by marriage. Becoming rich from the commerce which developed after the opening of the Santa Fe Trail, many looked with favor upon closer ties with the AngloAmericans. Twitchell reports, for example, that the wives and daughters of the ricos had begun to acquire elements of the language, style of dress, and mannerisms of the Anglo-American world prior to the conquest. With the arrival of American women in the province, this process was greatly accelerated. In 1866 Colonel J. F. Meline noted that the rebozo had almost disappeared in Santa Fe and that hoop skirts, on sale in the stores, were being widely used. The costume of the caballero had begun to disappear in the 1830s, with the serape and sombrero being the last to vanish. Geographically segregated from the Anglo-Americans, who moved into the south-central part of the state, the villagers were at first not greatly affected by the transition. In fact some improvement could be noted. New markets were opened; the Indian raids were eventually eliminated; and the railroads brought opportunities for jobs as construction workers, section hands, and maintenance employees. Up to 1880, there was little Anglo-Hispano competition for land or resources; but, with new markets, land values began to rise and the Spanish-speaking element began to feel, at a dozen different points, the pressure of Anglo-American competition. For a time commercial cattle-raising assumed something like bonanza proportions in New Mexico. The number of cattle increased from 160,000 in 1880 to one million head in 1900. Competition

52  North from Mexico

for grazing lands became keen, with control of “waterholes” being used by wily Anglo-Americans as a means of acquiring ownership of the available rangelands. A similar expansion took place in commercial sheep-raising. The Hispanos also began to feel the competition of dry-land farming which the Anglo-Americans introduced to the eastern portions of the state. Later, with the passage of the Reclamation Act in 1902, competition for agricultural lands became intense. The consequences of these changes, so disastrous in human terms, did not become fully apparent until the middle 1920s. In the end, the Hispanos were caught up in the meshes of Anglo-American banking, finance, and legal intrigue. Prior to the conquest, there had been no land tax in New Mexico; but, with Anglo rule, came taxes, litigation over land titles, mortgages, and the other incidents of a monetary economy. However, the process by which Hispanos were entangled in the cash system was much slower in New Mexico than in Texas or in California. The debacle that engulfed the Spanish-speaking in California in the first decade after 1848 did not become fully apparent in New Mexico until seventy years after the conquest. In New Mexico, land grants had been loosely defined, with boundaries often being fixed by reference to trees, rocks, and mountain peaks. Under Spanish and Mexican rule, this vagueness of boundaries had caused little trouble or confusion; for use and occupancy, rather than ownership, were the important considerations. Under American law, the filing of a claim based on a Mexican or Spanish grant automatically prevented the land from being considered a part of the public domain until the validity of the grant was determined; in the meantime, however, the lands could be grazed. As a consequence, of course, all sorts of bogus claims were filed. George W. Julian, appointed surveyor-general by President Cleveland, contended that gross frauds had been committed through the machinations of Anglo-Americans in close alliance with a few large Spanish-American landowners. While Twitchell refers to Julian as a “mountebank,” his charges were certainly not groundless, for Blackmar, some years later, also referred to “the wily intrigues” by which Anglo-Americans were exploiting the confusion in land titles. Many of the villagers neglected to bring their papers into court and often had lost evidences of title. Most of them lacked funds to defend titles; or, if they retained an Anglo-American lawyer, a large part of the land went in payment of court costs and fees. The confusion became so great that in 1891 a Court of Private Land Claims was established to pass upon the land grants in New Mexico. Needless to say, the members of this court were all AngloAmericans; and, as nearly as I can determine, there was not a single SpanishAmerican lawyer in the territory. Litigation over land titles was highly technical and involved; cases dragged on in the courts for years; and, in the general process of settling titles, control of resources shifted to the Anglo-Americans.

Heart of the Borderlands   53

Erna Fergusson has given a graphic description of one of numerous processes by which Anglo-Americans encroached upon the property rights of the Spanish-speaking people after 1846. Under Spanish law, the children inherited in equal shares, always with a small frontage on the stream or river. Consequently, even a small grant might in time have a hundred owners, each possessing a small strip of land running back from the river or “mother ditch” to the hills, while the grazing lands were held in common by all the heirs. Then some far-seeing Anglo-American stockman would purchase one of the individually owned parcels and claim an unlimited right to use the commonly owned grazing lands. In this way, she writes, “some men so achieved the use of millions of acres” for grazing purposes. Later the Anglo-American owner would induce the Spanish-Americans to petition for a division of the grant. In the end, title to the grazing lands had passed to the Anglo-American and the Spanish-Americans were left with their small irrigable plots and a portion of the proceeds from the sale of the commonly held lands. In many cases, the Spanish-Americans could not pay land taxes of $1.50 an acre, or more, levied against grazing lands. Anglo-Americans would then buy up the lands at tax sales and promptly have the land tax reduced to thirty or forty cents an acre. Often the villagers would be permitted, for a time, to continue grazing their small flocks on the range; but the pinch came with the appearance of ever-larger commercial flocks which monopolized the range. Overgrazing of the range, by large-scale commercial sheep-raisers, destroyed the cover of vegetation, ushering in a period of floods and soil erosion. In literally hundreds of cases, one can see this process at work in New Mexico. The Anton Chico grant near El Cerrito, embracing 275,000 acres, was confirmed in 1860. By 1926 the grant had become chronically tax-delinquent. As more and more land was sold to pay taxes, the grant was reduced to 85,000 acres by 1939, of which 22,000 acres were under lease. Seven hundred families were, by 1939, dependent on the remaining portions of the grant. For a time the Hispanos were able to eke out an existence by part-time supplemental labor: in the mines, on the railroads, and as migratory agricultural workers. By 1920 approximately 12,500 New Mexicans were leaving the state every year for seasonal work. But, with the depression, this number declined to 2,500. During the depression years from sixty to seventy percent of the villagers were on relief. It was around 1926, in fact, that a real “crisis” developed in the village economy of New Mexico. A few facts tell the story: in 1930, New Mexico, with a death rate of 13.8, had almost 3 deaths more per 1,000 population than the national average; the counties with the highest death rates were uniformly those where the Spanish-speaking people constituted more than fifty percent of the population (in Mora County almost eighty percent of the deaths were reported as occurring from unknown causes!); the counties with the largest Spanish-speaking population were the poorest counties—the

54  North from Mexico

higher the percentage of Spanish-speaking people the lower the per capita assessed valuation (in 1933, eighty-five percent of the taxpayers in the Spanishspeaking counties were assessed for less than $100). Compared with a national rate of 57 deaths per 1,000 live births, New Mexico had a rate of 118—the highest in the nation. In 1940 New Mexico reported 26,488 residents, twentyfive years of age or over, who had not completed one year of schooling.3 As this crisis developed, a change began to occur in Anglo-Hispano relations. For it was about this time that such terms as “Spanish-speaking,”   “SpanishAmerican,” and “Spanish-Colonial” came into use to designate the native New Mexicans. “Mexicans,” writes Erna Fergusson, “was the term universally applied to them within the memory of most of us. Suddenly, nobody knows just when or why, it became politic to use the hybrid-term, ‘Spanish-American.’” In general, this change coincided with a similar change in nomenclature in south Texas, where the term “Latin-American” began to be used about 1927. Writing in 1931, Ruth Laughlin said that it had only been “within the last generation” that the Spanish-speaking element in New Mexico had “rebelled at being called Mexicans and spoke of themselves as Spanish-Americans,” thereby reversing a usage which had been quite common for nearly a century. Mary Austin, concurring, said that the change dated from the first World War. While the change in usage is related to the large influx of Mexican immigrants—the native New Mexicans sought to distinguish themselves from the immigrants—its emergence also marks a growing self-consciousness on the part of the Spanishspeaking and a desire, on their part, to escape from a subordinate status. By 1918 or 1920, the first “school generation” had begun to reach maturity in New Mexico (the public school system dates from 1890); and the first World War had broken, at numerous points, the insularity of the state. During World War I, wrote Natalie Curtis Burlin, “the whole Southwest found itself abruptly seized by the collar and jerked out of its isolation.” By 1918, also, a middle-class business element had begun to emerge in New Mexico4; and, as elsewhere in the borderlands, prejudice has been most keenly resented by the emerging middle class. Having always been called “Spanish,” the rico element had experienced little discrimination or prejudice; while the poorer classes, being segregated by poverty and geographic location, had partially adjusted to a bicultural relationship. But the middle-class elements, small in number, lacked the social prominence to win exemption from discrimination and, at the same time, sought to distinguish themselves from the “Mexican” lower class. While the two upper-class groups have always hobnobbed together, exchanging compliments and courtesies and genuflecting before the “Spanish” monuments of the past, the respective middle-class elements have not gotten along so well together. Most of the service clubs in New Mexico—notably those in Albuquerque—systematically exclude Spanish-Americans and the Junior Service League of Albuquerque some years ago refused the application of

Heart of the Borderlands   55

Senator Dennis Chavez’s daughter. The fraternities and sororities at the University of New Mexico draw a sharp line against Hispano students; and a rigid taboo excludes these students from much of the social life of the campus. Carolyn Zeleny, who observed the campus life for two years, reports that the Spanish-speaking students do not, as a rule, attend school dances and that “an unwritten code does not permit an Anglo girl to date an Hispano boy.” New Mexico occupies a uniquely important position in the pattern of American culture. Protected by geographic, social, and cultural isolation, the Spanish-speaking element was given a sufficient margin of time in which to make the transition from Hispano to Anglo rule so that much of their cultural heritage has been preserved. The time-lag made possible the preservation of important elements of the culture through a process of slow adaptation. The semiarid character of the environment will continue to serve, as it has in the past, as a mantle for the protection of the Spanish-speaking people. The average density of population in New Mexico today is five persons per square mile, about one-ninth the density for the United States as a whole. One hundred years after the American conquest, the population of the state is only slightly in excess of five hundred thousand, and the Spanish-speaking element comprises fifty percent of this total. The Hispano element is too numerous, therefore, in relation to the Anglos, to be absorbed piecemeal and Hispano cultural influences are now too deeply impressed upon the land to be easily obliterated. Both in New Mexico and in Arizona, the Indian population should be regarded, in some respects, as part of the Hispanic element; for they are similar in racial background, language, and religion. Infant mortality rates are falling for both groups while their birthrates remain among the highest in the nation. Add the Indian population to the Spanish-speaking total, and it becomes quite apparent that Spanish-Indian-Mexican elements have a long life expectancy in the Southwest. New Mexico is the anchor for these elements: the rock upon which Spanish culture rests today.

5

The Broken Border

The Spanish scheme for colonizing the borderlands called for a strong central colony in New Mexico, the establishment of widely separated outposts in California, Arizona, and Texas, and, eventually, the linking of these settlements into a broad band across the northern part of New Spain. The central colony of New Mexico was finally anchored, after great effort, but more than a hundred years passed before the colonization of the three outlying provinces could be undertaken. While these salients were ultimately established, the colonies were never consolidated. During their existence as Spanish outposts, they went their separate and different ways, with little intercommunication or exchange; each with its own pattern, its own special problems. The failure of Spain to consolidate the borderland outposts has had important latter-day consequences. For the Spanish-speaking of the borderlands remain, to some extent, separate and disparate groups, sharing a common heritage but never having known the experience of functioning together. Spanishspeaking people in California know little of the experience of their compatriots in New Mexico; and those in New Mexico are unacquainted with conditions in Arizona and Texas. No effective liaison has ever existed between these groups; their experiences have run parallel but have never merged. For the border was broken, the links were never forged.

1. Pimeria Alta The Mission Nuestra Señora de los Dolores in Sonora was the “mother mission” for the settlements in southern Arizona. From this base, Father Kino established a chain of missions along the upper waters of the San Miguel, Altar, Santa Cruz, and San Pedro rivers: Guevavi (founded in 1692), Tumacacori, San Xavier del Bac. The discovery of huge silver nuggets just south of the present border in the Altar Valley, near a place which the Indians called Arizonac, encouraged the Spaniards, for a time, to make something of the southern Arizona settlements. During this spurt of activity, the missions in southern Arizona, or

58  North from Mexico

Pimeria Alta as it was called, enjoyed a brief period of prosperity and expansion. But they were virtually extinguished in 1751 when the Pima and Papago Indians joined in a general uprising. At this time, there were no Spanish garrisons north of Fronteras in northern Sonora and Pimeria Alta was almost wholly unprotected. In the following year, a garrison was established at Tubac in the Santa Cruz Valley and some of the priests, soldiers, and settlers returned. But the Apaches, who had been raiding the Sonora missions for a century, prevented any effective recolonization. In a report on the state of the province in 1777, De Anza pointed out that fifteen years of incessant Apache raiding had depopulated the settlements; that the mines, haciendas, and missions, if not ruined by these attacks, had been abandoned for fear of them. In a decade devoted to counteroffensives against the elusive Apache, the Spanish had managed to kill only 276 Indians. At this rate it was apparent that the Arizona settlements were doomed to failure. Always a lightly garrisoned province, the Sonora-Arizona frontier was exposed to renewed Apache attacks when the government withdrew its troops during the struggle for Mexican independence. Taking advantage of this opportunity, the Apaches laid waste to the entire province: mines were abandoned; haciendas were deserted; and any stock that the Apaches failed to kill roamed wild along the border. In the wake of this disaster, the discovery of gold in California induced thousands of Sonorans to desert their Apacheridden land. Not more than three hundred Mexicans were left in Arizona by 1856 and most of these were huddled in abject terror in the walled town of Tucson. But, within a few years, the population of Tubac and Tucson increased to 1,500 or 2,000, as some of the ex-soldiers returned to their milpas and the miners drifted back from the goldfields. For a brief time, the orchards of Tumacacori blossomed once again and the attractive fields and gardens which Father Kino had laid out near the missions came back into cultivation. With the commencement of the American Civil War, however, the government withdrew the troops which it had stationed along the border. Once again the Apaches made a bloody havoc of every ranch-house, village, and mining camp in the region of the Santa Cruz. From 1861 to 1871, the famous Apache chief, Cochise, ravaged the entire area from the Gila River far into Sonora and eastward to the Mimbres River in New Mexico. “Throughout Sonora,” wrote Sylvester Mowry, “the Apaches gradually extirpated every trace of civilization and roamed uninterrupted and unmolested, sole possessors of what was once a thriving and populous Spanish province.”1 J. Ross Browne, who visited both sides of the border in the seventies, reported that the mines had been abandoned; that the stock had been driven from the ranches; and that the Mexicans who remained in the province were apathetic, ridden by despair and a feeling of utter helplessness.

The Broken Border    59

While there had once been a few great estates and haciendas in southern Arizona, the territory was miserably poor and indescribably primitive in 1848. For over a hundred years, Arizona had been beset with calamities and misfortunes. It was the orphan, the pauper of the Spanish provinces. Too remote to participate in the trade which developed in New Mexico after the opening of the Santa Fe Trail, it was also too far removed from California to feel the leavening effects of the clipper ship trade. Its missions never attained the prosperity or stability of those in California; in short the settlement of the province was abortive. The Mexicans who were living in southern Arizona at the time of the American conquest were a miserable, landless, bewildered people, living in mortal fear of the Apaches. Twenty years after the conquest, Mowry estimated the Mexican population of Arizona at approximately two thousand, all of whom were pobres for the rich had long since vanished. The first Anglo-American settlers to arrive in Arizona were mostly from the states of the late Confederacy. Perhaps because of this circumstance, they lost little time in making Arizona “a white buffer state” between the Spanishspeaking people of New Mexico and those of Sonora. Granted separate territorial status in 1863, Arizona thereafter resisted a long series of proposals to admit it, along with New Mexico, as one state to the union. Since the colonization of Arizona had never been effective, the Spanish-speaking people of the state lacked foundations on which to build. Until the copper mines, the railroads, and the large reclamation projects began to attract Mexican immigrants, Indians remained a more important influence in Arizona than the Spanish-speaking.

2. The Tejanos While a few missions were established in eastern Texas in 1716, they were soon abandoned and the principal settlements remained those at San Antonio, a combination presidio-mission-and-pueblo; Goliad or La Bahia; and Nacogdoches. Exposed to Indian raids on all sides, none of these settlements prospered. The great rolling plains, stretching in all directions, made it impossible for the Spaniards to subdue the Comanches, who showed a marked disinclination to be enrolled as neophytes in the missions. Between 1722 and 1744, the Spanish spent three million pesos in an effort to colonize Texas but the number of colonists was less at the end than at the beginning of the period. By 1791 most of the Indians had fled from the missions and the few who remained were dispersed some years later. Poorly organized, feebly garrisoned, chronically neglected, the Texas settlements were swiftly engulfed in the tide of Anglo-American invasion. According to Bancroft, there were not more than 5,000 Mexicans in Texas in 1836: about 2,000 at San Antonio; 1,400 at La Bahia; and perhaps 500 at Nacogdoches. If one is to believe contemporary accounts, these Mexicans were a sorry lot:

60  North from Mexico

backward, illiterate, impoverished. The former Mexican soldiers, in particular, were a most bedraggled crew: “the lowest type of humanity that could be picked up,” according to Dr. George P. Garrison. Even the civilians were a source of considerable embarrassment to the Mexican officials who inspected the province. By the 1870s those who remained in the old settlements were completely submerged, surviving only as a “picturesque” element in the population: tamale vendors, chili vendors, peddlers of sweets (nueces dulces). The historians all agree, however, that quite a different situation prevailed along the border. Beginning in 1748, the rancheros of Santander (Tamaulipas) had been encouraged to settle along the Rio Grande in an effort to build a line of defense against the Indians. Most of these settlers came from such Mexican communities as Guerrero, Camargo, and Miero. Over a period of some years a few towns began to appear on the Texas side of the river: Dolores in 1761; Rio Grande City in 1757; Roma in 1767. Once Mexico had achieved its independence, the government parceled out most of the land lying between the Rio Grande and the Nueces in the form of large grants to favorites of the new regime and the movement of settlers into the region became more rapid. Due to the troubled state of affairs which prevailed during the decade of the Texas Republic, most of the Anglo-Americans settled north and east of the Nueces and this pattern prevailed for some years after the conquest. During this interregnum, however, Mexicans continued to cross the river and to settle between the Rio Grande and the Nueces. From 8,500 residents in 1850, the population of the border counties increased to 50,000 in 1880 and then to 100,000 in 1910. A large part of this increase was made up, of course, of immigrants from Mexico. The early Anglo-American settlers in the border counties were principally large cattle-operators (all of the land originally embraced in Mexican grants in Kleberg County became part of the famous King Ranch); and cattle-raising did not attract a large English-speaking population. As a consequence, sixty percent of the property-owners in Starr, Zapata, and Cameron counties, as late as 1930, were descendants of the original Mexican grantees. In the Lower Rio Grande Valley a way of life developed that was quite similar to that which had prevailed in early California. Here was to be found the same patriarchal setup in which a few large Mexican landowners lived an idle and lordly existence based on a system of peonage, vestiges of which still survive in the region. The peons were Mexican-Indian, being really more Indian than Mexican; while the landowners and vaqueros were mestizo. The peon was always in debt; in fact, he usually inherited the debts of his father. Landowners sold high-priced goods on credit to their peons, often refusing them permission to make purchases in the towns. Paid six reales a day (twelve cents in American money), the peon was not permitted to cultivate land, even to supply his own table needs; and his ownership of stock was limited to a few

The Broken Border    61

chickens, pigs, and goats. Throughout the lower valley, the peons lived in oneroom thatched-roofed dirt-floor jacales, with a portal or arbor made of dry cornstalks from which the inevitable olla and its gourd, used as a dipper, were suspended. Closely resembling the California pattern, Mexican ranch life in Texas survived well into the present century. Taking great pride in their Mexican culture, the landowners encouraged their families to keep the old customs and traditions alive and ruled their establishments like feudal lords. Their square, flat-roofed homes, usually made of stone, were often furnished with luxurious items purchased in Matamoros and Laredo. Here, as elsewhere in the borderlands, the kitchen was built around a huge fireplace and the oven was located in the yard. Since the border ricos consciously sought to retain their Mexican culture, the children were sent to private or parochial schools along the border or in Mexico. One of these schools, El Colegio Altamiro, founded in 1897, is still in existence. As more and more Mexicans came to the border counties, they were “Mexicanized,” not Americanized; for they had few contacts, at the outset, with the Anglo-Americans. Many of the border towns, such as Eagle Pass, developed as Spanish-speaking communities, with street signs and store names in Spanish. “The people are Texans,” wrote Lee C. Harby, “but do not speak English and have kept their blood, language, and manners.”2 The Tejanos, like the New Mexicans, took little part in the Mexican independence movement and had no tradition of self-government. Manhood suffrage was, of course, unknown. The lack of democratic traditions, the system of peonage, and the persistence of the patron-peon relation, combined to produce a type of political bossism similar to that which prevailed for so many years in New Mexico. The Anglo-American cattle-barons naturally assumed the prerogatives of the Mexican ricos and were accepted by the peones as patrons and protectors. Jim Wells, after whom one of the southern counties is named, is said to have bossed the border counties from 1880 until his death in 1920. Property-owners not only “voted their Mexicans,” but, when occasion demanded, brought droves of Mexicans from across the border, held them under guard in corrals and stockades, and voted them, too, on election day. Nor were all the bosses Anglo-Americans: Don Manuel Gerra was the undisputed political boss of Starr County until his death in 1915. As in California, many marriages took place, at an early date, between Anglo-American men and Mexican women in the border counties. Such names as Lacaze, Laborde, Lefargue, Decker, Marx, Bloch, Monroe, Nix, Stuard, and Ellert, according to Jovita Gonzales, represent families which claim the Spanish language as their own and boast of their Spanish blood. “The descendants of the Americans who married Mexican wives in the 1800’s,” she writes, “are more Mexicanized than the Mexicans.” Some of these marriages, it is interesting to note, were a by-product of the Mexican-American War. One of

62  North from Mexico

the early settlers in Rio Grande City, Henry Clay Davis, was a Kentuckian who returned to south Texas after the war to marry a Mexican-American. Retarded by a hundred years of border warfare, the economic development of the Lower Rio Grande Valley did not get under way until the completion of the St. Louis-Brownsville-Mexican rail line in 1904. Thus for more than a hundred years the Tejanos lived a life apart, cultivating their own customs and traditions. Even after 1848 they knew very little about what was going on in the United States and cared less. When they traveled, they went to Mexico. If they attended school, they were instructed in the Spanish language; if they read a newspaper, it was printed in Spanish. Like their compatriots in the mountain villages of New Mexico, they were hardly conscious of a change in citizenship. The moment the economic development of the region was undertaken, however, this feeling of indifference was soon dissipated. After the turn of the century, the economic development of the lower valley enabled the peons and jornaleros to profit by the change in rule. Never wholly escaping from a kind of peonage, they nevertheless managed to evade the more obvious forms of control. Those who got jobs in the citrus groves and on the truck farms began to purchase small lots in the towns and to acquire a measure of independence. The position of the landowners, on the other hand, began to deteriorate with the economic transformation of the region. The ricos were extremely annoyed to see Anglo-Americans come into the valley and, with the aid of farm machinery and the extension of irrigation systems, reclaim lands which they had long regarded as of little agricultural value. Nor was it long before the Mexican-American middle class, which had emerged in the valley, became intensely dissatisfied with the new dispensation. “The friendly feeling which had slowly developed between the old American and Mexican families,” writes Miss Gonzales, “has been replaced by a feeling of hate, distrust, and jealousy on the part of the Mexicans.” By the middle twenties, the word “white,” used to distinguish Anglo-Americans, affected the Tejanos “like a red flag to a bull.” South Texas was one of the first areas in the borderlands to develop a Mexican-American middle class. The retarded economic development of the region kept the Anglo-American element at a minimum prior to 1900, thereby giving the Tejanos a margin of time in which to develop a middle class of their own. The need for a middle class, moreover, was much greater in the border counties than in the remote mountain villages of New Mexico, where the possibilities of producing a surplus for trade were far less favorable. The emergence of a Mexican-American middle class in Texas has had an important effect on Anglo-Hispano relations. On August 24, 1927, these middle-class elements formed the League of Latin-American Citizens at a meeting in Harlingen. The formation of the “Lulacs,” as they are called, has been described as “the first attempt on the part

The Broken Border    63

of Mexican-Americans to organize themselves for the purpose of giving voice to their aspirations and needs as citizens of the United States.”3 From Harlingen the Lulac movement has spread to the other Texas towns but the nature of the “broken border” has minimized its influence in New Mexico and Arizona, California and Colorado. True to its origins, it has remained largely a middleclass organization.

3. The Californios The Spanish-speaking settlements in California differed in a number of respects from those in the rest of the borderlands. Unlike the other borderland settlements, California was both a sea and land frontier. The California missions, particularly those in the southern counties, were well administered (from the Spanish point of view) and became quite prosperous. When the first secularization decrees were issued in 1834, the lands and holdings of the missions were valued at $78,000,000. Mission San Gabriel, for example, operated seventeen large ranchos, worked 3,000 Indians, and owned 105,000 head of cattle, 20,000 horses, and 40,000 sheep. While the Spaniards had a great deal of trouble with the Indians of the Central Valley, they encountered little opposition from the coastal tribes. Furthermore, California was inherently a richer province than the other borderland settlements, with a milder climate and a great superiority in natural resources. For these and other reasons the character of the people of California, as Blackmar wrote, “differed from that of every other Spanish province. Owing to its isolated position, there was but little communication with the remainder of the Spanish dominion, and there sprang up an independent spirit not observed elsewhere in the Spanish Americas.”4 Revolutions were a matter of more or less normal occurrence in California. The population of California, however, was divided by the same sharp class and status lines. At the top of the hierarchy were the Spanish Franciscans, the Spanish officials, and the Spanish officers of the troops garrisoned in the province. Included in this category were some soldiers and noncommissioned officers who did not rate the distinction of being soldados distinguidos but were nonetheless a cut above the average in the borderlands. Some of the most prominent families in California—the Castros, Picos, Bandinis, Alvarados, Ortegas, and Noriegas—belonged in this category. While there were some distinguished families in the province, the number having “pure Spanish blood” has been grossly exaggerated. “A very small percentage,” wrote Charles Dwight Willard, “were pure-blooded Spaniards, although few were ready to admit they were anything else.” Most of these gente de razón families, like the ricos of New Mexico, were related by marriage and constituted a kind of ruling-class elite. After the opening of the clipper trade, a few of the homes in the presidial towns had an atmosphere reminiscent of the social graces, refinement, and

64  North from Mexico

elegance of the Old World. But most of the gente de razón lived the same rough, semiprimitive existence as their counterparts in New Mexico. Always outnumbered by the lower classes in the ratio of ten to one, the number of gente de razón families was never large. Included in the first generation were some welleducated, capable, energetic individuals; but they were certainly a minority element. Accustomed to indolence and denied an opportunity for education, the second generation was demoralized by the ubiquitous use of cheap Indian labor. The inefficiency of this element was unmatched in Spanish America. Below the gente de razón were the Mexicans: soldiers, artisans, colonists, the cholos of the province. Recruited from among the riffraff of Sonora and Sinaloa, they were certainly a nondescript lot. “Presidial society looked down upon these rustic villagers,” writes Dr. John W. Caughey, “and the missionaries regarded them askance, as being likely to corrupt the neophytes.”  * In the towns and on the ranches, the Mexicans were sharply set apart from the gente de razón. Largely illiterate, speaking a different dialect, they thought of themselves as Mexicans, not as Spaniards. Marriage between the Mexicans and the gente de razón elements was, of course, unthinkable. Seldom, if ever, did the Mexicans rise to positions of prominence. Unlike New Mexico, there were no small farming villages in California so that the state never produced a class of independent, self-sufficient paisanos. In many respects, the social structure of Spanish California resembled that of the Deep South: the gente de razón were the plantation-owners; the Indians were the slaves; and the Mexicans were the California equivalent of “poor white trash.” These sharply differentiated groups reflected a division of labor which had become traditional. The Mexicans were the artisans, vaqueros, and majordomos of the ranchos; the craftsmen and pobladores of the pueblos. The gente de razón held all the government positions, made up the officer class of the military, and controlled the great ranchos. While showing a lively interest in cattle and horses (the care of which they feared to entrust to the Indians), they were never interested in farming so that the agriculture of the province remained largely undeveloped and primitive. At the base of the pyramid were the Indians, upon whose unpaid labor the entire economy was based. To these elements must be added, however, a unique strain made up of the American, British, Scottish, German, and French adventurers who had infiltrated the province prior to 1846. There were only a hundred or so of these adventurers but they played a role of crucial importance at the time of the conquest. With scarcely a single exception, these curiously assorted characters had married daughters of the gente de razón after first joining the Catholic Church and accepting Mexican citizenship. Once related to the “best families”

* From California by John Walton Caughey. Copyright, 1940, by Prentice-Hall, Inc.

The Broken Border    65

by marriage, they became eligible for land grants and were permitted to engage in trade. Embracing the daughters of the land, they also made a pretense of embracing its customs, adopting the prevailing style of dress and Hispanizing their surnames. At the time of the conquest, of course, they went over to the American side en masse and, in many cases, induced their in-laws to collaborate with those who were directing the American invasion. No matter what his social status was–doctor, trader, sailor, or smuggler— the ultramontane adventurer had little difficulty in joining the gente de razón, provided he joined the Catholic Church and became a citizen of Mexico. As long as California remained isolated from the rest of the country, these interlopers were quite willing to pose as hijos del país; but with the influx of AngloAmerican women they quickly discarded their Spanish trappings, and often their Spanish wives, and reverted to type. Conversely, those settlers—and there were a few—who married Indian women never achieved the status of gente de razón. To the same point, an early-day resident of Los Angeles once remarked that he had never known “of a Spaniard or Mexican of this section marrying an American wife.” With the breakup of the ranchos, the intermarriage and limited cultural fusion of this earlier period came to an abrupt cessation. Essentially the difference between California and New Mexico, as Spanish provinces, was that the former was the home of the cattle ranch, the latter of the sheep ranch. In New Mexico, the pastores were permitted to have their own subsistence farms and to graze a few sheep along with those of their patron; but all economic activity in California was centralized, first in the missions, and later in the rancho establishments. The opening of the clipper trade with Boston gave, of course, a great impetus to cattle-raising. The more money there was to be made in cattle, the greater became the clamor to secularize the missions. Once the missions were secularized, the number of range cattle soared to new heights. Between 1830 and 1846, the period of secularization, eight million acres of land in California passed into the ownership of less than eight hundred grantees. Many of these ranchos were baronial in extent, with cattle grazing “on a thousand hills.” With the discovery of gold, the price of cattle promptly soared from $2 and $4 a head to $20 and $50 and the cattle ranches and vineyards of Southern California became immensely profitable. Of short duration, this soaring prosperity had disastrous consequences for the ranchero class. The moment herds of cattle began to be driven overland to the mines, the price of California cattle quickly dropped. Soon great herds of beef cattle were being raised and pastured in the San Joaquin Valley and few cattle buyers bothered to make the long trip south of the Tehachapi to purchase scrawny Spanish steers from the rancheros. During two years of ruinous drought, in 1862 and 1864, nearly three million cattle perished in the “cow counties” of Southern California and nearly five-sixths of the land was reported tax-delinquent. Forty percent of the land held in Mexican grants was sold to meet the costs and

66  North from Mexico

expenses involved in confirming land titles after the conquest. The Rancho de los Alamitos, consisting of 265,000 acres, was sold for delinquent taxes of $152–one of many similar cases. Interest rates of five percent compounded monthly were not uncommon. The Rancho Santa Gertrudes, worth a million dollars, was forfeited for nonpayment of a $5,000 debt. So general was the debacle that by 1891 not more than thirty gente de razón families in the northern part of the state had managed to preserve even a semblance of their former prestige and power. In Southern California, however, the gente de razón retained a measure of their former power and influence for some years after the conquest. Here they were concentrated in sufficient number so that they remained an important political factor through the 1880s. In most elections, from 1849 to 1880, the newcomers were pitted against the Spanish-speaking. “Down to the end of the 1870’s,” writes Owen O’Neil, “local politics in Southern California were complicated by a natural tendency to diverge on racial lines. Vast and complex family connections would make it impossible to trace these cleavages by any process so simple as noting Spanish names, but they were a real and potent factor which became more evident after 1865, when so many of the old Californians, once magnates of the land, were being crowded to the wall by economic misfortune.” Among the first representatives of Santa Barbara County in the state legislature were such individuals as Pablo de la Guerra, Antonio María de la Guerra, Romauldo Pacheco (later lieutenant-governor), and J. Y. Cota. An Estudillo and a Coronel became state treasurers and, in Los Angeles, a member of the Sepulveda family was elected to the bench. As late as 1870, native Californians outnumbered Anglo-Americans in Santa Barbara, owned more than a third of the property, and occupied most of the political positions; but, by the end of the decade, the native element was almost entirely eclipsed. Unlike New Mexico, California was engulfed by a tidal wave of AngloAmerican immigration after 1848. While the northern counties received the bulk of this immigration at the outset, the tide shifted to Southern California in the 1880s. “This overwhelming horde of new arrivals,” wrote Willard, “took possession of the land and proceeded to make things over to their own taste.” The Spanish-Mexican appearance of the Southern California towns changed overnight. As much as anything else, this transition was symbolized by the rapid disappearance of the adobes. “Death and emigration,” wrote J. P. Widney in 1886, “are removing them [the Californians] from the land. . . . They no longer have unnumbered horses to ride and vast herds of sheep, from which one for a meal would never be missed. Their broad acres now, with few exceptions, belong to the acquisitive American. . . . Grinding poverty has bred recklessness and moroseness.” If this process of change bore heavily upon the gente de razón, it had a simply crushing effect upon the Mexicans. One after another the economic functions for which they had been trained were taken from them. The Mexicans

The Broken Border    67

were excellent and well-trained vaqueros but this function disappeared with the collapse of the ranchero regime. The rapid rise of the sheep industry after 1860 momentarily provided employment as herders and shearers; but the period of bonanza sheep-raising soon came to an end. The Mexican then reappears in the local annals as a farmworker and livery-stable hand. Long before the livery stables disappeared, however, the Chinese began to displace the Mexicans as farmworkers. Visiting Southern California in 1888, Edward Robert noted that the “houses of the Spanish-speaking people are being taken over by the Chinese, who have invaded the adobe cottages.” Anglo-Americans infiltrated New Mexico; they engulfed California. The difference in impact was also a function of the size of the Spanish-speaking element in the two states: 60,000 in New Mexico, 7,500 in California. In California, moreover, there was no buffer group to stand between the Spanish-speaking and the Anglo-Americans in the manner that ten thousand well-settled Pueblo Indians stood between Anglos and Hispanos in New Mexico. With the eclipse of the Spanish-speaking element after 1880, few visible evidences of Spanish culture could be noted in California. Some Spanish words had been incorporated into the speech and important elements of Spanish-Mexican jurisprudence had been woven into the legal fabric of the state. A considerable amount of Spanish-Mexican blood flowed in the veins of local residents with such names as Travis, Kraemer, Reeves, Locke, and Rowlands. Most of the Spanish street names had been Anglicized, although few of the place-names were changed. At the turn of the century it appeared—in fact it was generally assumed—that the Mexican influence had been thoroughly exorcized. But what had really happened was that the “old life”—the Mexican life—of the province had retreated “along the coastal plains that reach from Los Angeles to Acapulco.” Just as the Spanish-speaking had retreated from the northern counties to the southern, so they later withdrew, to some extent, to Mexico. But the number of Spanish-speaking residents in Southern California was at all times sufficient to keep vestiges of the earlier life and culture alive. Later, in the period from 1900 to 1920, these surviving elements of the old life were renewed and revived by a great influx of Mexican immigrants and the longdormant conflict of cultures entered upon a new phase.

4. Lost Provinces In two parts of the borderlands, southern Colorado and western Texas, the Spanish settlements which existed were even more severely isolated than those in New Mexico. These were the real “lost provinces.” Although both De Vargas and De Anza had explored southern Colorado, the first permanent colony was not established in the Costilla Valley until 1849. The town of San Luis, founded in 1851, is said to be the oldest settlement in the state. Most of the Colorado

68  North from Mexico

colonists came from New Mexico, many of them from Taos, which was, in a sense, the parent colony. The Colorado settlements had been planned as a series of listening posts by means of which New Mexico might be forewarned of Indian raids. To induce the colonists to settle in these lonely outposts, the provincial government had made small land grants to individuals and to families. The first settlers in southern Colorado lived in crudely built jacales which were later replaced with adobe structures. The form of settlement, both in Colorado and west Texas, was the plaza: a series of flat-roofed adobe houses joined together in the form of a square or rectangle with an opening at each end or on the sides. In effect, the plaza was a form of walled village inhabited by a number of families. The enclosed patio frequently served as a corral or stockade into which the stock were driven at night. The thick exterior walls, without doors or windows, had an extension around the roof, called the pretil, which was used as a barricade. Since no glass was available, the windows on the patio side were covered with a parchment or pergamino, made from sheepskin. Doors were ponderous affairs without hinges or locks and the floors were of beaten earth. The well-to-do lived in single-family dwellings, called plazuelas, located at some distance from the plaza. Farming and stock-raising activities were limited in Colorado by the incessant danger of Indian raids. As late as 1854 there were only two guns to be found in Costilla, one an old musket; and for years the principal weapons were bows and arrows. Everything the colonists possessed was homemade: their brooms and utensils; their household furnishings; their clothing; their agricultural implements. Long after the Indian menace had been eliminated, the thick walls of the patio served to enclose the lives of the people; to set them apart in a world by themselves. In much the same way, the self-sufficient character of their economy made it possible for them to carry over into the present many elements of their Spanish culture. Walled off by mountain ranges, the San Luis Valley has always had stronger economic and cultural ties with New Mexico than with the rest of Colorado. Never having attracted a large number of Anglo-American settlers, it remains predominantly Spanish-speaking (about fifty-eight thousand Spanishspeaking people live in southern and southeastern Colorado today). If possible more isolated and landlocked than the mountain villages of New Mexico, a folk culture has survived in the San Luis Valley, some phases of which stem directly from the Middle Ages. Olibama López, a native of the region, has given the following vivid description of what happened, a few years ago, when one of the communities reenacted the entire story of the capture, trial, and crucifixion of Christ: On Wednesday evening a group representing soldiers and Jews led by the armored centurion, riding a spirited horse and flanked by two lacayos or lackeys, to the accompaniment of martial music produced by a fife and

The Broken Border    69

drum, went forth to capture “Christ.” The latter was represented by the Image of Christ that belonged to the church, and the priest acted as the voice. The capture occurred in the cementerio of the church, a large patio where the dead were buried during the first years of settlement. The Image was carried in a procession; then it was brought back to the church. On Thursday morning the story of the capture and trial was read at church. On Friday afternoon, after a procession through the village, in which several men dragged an enormous cross, the Image was crucified in church. A group of piteros—fifers—played a dirge, a plaintive melody that penetrated the very hearts of the listeners, and made the scene they had just witnessed very real to them. On Friday night, after the last mass, occurred the tinieblas, representing the darkness that fell over the earth as Christ expired. . . . The church was left in darkness except for several candles burning behind a heavy curtain where the rezadores—men who prayed—were stationed. Before the prayers began, one of the rezadores would say in a loud voice, salgan, vivos y difuntes, que aquí estamos todos juntos—come forth ye dead and living for we are all here together. No sooner had he said this, than chains were dragged across the floor, the matracas were rattled, and the pitas were blown behind the curtain. This represented the earthquake and the opening of the graves which occurred as Christ died. . . . After the tinieblas, the Christ was taken from the cross and laid in a coffin. A bachillero—a talkative fellow—was left to guard it all night, though the soldiers returned every hour to see that the body was still there.

Here in the Spanish-speaking towns and villages the children still play dozens of Spanish games: la pelota, las cazulejas, pitarilla, el canute, el coyotito, la ponsona, la cabra. In former times, the villagers were visited once a year by traveling troupes of maromeros or tumblers from Taos, each with its payaso or clown. Old folk songs, such as “La Cautiva Marcelina,” and “El Vaquero Nicolás,” are still popular. Each village has its poeta, or poet, adept at composing coplas for all occasions. While Mexico and Spain are “foreign” countries to most of the residents, they are addicted to a kind of españolismo: a complacent self-satisfaction with everything Spanish. While the San Luis Valley is made up principally of small agricultural holdings, it has had, in times past, one or two princely estates. Casimiro Barela, after whom the town of Barela is named, served Las Animas County in the state legislature for forty years (1876 to 1916). His home at Rivera, near Barela, was patterned after the large Mexican hacienda and maintained by a retinue of servants. The owner of important properties in Mexico and a coffee plantation in Brazil, Barela was the “boss” of the region during most of his lifetime. In the southern part of the state, also, is the famous Trinchera Ranch, once part of the 1,038,000-acre Sangre de Cristo grant.

70  North from Mexico

For years New Mexican ciboleros or buffalo-hunters had roamed the famous El Llano Estacado, or Staked Plains of the Panhandle, hunting buffalo with the lance. Crews in huge cumbersome wooden-wheeled carretas followed the ciboleros to skin the animals and strip the meat, hanging it out to dry in the sun. This dried meat the Mexicans called charquí which, on the tongue of the Anglo-Americans, became “jerky.” The routes of the ciboleros were marked by buffalo skulls; hence the name “Staked Plains.” Moving down the Canadian from Las Vegas to a point near the present town of Canyon, the ciboleros had various points of rendezvous along the river. Out of this buffalo-hunting came the trade with the Comanches, with the comanchero gradually replacing the cibolero. To the plains, the comanchero brought merchandise which was traded to the Comanches at the old rendezvous points along the river: Las Tecovas, Las Linguas (“The Tongues,” so named because many languages were spoken there), and other points. The comanchero trade was long a source of great friction between Anglos and Hispanos in the Southwest, for the Anglos charged, and with some truth, that the stock which the Comanches offered in exchange for New Mexico merchandise was contraband stolen from Texas ranches. Eventually the Anglo-American buffalo-hunters drove the Mexican ciboleros and comancheros from the plains but as late as 1875 one José Taffola was found, with a full caravan of merchandise, roaming the plains in search of Comanches who had failed to keep a rendezvous. At an early date, perhaps as early as 1830, a few New Mexico families, principally from Taos, began to settle in plazas along the Canadian River. The famous early-day town of Tascosa, from atascosa (“boggy”), was an outgrowth of such a settlement. In 1876 Casimero Romero moved into the region with a huge caravan of fourteen prairie schooners to be followed, the next year, by a number of other families from New Mexico. Down the river the plazas bore the names of the various families: Trujillo, Valdez, Ortega, Chavez, Romero, Sandoval, Domingo, Callinas, Joaquin, Ventura, Montoya. For over fifty years, these Mexican families were about the only settlers in the region, living in their low, flat-roofed stone plazas, grazing their sheep on the plains. Later, when large Anglo-American cattle interests invaded the Panhandle, the Mexicans retreated to New Mexico. “Two of the large ranch outfits,” writes the historian of Tascosa, “set about moving the Mexican plaza residents as soon as they had completed their fences.”5 The last survivor of the old Mexican life in the Panhandle, one Sandoval by name, died near the turn of the century.

6

“Not Counting Mexicans”

When asked how many notches he had on his gun, King Fisher, the famous Texas gunman, once replied: “Thirty-seven—not counting Mexicans.” This casual phrase, with its drawling understatement, epitomizes a large chapter in Anglo-Hispano relations in the Southwest. People fail to count the nonessential, the things and persons that exist only on sufferance; whose life tenure is easily revocable. The notion that Mexicans are interlopers who are never to be counted in any reckoning dies but slowly in the Southwest. To this day Mexicans do not figure in the social calculations of those who rule the Border States. As I write these lines, the Mexican consul-general in Los Angeles has just entered a vigorous protest against the insulting behavior of custom inspectors at the municipal airport. A majority of the present-day residents of the Southwest are not familiar with the malignant conflict of cultures which has raged in the borderlands for more than a century. Blinded by cultural myths, they have failed to correlate the major events in a pattern of conflict which has prevailed from Brownsville to Los Angeles since 1846. Once this correlation is made, it becomes quite apparent that the Mexican-American War was merely an incident in a conflict which arose some years before and survived long after the Treaty of Guadalupe Hidalgo. It is only within the framework of this age-old conflict that it is possible to understand the pattern of Anglo-Hispano cultural relations in the Southwest today. In summarizing the history of this conflict, one necessarily starts with Texas, for there the first blood was shed.

1. Los Diablos Tejanos In Texas the Spanish-Mexican settlements were directly in the path of AngloAmerican expansion. Unlike the rest of the borderlands, Texas was not separated from the centers of Anglo-American population by mountain ranges and desert wastes; geographically it invited invasion. In a series of belts or strips, its rich, alluvial plains stretched from the plateaus to the gulf. The rivers that

72  North from Mexico

marked these belts could be crossed, at all seasons, at almost any point, without much trouble. On the other hand, between the most southerly settlements in Texas and those in Mexico, there was, as Dr. Samuel Harman Lowrie has pointed out, “a great expanse of semi-arid land which at that time served as a more or less natural, though temporary barrier to the effective extension of Mexican influence and control.” Texas was 1,200 miles removed from its capital, Mexico City. By 1834 the Anglo-Americans outnumbered the Mexicans in Texas: thirty thousand to five thousand. Most of the Mexicans were concentrated in the old Spanish towns or along the border, while the Anglo-Americans were to be found on the farms and ranches. Mexican townspeople had few opportunities for acculturation for they saw very little of the Anglo-Americans. From the outset, moreover, relations between the two peoples were clouded by the fear of war. The Anglo-Americans bore the brunt of Mexico’s hostile distrust of the United States and were, in turn, encouraged to take an unfriendly attitude toward the natives by the unconcealed, aggressive designs of the jingoes in Washington. As might have been expected, each group formed a highly unfavorable initial impression of the other. To the early American settlers, the Mexicans were lazy, shiftless, jealous, cowardly, bigoted, superstitious, backward, and immoral. To the Mexicans, on the other hand, the Texans were “los diablos Tejanos”: arrogant, overbearing, aggressive, conniving, rude, unreliable, and dishonest. The first Mexican ambassador to the United States had complained in 1882 of the “haughtiness of these republicans who will not allow themselves to look upon us as equals but merely as inferiors.” Still another Mexican official had charged that the Americans in Texas considered themselves “superior to the rest of mankind, and look upon their republic as the only establishment upon earth founded upon a grand and solid basis.” Full of brag, bluster, and spread-eagle chauvinism, the Americans of the 1800s were hardly the most tactful ambassadors of goodwill. The truth of the matter is that the border residents were not a credit to either group. Under the most favorable circumstances, a reconciliation of the two cultures would have been difficult. The language barrier was, of course, a constant source of misunderstanding; neither group could communicate, for all practical purposes, with the other. The Mexicans knew almost nothing of local self-government, while the Americans, it was said, traveled with “their political constitutions in their pockets” and were forever “demanding their rights.” Although tolerant of peonage, the Mexicans were strongly opposed to slavery. The Anglo-Americans, most of whom were from the Southern states, were vigorously proslavery. The Anglo-Americans were Protestants; the Mexicans were Catholic. Speaking of a Mexican, a Protestant missionary is said to have remarked: “He was a Catholic, but clean and honest.” Both groups lacked

“Not Counting Mexicans”   73

familiarity with the existing Mexican laws, for there was no settled government in Texas. Anglo-Americans found it extremely difficult to respect the laws of Mexico in the absence of law-interpreting and law-enforcing agencies. Thus it was, as Dr. Lowrie writes, that “cultural differences gave rise to misconceptions and misunderstandings, misunderstandings to distrust, distrust to antagonism, and antagonism on a very considerable number of points made open conflict inevitable.” The first Anglo-Americans literally fought their way into Texas. While most of these early filibustering expeditions were defeated, they succeeded in laying waste to the country east and north of San Antonio. Both Mexicans and Americans were killed by these invading private armies. No sooner had the Mexicans driven out the filibusters, than the Comanches raided the entire stretch of country between the Nueces and the Rio Grande. According to one observer, the whole region was “depopulated, great numbers of stock were driven off, and the people took refuge in the towns on the Rio Grande.” Preoccupied with revolutionary events in Spain and Mexico, the government could give little attention to the Texas settlements. After 1821, however, a measure of protection was provided against the devastating raids of the Comanches and many of the settlers moved back across the Rio Grande.

2. Alas! The Alamo With the Texas Revolution came the embittering memories, for the Texans, of the slaughter of Anglo-Americans at the Alamo and Goliad; and, for the Mexicans, of the humiliating rout and massacre at San Jacinto. Prior bitternesses were now intensified a thousandfold. “Towards the Mexicans remaining within the limits of the Republic,” writes Dr. Garrison, “the feeling of the Texans was scarcely better than towards the Indians.” Memories dating from this period still poison relationships between Anglos and Hispanos in Texas. Some years ago a district judge told of how, as a child, he had heard an old man give an eyewitness account of the slaughter at the Alamo. “I never see a Mexican,” he confessed,“without thinking of that.” José Vasconcellos, the well-known Mexican educator and philosopher, tells in his autobiography of how these same memories poisoned his boyhood in Eagle Pass. After the Texas Revolution, as Erna Fergusson has pointed out, “Texans could not get it out of their heads that their manifest destiny was to kill Mexicans and take over Mexico.” Throughout the decade of the Texas Republic (1836–1846), the shooting war continued in “the Spanish country” south of the Nueces. Murder was matched by murder; raids by Texans were countered by raids from Mexico. Since a peace treaty was never negotiated, no boundaries could be fixed. Texas claimed to the Rio Grande, while Mexico insisted that its boundary rested on the Nueces. In the bloody zone between the two rivers an uninterrupted guerrilla warfare

74  North from Mexico

continued throughout the life of the Texas Republic. In 1839 General Don Antonio Canales launched a revolution on Texas soil against Santa Anna and raised the banner of the Republic of Rio Grande. Of the 600 men who rallied to his standard, 180 were Texans. Awakening to the fact that Texans were using his insurrection as a cover for an attack on Mexico, General Canales finally surrendered but not until his troops had fought several engagements along the border. At the head of a raiding party of five hundred men, General Vásquez captured San Antonio in 1842 and held it for two days. These are but two of many similar episodes that occurred during the hectic life of the new republic. Throughout the period of this border warfare, the Texas-Mexicans were caught between opposing forces. “When the Americans have gone there,” explained a delegate at the Texas constitutional convention, “they have preyed upon the Mexicans; they have been necessarily compelled by force or otherwise to give up such property as they had. So vice versa, when the Mexicans have come in, they have been necessarily compelled to furnish them the means of support. . . . Since 1837 they [the Texas-Mexicans] have been preyed upon by their own countrymen as well as by ours.” The Texans constantly suspected the Mexicans of inciting the Indians against them and every Indian raid provoked retaliation against the Tejanos. The Mexicans naturally regarded the Texas Revolution as American-inspired and the prelude to the conquest of Mexico. However all Mexicans were not equally affected by this complex warfare. A sizable number of the upper-class settlers quickly became identified with the Texans. These Texanized Mexicans or “the good Mexicans” were called Tejanos and were invariably of the rico class. Two of the fifty signers of the Texas Declaration of Independence were native Mexicans and a third, born in Mexico, became the first vice-president of the republic. At a later date, Captain Refugio Benavides commanded a company of Texas-Mexicans which operated along the border against Mexican raiders and marauders.

3. The Mexican-American War Provoked by the annexation of Texas in 1846, the Mexican-American War represented the culmination of three decades of cultural conflict in Texas. To the Mexicans, every incident in Texas from the filibustering raids to the Revolution of 1836 was regarded, in retrospect, as part of a deliberately planned scheme of conquest. To the Anglo-Americans, the war was “inevitable” having been provoked, in their eyes, by the stupidity and backwardness of the Mexican officials. Not only did Mexico forfeit an empire to the United States, but, ironically, none of the signers of the Treaty of Guadalupe Hidalgo realized that, nine days before the treaty was signed, gold had been discovered in California. That they had unknowingly ceded to the United States territories unbelievably rich in gold and silver—the hope of finding which had lured Coronado and De Oñate

“Not Counting Mexicans”   75

into the Southwest—must have added to the Mexicans’ sense of bitterness and defeat. Furthermore the way in which the United States fought the MexicanAmerican War added greatly to the heritage of hatred. A large part of our invading army was made up of volunteers who, by all accounts, were a disgrace to the American flag. General Winfield Scott readily admitted that they had “committed atrocities to make Heaven weep and every American of Christian morals blush for his country. Murder, robbery and rape of mothers and daughters in the presence of tied-up males of the families have been common all along the Rio Grande.” Lieutenant George C. Meade, of later Civil War fame, said that the volunteers were “driving husbands out of houses and raping their wives. . . . They will fight as gallantly as any men, but they are a set of Goths and Vandals without discipline, making us a terror to innocent people.” How bitterly these outrages were resented is shown by a passage which Lloyd Lewis has culled from one of the Mexican newspapers of the period: “the horde of banditti, of drunkards, of fornicators . . . vandals vomited from hell, monsters who bid defiance to the laws of nature . . . shameless, daring, ignorant, ragged, bad-smelling, long-bearded men with hats turned up at the brim, thirsty with the desire to appropriate our riches and our beautiful damsels.” The year 1844 had seen the rise of a Native American Party in the states and much anti-Catholic feeling found expression during the war. Mexicans charged that the volunteers had desecrated their churches, “sleeping in the niches devoted to the sacred dead . . . drinking out of holy vessels.” Two hundred and fifty American troops, mostly of Catholic background, deserted and joined the Mexican army to form the San Patricio battalion. The barbarous manner in which eighty of these deserters were executed in San Angel, a suburb of Mexico City, was long cited by the Mexicans as further proof of Yankee cruelty. Nothing was more galling to the Mexican officials who negotiated the treaty than the fact that they were compelled to assign, as it were, a large number of their countrymen to the Yankees. With great bitterness they protested that it was “not permissible to sell, as a flock of sheep, those deserving Mexicans.” For many years after 1846, the Spanish-Americans left in the United States were known in Mexico as “our brothers who were sold.” As late as 1943 maps were still used in Mexican schools which designated the old Spanish borderlands as “territory temporarily in the hands of the United States.” It is to the great credit of the Mexican negotiators that the treaty contained the most explicit guarantees to protect the rights of these people, provisions for which they were more deeply concerned than they were over boundaries or indemnities. It should never be forgotten that, with the exception of the Indians, Mexicans are the only minority in the United States who were annexed by conquest; the only

76  North from Mexico

minority, Indians again excepted, whose rights were specifically safeguarded by treaty provision. Just as the end of the Texas Revolution did not terminate hostilities in Texas, so the Treaty of Guadalupe Hidalgo failed to bring peace to the borderlands. Under the terms of the treaty, it became the obligation of the United States to police 180,000 Indians living in the territories which we acquired from Mexico. This obligation the United States failed to discharge for many years. Taking advantage of the confusion which prevailed, the Indians launched fierce raids on both Anglo and Hispano settlements, conducted marauding expeditions deep in Mexican territory, and cunningly exploited the hatred that had been engendered between Anglo and Hispano. The Anglos promptly attributed these raids to Mexican duplicity and instigation; the Hispanos as promptly charged them up to the malice or carelessness of the Americans. Hard-pressed on all sides, the Indians had come to live off the plunder seized in these raids which, with the confusion and demoralization which prevailed in Mexico, were conducted on a larger scale than ever before. It was not until about 1880 that the United States finally managed to bring the Indians of the Southwest under close police surveillance. Nor were Indians the only troublemakers in the postwar decades. Between 1848 and 1853, various American filibustering expeditions violated Mexican territory in Sonora, Lower California, and at various points along the border. When word of the discovery of gold reached the Eastern states, swarms of emigrant gold-seekers passed along the southern routes to California, often traveling in Mexican territory without passports, and not infrequently helping themselves to Mexican food and livestock en route. In 1850 José M. Carvajal organized a revolution in Mexico, sponsored by American merchants, which aimed at converting the State of Tamaulipas into the Sierra Madre Republic. Carvajal was a Texan by birth who had been educated in Kentucky and Virginia. Backed by Richard King and Mifflin Kennedy, two of the great cattle-barons of south Texas, the Carvajal revolution was supported by bands of armed Texans who crossed the Rio Grande. The American ambassador reported that these raids, in which as many as five hundred Texans participated, had “awakened a feeling of intense prejudice against everything connected with American interest.” The fateful strip of territory between the Nueces and the Rio Grande once again became the home of numerous outlaw bands who preyed indiscriminately upon both Mexican and American settlers. In the face of these staggering blows—filibustering expeditions, Indian raids, revolution, war, and constant guerrilla fighting—the Mexicans in Texas constantly retreated and their retreat, of course, gave rise to the notion that their conquerors were pursuing a mandate of destiny. Major Emery, writing in 1859, said that the “white race” was “exterminating or crushing out the inferior race”; and an American soldier

“Not Counting Mexicans”   77

wrote home that “the Mexican, like the poor Indian, is doomed to retire before the more enterprising Anglo-Americans.”

4. Slaves and Peons As early as 1839 fairly large numbers of Negro slaves had escaped from their Texas owners by crossing the Rio Grande and a sizable colony of ex-slaves had sprung up in Matamoros. During the Civil War, the Texans suspected that native Mexicans were implicated in the flight of fugitive slaves, an accusation that found circumstantial confirmation in the known opposition of Mexicans to slavery (Mexico had sought to insert a provision in the treaty barring slavery forever from the territory ceded to the United States). “The possession of slaves in Western Texas,” wrote Colonel Ford, “was rendered insecure owing to the contiguity of Mexico, and to the efforts of the Mexicans to induce them to run away. They assisted them in every way they could.” To some extent, the movement of Negro slaves across the border was matched by the flight of Mexican peons into Texas. According to Dr. Paul S. Taylor, some 2,812 servants with families numbering an additional 2,572 persons, escaped to Texas from Nuevo Leon and Coahuila in the period from 1848 to 1873. The loss in unpaid debts, which the flight of these peons represented, was estimated by the Mexican government to be in excess of $400,000. In 1856 a Negro insurrectionary plot was uncovered in Colorado County. According to the Texans, the Negroes had planned to rebel, kill their masters, and, with the aid of native Mexicans, fight their way across the border. Without exception every Mexican in the county was “implicated” and over two hundred slaves were arrested and punished (two were whipped to death). Mexicans were ordered to leave Matagorda and Colorado counties immediately and in Uvalde they were forbidden to travel the roads without passes. “Anti-Mexican sentiment,” writes Dr. Taylor, “based on the belief that the peons imperilled the institution of slavery, broke out in meetings which in Austin, Gonzales, and other towns, passed resolutions protesting against their employment. At Goliad the resolution declared that ‘the continuance of the greaser or peon Mexicans as citizens among us is an intolerable nuisance and a grievance which calls loudly for redress.’” As always, the circumstance that Mexicans were concentrated in the strip of territory immediately north of the border aroused the most dire forebodings. The Negro insurrection was quickly followed by the Cart War which broke out in 1857. Prior to this time, Mexican ox-cart freighters had been hauling— between San Antonio and the coast and from San Antonio to Chihuahua— an annual cargo of goods and merchandise valued at several million dollars. The Cart War involved a systematic campaign on the part of Anglo-Americans

78  North from Mexico

to force Mexican freighters out of this lucrative business. For over a year, organized bands of Texans preyed on the Mexican freight trains, killing the drivers, stealing the merchandise, and generally disrupting the traffic. So tense did the situation become, with the Mexican ambassador filing one vigorous protest after the other, that federal troops were finally dispatched to protect the cartmen.

5. “Red Robber of the Rio Grande” In the wake of the Cart War came the highly significant Cortina episode. Juan Nepomuceno Cortina—”the red robber of the Rio Grande”—was born near Brownsville. A blocky, powerfully built, red-bearded Mexican, Cortina came from a prominent and well-to-do family. Like so many Mexicans in Texas, he was a magnificent horseman. The Cortina War, which was to last a decade, started on July 13, 1859, when a deputy sheriff arrested a Mexican who had been a servant of the Cortina family. Contending that the arrest was merely another example of gringo arrogance, Cortina shot the deputy and freed the prisoner. On the morning of September twenty-eighth, Brownsville awoke to the cry of “Viva Cortina! Viva Mexico! Maten los Gringos!” as Cortina, at the head of an armed force, swept into the town, killed five Americans, released the pelado culprits from the jail, and plundered stores and shops. By 1860 Cortina had laid waste to the country from Brownsville to Rio Grande City—a distance of a hundred and fifty miles—and inland as far as Arroyo Colorado. Fifteen Americans and eighty “friendly” Mexicans were killed in these raids, while Cortina is said to have lost a hundred or more of his men. For fifteen years, Cortina was the scourge of the Lower Rio Grande Valley, defying capture, constantly eluding his pursuers. At one point in the Cortina War, Captain McNelly of the Texas Rangers crossed the Rio Grande, in defiance of orders, and gave Cortina’s forces a severe defeat in a pitched battle at Las Cuevas. Incensed by these continued raids, the Texans burned the homes of all Mexicans suspected of being implicated or of giving aid and comfort to Cortina’s forces. On his part, Cortina terrorized the Mexican residents and made short shrift of those suspected of being informers. This continued terror naturally silenced the Mexicans—a circumstance which only confirmed the Texans’ belief in their innate duplicity and treacherousness. A real expert in border warfare, Cortina hoisted the Mexican flag in Texas, and, so it was said, often raised the American flag in Mexico. Both Texas Rangers and Mexican troops from Matamoros on more than one occasion met defeat at his hands. Although he was a bandit and a cattle-thief, there was unquestionably something of the Robin Hood about Cortina. He had become a desperado, so he said, because the Anglo-Americans had tried “to blacken, depreciate, and load with insults” the Mexican residents of Texas. In one of

“Not Counting Mexicans”   79

numerous manifestos, he pointed out that “a multitude of lawyers” in Texas sought to rob the Mexicans of their lands. In particular, he charged that one Adolph Glavecke, a deputy sheriff, acting in collusion with certain lawyers, had spread terror among the Mexicans, threatening to hang them and to burn their homes unless they abandoned the country. “Our personal enemies,” he said, “shall not possess our lands until they have fattened it with their gore.” Major Heintzelman, on the border at the time, stated that after the Brownsville raid Cortina was a great hero in the eyes of the people. “He had defeated the gringo and his position was impregnable. He had the Mexican flag flying in his camp and numbers were flocking to his standard. He was the champion of his race— the man who would right the wrongs of the Mexicans and drive the hated Americans to the Nueces.” While some Mexicans undoubtedly sympathized with Cortina and gave him aid, it is also a matter of record that others, at great personal peril, joined in the fight to defeat him. Despite this fact, however, the Anglo-Texans believed that every Mexican along the border was in league with Cortina and would, if given a chance, “murder every white inhabitant.” At the request of the American government, Díaz finally brought the Cortina War to a close in 1873 by making Cortina his prisoner; but, as Walter Prescott Webb has written, “the evil consequences lived on.”

6. “The Dead-Line of Sheriffs” In the period from the close of the Civil War to 1880, there was nothing resembling “law and order” in the territory between the Nueces and the Rio Grande where friction between Anglo and Hispano was intense and continuous. In this strip of territory resided about eighty percent of the Mexicans then living in Texas. For fifty years after 1846, this territory was known as “the dead-line of sheriffs.” American officials refused to provide Catholic bishops safe conduct through the area and even the Texas Rangers hesitated to enter it. Neglected during the Civil War, great herds of cattle roamed wild in the brush country and plundering expeditions crossed and recrossed the border as cattle-stealing became an accepted business. To complicate matters, Mexico had established a narrow strip of territory on its side of the Rio Grande in which goods could be sold free of custom charges and duties. The existence of this “free zone” was an open invitation to smuggling and greatly annoyed American merchants in the Texas towns. From 1871 to 1875 the whole border was aflame with a type of lawlessness and violence even worse than open warfare. Historians have despaired of listing the murders committed by both sides and have never succeeded in counting the number of raids. On dozens of occasions, American troops were sent on expeditions into Mexico; nor were Mexicans much more respectful of American sovereignty.

80  North from Mexico

On May 8, 1874, a band of Mexican outlaws murdered four Anglo-Texans at Penascal; and, earlier, Albert Garza, a Mexican cattle-thief, raided far and wide in Texas. “Ghastly murders,” writes Dr. Taylor, “and shootings of Mexicans, each attributed, probably often correctly, to the other race, became not infrequent occurrences.”* General Ord reported that the raids were so frequent that the whole territory was entirely in the possession of cattle-raiders and bandits and that it was utterly impossible to “execute the laws at all.” Formerly Mexicans had borne the brunt of frontier lawlessness but, after 1870, the tide was somewhat reversed, and it was the Anglo-Texans who were on the receiving end. The raids culminated in March 1875, when a band of 150 Mexicans crossed the border near Eagle Pass and raided as far east as Corpus Christi. For this daring raid, the retaliation of the Texans was swift, violent, and indiscriminate. Bands of Texans raided Mexican settlements, burning houses, shooting Mexicans, spreading terror throughout south Texas. Large parties of mounted, well-armed men, wrote N. A. Jennings, “committed the most brutal outrages, murdering peaceful Mexican farmers and stockmen who had lived all their lives in Texas.” The adjutant-general of Texas, who can hardly be accused of pro-Mexican bias, reported that parties of Anglo-Americans had “banded together with the object of stopping the killing of cattle for their hides, but have themselves committed the greater crimes of murder and arson.” Merchants in Corpus Christi began to complain that “every good Mexican is afraid to navigate the roads on horseback or with carts, and the business in these parts has commenced turning into another channel, where less risk is found.” The temper of the feeling, on both sides, is indicated in a report of General Steele in 1875 in which he said that “there is a considerable Texas element in the country bordering on the Nueces that think the killing of a Mexican no crime” and a collection of “Mexican thieves and cut-throats who . . . think the killing of a Texan something to be proud of.” Reminiscing about the period, in 1929, a Texan told Dr. Taylor that “Mexicans despised us, and we hated the Mexicans like a human hates a rattlesnake.” Throughout this period, there were elements in Texas who were deliberately fomenting disorder and violence in the hope that the United States would take another slice of Mexican territory; and it is a matter of record that in 1876 President Hayes toyed with the idea of provoking a war with Mexico to divert attention from the shady deal by which he had robbed Tilden of the presidency.1 For this period as a whole, there is simply no telling, as J. Frank Dobie has said, “how many Mexicans bit the dust.” Naturally, robbery and theft went * Reprinted from An American-Mexican Frontier by Paul S. Taylor. Copyright, 1934, by The University of North Carolina Press.

“Not Counting Mexicans”   81

hand in hand with physical violence. “There is a disposition,” as one witness testified before a congressional committee in 1875, “on the part of some Americans, which crops out every once in a while, not to respect the property rights of Mexicans living southwest of the Nueces River.” The “beef packeries” on the American side of the Rio Grande often winked when Texas cattlemen brought in large herds of cattle clearly marked with Mexican brands. It was during this period, writes Garrison, that “large bodies of land that now have enormous value were then secured [from Mexican settlers], sometimes legally and sometimes illegally, for almost nothing.” On the other hand, Mexican cattle-thieves, in raiding Texas herds, laughingly said that they had come to collect “las vacas de tata” or “grandfather’s cattle.” One of the last large-scale raids was that organized by Catarina Garza in 1892 whose raid on the Norias Ranch is celebrated in a famous corrido.2 “Not only were the Mexicans bamboozled by the political factions,” writes Walter Prescott Webb, “but they were victimized by the law. One law applied to them and another, far less rigorous, to the political leaders and to the prominent Americans. The Mexicans suffered not only in their persons but in their property. The old land-owning Mexican families found their titles in jeopardy and if they did not lose in the courts, they lost to their American lawyers. The humble Mexican doubted a government that would not protect their person and the higher classes distrusted one that would not safeguard their property.” In an official investigation in 1878, the American consul in Matamoros testified that the authorities were never interested when a Mexican was killed in Brownsville; but, if a “white man” was molested in any way, “there is generally a great fuss made about it by those not of Mexican origin.” As late as 1879 the Anglo residents of Crio Canyon posted an order commanding all Mexicans to leave the area within three days. “In passing through Bee county,” said Senator Dwyer, “we heard of a Mexican, a quiet citizen, who had been brutally murdered a few days before our arrival, by several Americans because the Mexican would not go and play the fiddle for them.”

7. The Salt War In 1877 there were about twelve thousand people living along the Rio Grande at El Paso, all but eighty of whom were Mexicans. The Mexicans, writes Dr. Webb, “felt that Texas was by right still a part of Mexico” for El Paso had been “Mexican” for nearly three hundred years. After the Mexican-American War, a few Anglo-Americans appeared on the scene, monopolizing the government positions and showing a general tendency to take over. About a hundred miles east of El Paso was a salt mine which the Mexicans had discovered in 1862. By general consensus the Mexican residents of El Paso had been accorded the privilege of digging salt at the mine, without charge, for their personal needs.

82  North from Mexico

An ambitious American acquired control of the salt mine, by a series of devious maneuvers, and announced that henceforth it would be operated as a private monopoly. Outraged by this action and inflamed by the demagoguery of Father Borajo, a local priest, a mob of El Pasoans seized the city on October 10, 1877, killed three Anglos, and committed property damage that ran into thousands of dollars. In the process of restoring “law and order,” the usual retaliations were committed with a number of Mexicans being killed and several more being lynched. As might have been expected, “bitter hatreds were sown” as a consequence of the short-lived Salt War. From the inception of the Díaz regime, a degree of quiet prevailed along the border until the outbreak of the Mexican Revolution in 1910. But, even during this period, incidents were constantly occurring. When a Ranger shot and killed Ramón de la Cerda of the King Ranch in 1902, a general resentment against the Rangers flared up all along the border. Annoyed by this protest, a Ranger proceeded to whip a Mexican boy with a quirt and, shortly afterwards, Albert de la Cerda, a brother of the slain man, was shot by the Rangers. Incidents of this kind, with which the record is studded, served to keep the old antagonisms alive.

8. Open Season on Mexicans From 1908 to 1925, the whole border was aflame, once again, as revolution engulfed Mexico. No one knows how many American and Mexican civilians were killed along the border in these years but the estimates, according to Dr. Webb, range from five hundred to five thousand. As war approached in Europe, the Texans inevitably suspected the Mexicans of being in league with the Germans. “There is a fear constantly stored away in the back of the El Pasoan mind,” wrote Tracy Hammond Lewis, “that these Mexicans will take it into their heads to have an especially-appointed uprising at the expense of the Americans.”3 Fearful of the revolution, Mexican cattle-owners drove tens of thousands of cattle across the border to cash in on wartime beef prices in the United States. Once the ranges were deserted, the feed became, of course, very good on the Mexican side. Raiders then crossed the border, rounded up thousands of American-owned cattle, pastured them in Mexico, and later sold them to one or another of the various factions fighting in Mexico. As a consequence, property losses ran into the hundreds of thousands of dollars. From Brownsville to Calexico, raiders crossed and recrossed the border, exploiting the confusion which prevailed on both sides of the line. On March 9, 1916, Francisco Villa spread terror up and down the border with his raid on Columbus, New Mexico; and, before much time had passed, the Pershing expedition was deep in Mexican territory. Over two thousand postcards a day were sold in El Paso

“Not Counting Mexicans”   83

depicting “Mexican atrocities” while American troops marched through the streets singing: It’s a long, long way to capture Villa; It’s a long way to go; It’s a long way across the border Where the dirty greasers grow.

In this bloody seventeen-year period, hundreds of innocent civilians were killed. “Americans,” writes Ernest Gruening, “continued to be killed by the vengeful Villistas, at times for no other reason than that they were ‘gringos.’ Mexicans likewise were killed in Texas chiefly because they were ‘greasers.’” In an article in World’s Work, George Marvin reported that “the killing of Mexicans .  .  . through the border in these last four years is almost incredible. . . . Some rangers have degenerated into common man-killers. There is no penalty for killing, for no jury along the border would ever convict a white man for shooting a Mexican. . . . Reading over the Secret Service records makes you feel almost as though there were an open game season on Mexicans along the border.” Carranza, in a well-documented report, charged that 114 Mexicans had been murdered on the American side and a number of American officials acknow­­ ledged the accuracy of the charge. A formal protest filed by the Mexican ambassador in 1912 complained of the mistreatment of Mexicans in California and Texas and listed any number of lynchings and murders. On November 11, 1922, a Mexican, Elías Zarate, was lynched in Weslaco, Texas. Zarate had been arres­­ ted after a fist fight with an American and the Mexican consul had warned the authorities of the danger of mob violence. In its issue of July 12, 1922, The Nation documented a series of cases, all occurring in Texas, in which Mexicans had been brutally assaulted; in some cases, murdered. Following an old-established pattern, the authorities in Breckenridge, Texas, warned all Mexicans to depart overnight. The lawlessness became so widespread that Secretary of State Hughes had to warn the governor of Texas that some action would have to be taken to protect Mexicans. In an editorial of November 18, 1922, the New York Times said that “the killing of Mexicans without provocation is so common as to pass almost unnoticed”—nearly a hundred years after the signing of the Treaty of Guadalupe Hidalgo. Much of the lawlessness against Mexicans in Texas had an official or semiofficial status, for the Texas Rangers had become a kind of “black-and-tan” constabulary bent on terrorizing the Mexican population. Jovita Gonzáles quotes a local historian to the effect that the Rangers had executed, without due process of law, between one hundred and three hundred Mexican residents of the border counties. “The Rangers,” wrote Tracy Hammond Lewis,

84  North from Mexico

“are only cold-blooded where the Mexicans are concerned, and this solely because they have learned it is the one manner in which they can be properly handled.” J. T. Canales, a Mexican-American member of the Texas legislature from Brownsville, filed formal charges against the Rangers over the mistreatment of Tejanos and there must have been some merit to these charges for the number of Rangers was shortly reduced and, some years later, the organization was disbanded. This eruption of violence against persons of Mexican descent had important international repercussions. For example, Carranza at first refused to permit Mexico to join the League of Nations on the ground that the Covenant of the League did not ensure racial equality. Nothing so much infuriated the influential anti-Yankee Latin-American publicist Manuel Ugarte as the conditions which he had observed along the border. “From the very frontier,” he wrote, “the irreconcilable opposition between the two communities presents itself vividly and obviously. The Anglo-Saxon, hard, haughty, and utilitarian, infatuated with his success and his muscular strength, improvises towns, dominates nature, imposes everywhere the impress of his activity and ambition; and, like the Romans in their palmy days, has as his auxiliaries and servants the subject races—Indians, Chinese, Africans—who gather up the crumbs of the feast in return for discharging their subaltern tasks. As opposed to him, the Mexican . . . continues in his easy-going customs and accepts the fruits of the earth.”4 The general attitude of Latin-Americans was reflected in an editorial of May 15, 1922, in Heraldo, published in Mexico City: It is thoroughly irritating that while in our country American citizens enjoy ample guarantees and when anything happens to them it is settled by the United States consuls, in that country, on the other hand, Mexicans are still being killed without any effort by the American authorities to punish the murderers. . . . Up to the present time, not a single person has been electrocuted for killing a Mexican, no matter how brutally or basely he might have perpetrated the crime.

Viewing this record in retrospect, one can thoroughly appreciate the comments, without sharing the conclusions, of F. L. Olmsted5: . . .between our South American and the Mexican there is an unconquerable antagonism of character, which will prevent any condition of order where the two come together. . . . The mingled Puritanism and brigandism, which distinguishes the vulgar mind of the South, peculiarly unfits it to harmoniously associate with the bigoted, childish, and passionate Mexicans. They are considered to be heathen; not acknowledged as “white folks.” Inevitably they

“Not Counting Mexicans”   85

are dealt with insolently and unjustly. . . . Guaranteed by the treaty of Guadalupe Hidalgo, equal rights with all other citizens of the United States and of Texas, the whole native population of county after county has been driven, by the formal proceedings of substantial planters, from its homes, and forbidden, on pain of no less punishment than instant death, to return to the vicinity of the plantations.

7

Gringos and Greasers

In the Southwest, Anglos have always been “gringos” to the Hispanos while Hispanos have been “greasers” to the Anglos. The two terms pretty accurately reflect the measure of mutual esteem which has prevailed. For many years the origin of the word “gringo” was traced to a song—“Green Grow the Rushes, O!”—which the Yankees sang in 1846 when they marched into Mexico. Actually “gringo” is to be found in all Spanish dictionaries. Defined as a corruption of griego or “Greek,” it is said to be a nickname applied to foreigners. Hablar en gringo is to talk gibberish; much as Americans would say, “It’s all Greek to me.” In popular Southwestern usage prior to the conquest, “gringo” referred to any foreigner who spoke Spanish with an accent and was first recorded in the New English Dictionary in 1884.1 It should be noted that the term, as used by Mexicans, is less insulting in its implications than “greaser.” The origin of “greaser” has been variously explained. It is said that a Mexican once maintained a small shop at the crest of Raton Pass where the ox-carts and wagons of the Santa Fe Trail were greased before they made the descent to the New Mexico plateau; hence a Mexican was literally a “greaser.” In California the term has been traced to the days of the hide-and-tallow trade and is said to have first been applied by American sailors to the Indians and Mexicans who loaded the greasy, tick-ridden hides on the clipper ships. It is also said to have had some relation to sheep-shearing. The term was certainly well known in early California for Harris Newmark tells of the lynching of an Anglo in Los Angeles in 1854 who objected most strenuously to being shuffled off “by a lot of greasers.”2 The term is defined by Vizetelly as “Mexican; an opprobrious term,” and a note states that it is “California slang for a mixed race of Mexicans and Indians” with Bret Harte’s “Carquinez Wood” being cited as authority for this usage. The definition to be found in the Century Dictionary is: “a native Mexican or native Spanish-American, originally applied contemp­­ tuously by the Americans of the Southwestern United States to Mexicans” (emphasis added). In any case, gringo and greaser it has always been, and it is in reference to the hostility and opposition which these terms imply that the

88  North from Mexico

pattern of Anglo-Hispano relations outside Texas will be discussed in this chapter.

1. The Bloodless Conquest The conflict between greasers and gringos never assumed the proportions in New Mexico that so long prevailed in Texas. Being off the main path of AngloAmerican expansion, New Mexico grew very slowly. At the time of the conquest the population of the territory was 61,525 which was made up almost entirely of Spanish-speaking people. From this figure the population increased to 87,034 in 1860 (79,249 native-born); and then to 91,784 in 1870 (83,175 native-born). Hence the population of the state remained ninety percent Spanish-speaking during the first two decades after the conquest. In fact it has only been of fairly recent date that the Anglo-Americans have achieved a slight numerical ascendancy. Furthermore the two groups have always been geographically segregated in New Mexico, with the Spanish-speaking element concentrated in the north-central part of the state and the Anglo-Americans in the larger towns and in the southern and eastern areas. Distance has made for a formality and civility the absence of which is most conspicuous in Texas. The Anglo-American immigration to New Mexico was of an entirely different character, in quality as well as quantity, from the immigration that so quickly engulfed the Spanish-speaking in Texas and California. Only a hundred or so Anglo-Americans had settled in New Mexico prior to 1846 and most of these had married into prominent “native” families. During the height of the commerce on the Santa Fe Trail, not more than 150 Anglo-Americans entered the province each year and most of these visitors departed as soon as their wares had been sold. After the conquest, the first Anglo-American settlers were federal officers, followed by territorial officials, lawyers, and merchants. Since there was never a large influx of farm families, active competition for resources was kept at a minimum until about 1880. The economic conquest of Texas and California was achieved by the Anglo-Americans overnight and by the use of superior force; in New Mexico the process was indirect and subtle and took the form of a gradual assertion of dominance through manipulation rather than by outright expropriation. At the outset there was less opposition to the Anglo-Americans in New Mexico than in Texas for neither the pobres nor the Indians felt much loyalty to Mexico. Twitchell described their allegiance to Mexico as “not overly strong,” while Bancroft said that it was “only nominal.” Nor were the ricos united in opposition to the Americans. Those who had profited by the commerce which developed with the opening of the Santa Fe Trail saw in American rule the hope of expanding markets and better communications. The resistance of those who opposed the conquest, moreover, was frustrated by the treachery of

Gringos and Greasers   89

their leader, Governor Manuel Armijo, who is said to have presented New Mexico to the American agent, James W. Magoffin, for a handsome consideration. Under the terms of the Treaty of Guadalupe Hidalgo, Mexican residents were given one year in which to leave the borderlands; those who remained automatically became citizens of the United States. According to Bancroft, not more than 1,200 residents of New Mexico left the province during the year of grace. There was still another reason for the lack of opposition to the conquest in New Mexico. In 1841 the Republic of Texas had sent an expedition of 270 soldiers to seize New Mexico. Intercepted by the New Mexicans, the invading army was defeated in battle and its leaders were sent as prisoners of war to Mexico City. By way of retaliation, the Texans began to raid the Santa Fe caravans with such regularity that all commerce was suspended for one year (1843). George Kendall’s famous account of the Texas expedition, emphasizing the cruelty of the captors, has been called the Uncle Tom’s Cabin of the Mexican-American War. This initial clash was merely the first of a long series of incidents which served to create and maintain a traditional antipathy bet­­ ween Texans and New Mexicans. The antipathy was so keen in 1846 that the New Mexicans were inclined to look upon the arrival of American troops as a measure of protection against further encroachments on the part of the Texans.

2. The Ghost of Father Martínez Although the pattern of antecedent conflict was lacking in New Mexico, the conquest was not entirely bloodless. Shortly after the arrival of federal troops in the province, the New Mexicans organized a revolt in the course of which the first American civil governor, Charles Bent, was assassinated. A most interesting New Mexican is said to have played a major role in organizing the brief revolt. Born in Abiquiu, educated in Durango, Father José Antonio Martínez had been named curate of Taos in 1830. A man of great energy and outstanding ability, Father Martínez was the undisputed boss of Taos. The owner of several ranches and a flour mill, he published the first newspaper printed in New Mexico, El Crepúsculo—“The Dawn.” Something of a freethinker, he denounced heavy tithes and church fees; founded the first schools in the province; and served as a member of the provincial assembly. In line with the prevailing moral standards of the New Mexican curates, Father Martínez had several wives and his progeny were numerous. Many of his offspring still live in Taos and are rightly proud of their paternity. A shrewd man, Father Martínez once observed that a republic was a burro on which lawyers jog along much better than priests. In the opinion of Erna Fergusson, Martínez still ranks as perhaps “the outstanding New Mexican.”

90  North from Mexico

Although the historians have been unable to define the precise responsibility of Father Martínez for the ill-fated revolt of 1846, they are agreed that he was one of the principal organizers of the revolt whose nominal leader was a peon, one Pablo Montoya. Shortly after the assassination of Governor Bent, federal troops arrived at Taos under Colonel Price and laid siege to the old church in which the rebels had barricaded themselves. About 150 New Mexicans were killed in this engagement; some twenty-five or thirty prisoners were shot down by firing squads; and many of those who surrendered were publicly flogged. Colonel Price’s troops are said to have been so drunk at the time that the Taos engagement was more of a massacre than a battle. In the restoration of order, Kit Carson, who, like Governor Bent, had married a Spanish-American, played a leading role, as did Domiciano Vigil, a native New Mexican of the rico class who became acting governor. While the revolt was quickly suppressed, it left a heritage of ill-will and bitterness which survived for many years. George F. Ruxton, the English traveler, reported that he had “found over all New Mexico the most bitter feeling and the most determined hostility against the Americans who . . . by their bullying and overbearing demeanor toward them [the natives] have in great measure been the cause of this hatred.” With the arrival of Bishop Lamy in 1851, Father Martínez was suspended as a priest and later denied communion, all as part of Lamy’s general effort to reform the worldly, pleasure-loving curates of New Mexico. Despite these retaliatory measures, however, Father Martínez continued to preside in his church until his death in 1867 at the age of seventy-four. Some years later, still another priest emerged as a leader of the native-born New Mexicans and achieved, for a time, considerable power and influence. This priest, Father Gallegos, was excommunicated by Bishop Lamy for concubinage. Perhaps because they were Mexican-born, both Father Martínez and Father Gallegos were close to the people, sharing their feelings and sentiments. The leadership which they provided for the native peoples is supposed to have had a marked influence on Catholic policy. Since Father Gallegos was excommunicated, the Church has been careful to select French, German, and Belgian priests for service in New Mexico.

3. Disturbance in Socorro While the Taos revolt was, as Harvey Fergusson has written, “the last battle on the Rio Grande between the old order and the new,” a number of later incidents indicate that the conflict continued in other forms. Francisco Perea, one of the ricos, once ran for office on a platform which opposed the building of the rail lines. “We don’t want you damned Yankees in the country,” he said; “we can’t compete with you; you will drive us out.” And, at a place called Cow Creek

Gringos and Greasers   91

Hill, a pitched battle was fought in 1880 between railroad construction workers and local Spanish-speaking residents.3 With the railroads, of course, came the cattlemen. In the 1880s a number of Texas cowboys began to drive their herds, and occasionally their neighbors’ herds, into the region around Socorro, north of El Paso on the Rio Grande. They were men, wrote William French, “who lost no opportunity of displaying their hatred of Mexicans. To them all Mexicans were ‘greasers’ and unfit associates for the white man.”4 A few brave words about Davy Crockett and the Alamo were all that was needed to work these men into a Mexican-killing mood. When the Texans began to invade Socorro, shooting up the town and terrorizing the natives, many of the Mexicans abandoned their homes and ranches and crossed the border into Mexico. In the town of Dona Ana alone some sixty New Mexican families moved out when the Texans moved in. John R. Bartlett, of the U.S. Border Commission, reported that “immediately preceding and after the war with Mexico, the Mexican population was greatly annoyed by the encroachments of the Americans and by their determined efforts to despoil them of their landed property. This was done by the latter either settling among them or in some instances forcibly occupying their dwellings and cultivated spots.” The murder of Manuel Otero by Texas cowboys at Estancia Springs was merely one of many similar incidents of the period.5 Control of the rich pasture lands of western New Mexico and eastern Arizona was the prize for which Anglos and Hispanos contended along the lower Rio Grande in New Mexico. Out of this conflict emerged one of the present-day folk heroes of New Mexico—Elfego Baca. A native New Mexican, Baca greatly resented the arrogance of the Texans and volunteered to serve as a deputy sheriff. When the Texans went on their next shooting spree, he proceeded to arrest a Texas cowboy. Once word reached the ranches that the unheard-of had happened—that a “Mexican” had actually arrested a “Texan”—the cowboys came trooping in for the kill. In the meantime, however, Baca had lodged his prisoner in the jail and had carefully barricaded himself in an adobe. When the dust of battle finally subsided, after a night and a day of blazing guns, Baca still held possession of the adobe. A truce was then negotiated and the Texans gathered up their dead and wounded and departed from Socorro. Once Baca had demonstrated that New Mexicans could shoot as accurately as Texans, the encroachment of Texas folkways in New Mexico came to a dead-halt. In the Panhandle of Texas still another Mexican folk hero emerged: Sostenes L’Archevéque, who was born in Santa Fe, of a French father and a MexicanIndian mother. After his father was shot down by Anglo-Americans in the little town of Sapello in eastern New Mexico, the young man lost little time in acquiring twenty-three “gringo” notches on his gun—two more notches than were to be found on the gun of Billy the Kid. In the Panhandle, however,

92  North from Mexico

Sostenes overreached himself. After killing and robbing an Anglo-American sheepman, as part of his bitter vendetta, he was lured into a trap by his own people. It was said that his executioners felt that they must avenge his crime in order to prevent wholesale retaliations. If this was their motive, then the sacrifice of Sostenes was needless, for the reprisals were swift and cruel. For a week or more, Texas cowboys ran wild in Tascosa, shooting up the Mexican plazas, killing several innocent Mexicans, and lynching those suspected of complicity in the murder of the sheepman. It was largely as a result of these reprisals that the Mexicans retreated from the Panhandle to New Mexico.

4. The Don System Confronted by an overwhelming Spanish-speaking majority whose property and political rights were guaranteed by the Treaty of Guadalupe Hidalgo, the Anglo-Americans were forced to proceed by stealth and indirection in their campaign to acquire control of New Mexico. They had, however, one great advantage: control of the territorial government and the courts through the appointive power. When one compares the celerity with which California and Nevada were admitted to the union with the prolonged struggle for statehood in New Mexico, it is readily apparent that forces were at work, both within and without the state, to delay admission until an Anglo-American majority had been established. “The mass of the people are Mexicans,” as Thaddeus Stevens explained to his colleagues in Congress, “a hybrid race of Spanish and Indian origin, ignorant, degraded, demoralized and priest-ridden.” One of the leaders in the fight to extend the franchise to the Negro freedmen in the South, Stevens showed little interest in democracy in New Mexico. Whatever the motive, the federal government ruled New Mexico for sixty-three years as a dependent province. Even under the forms of territorial government, however, it was necessary for the Anglo-Americans to cultivate the ricos through whom they could control the Spanish-speaking majority. Jealous of their power, the ricos used the patron-peon relationship to develop what came to be called “the don system” of politics in which the haciendado was said to have voted his sheep as well as his peons. For years the phrase “New Mexico politics” was synonymous with political corruption. The corruption had little to do with the character of the people: it was a function of the political system. Most of the governors and other territorial officials were, of course, Anglo-Americans; but the territorial legislature was dominated by Hispanos from its establishment until about the time New Mexico was finally granted statehood in 1912. Throughout its history, the territorial legislature was made up of members who belonged to, or were controlled by, some twenty prominent families of the rico class. Thus to control the legislature and the territorial delegate in Congress, it was

Gringos and Greasers   93

necessary for the Anglo-Americans to work through and to ally themselves with the rico class. Out of this alliance there developed what was called “the Santa Fe Ring,” a small, compact group of Anglo-American bankers, lawyers, merchants, and politicians who dominated the territory through their ties with the ricos who in turn controlled the votes of the Spanish-speaking. “The professional connections made by them,” writes Twitchell, “with the representative families of the Rio Abajo and Rio Arriba in guarding their interests in the matter of Spanish and Mexican land grants enabled them to maintain what seemed to be an invincible position in the control of all matters of business or of a political character.” It was the members of this “ring” who manipulated the Indian Bureau; controlled the allocation of contracts to supply the army posts; dictated territorial appointments; and exercised a great influence over the courts. On at least two occasions, however, the neat alliance between the Santa Fe Ring and the ricos was seriously challenged. As a boy, Colonel Francisco Chaves had been told by his father, a member of the rico class, that “the heretics are going to overrun the country. Go and learn their language and come back prepared to defend your people.” From New Mexico, young Chaves was sent to school, first in St. Louis, and then to New York, where he graduated from the College of Physicians and Surgeons. During the Civil War, he was one of a number of native-born New Mexicans who fought with the Union Army. Returning to New Mexico in 1867, he ran for the post of territorial delegate in Congress against an Anglo-American and was finally seated after a bitter campaign which had resulted in a contested election. In 1869 he was reelected as delegate, following another heated and bitterly fought campaign. Then, with the election of 1871, came the “Mesilla Riots” in which nine men were killed and forty or fifty were wounded. Most of the casualties were Spanish-speaking supporters of Colonel Chaves, who was defeated. These were statewide campaigns with the Anglos and their rico allies being lined up against the Spanishspeaking lower classes. Following his defeat, Colonel Chaves served for twelve years as president of the territorial state senate. Here, again, he waged an unrelenting campaign against Anglo-American bossism and opposed the alliance which existed between the Santa Fe Ring and the ricos. Later, as the first superintendent of public instruction in New Mexico, he put up a gallant although unsuccessful fight to bring about an adequate allocation of federal funds for the schools of the territory. In 1904 Colonel Chaves was murdered “under the most mysterious circumstances.” “The charge was made,” writes Twitchell, “that it was political assassination involving men of great prominence in Santa Fe and elsewhere.” On the death of Colonel Chaves, leadership of the Spanish-speaking element passed to Octaviano A. Larrazolo, the first person born in Mexico to serve as

94  North from Mexico

governor of an American state. It was largely as a result of his effective leadership that the New Mexico constitution incorporated the guarantees of the Treaty of Guadalupe Hidalgo safeguarding the rights of Spanish-speaking residents. “Whether for good or evil,” writes one historian, “it is because of the impress of his day that the cleavage between the descendants of the Spanish conquerors and colonists and those who came from other states continues to be felt in political life, and is felt even in business, in the professions, and in social activities.” From that day to this the demand by the Spanish-American group for at least one-half of the candidates on the tickets of the two major parties has been a sine qua non in every state campaign. Larrazolo was the second Spanishspeaking governor of New Mexico (Ezequiel de Baca, the first, was elected in 1916), and the first Spanish-speaking New Mexican to serve in the U.S. Senate. The careers of Colonel Chaves and Larrazolo serve to indicate that beneath an apparently calm surface much ethnic ill-feeling has long existed in New Mexico. When Bronson Cutting reversed the usual pattern of New Mexico politics by espousing the cause, not of the ricos, but of the Spanish-speaking majority, the Santa Fe Ring suffered its first serious defeat. The undisputed boss of New Mexico politics, Cutting was regarded by most Anglo-Americans as a renegade, a traitor to his “race” and class. The first Anglo-American politician to champion the cause of the Spanish-speaking element in the Southwest, Cutting is still revered in New Mexico. The following that he developed with such tact and skill is today represented in the U.S. Senate by Dennis Chavez, a native-born New Mexican. Throughout the borderlands Mexican-Americans are still struggling, often in a most confused and inchoate way, to overthrow a dual dominance: an Anglo-American hegemony imposed through the instrumentality of a quisling-like, upper-class “Spanish” element. “The poverty and social condition of the lower classes,” wrote Twitchell, “the system of peonage which prevailed, and above all the attitude of the upper or official class displayed and maintained toward those of their own people whom they regarded in every way as servitors, may well be said to account for social infirmities, encouraging and patronizing which it may safely be said was always some foreigner who was a more or less active partner in the enterprise.” While conflict between Anglos and Hispanos is less evident in New Mexico than elsewhere in the borderlands, it has long existed and, of late years, has become more pronounced. It is often masked by religious issues, as in the consistent opposition of the Catholic Church to the work of Protestant missions and in the opposition of Protestant groups to the employment of nuns and Catholic brothers as teachers in the public schools. One can even detect an ethnic cleavage within the Catholic Church. For the larger communities usually have one church for the English-speaking; another for the Spanish-speaking. From the earliest times, the Protestant sects have separated Spanish-Americans

Gringos and Greasers   95

from the Anglos in their churches, schools, and other institutions. Most of the larger towns are divided into an “Old Town” and a “New Town,” or an Anglo and Hispano section, with the high school being customarily located in the “new” or Anglo community. The cleavage is most apparent and the group-consciousness of the Spanish-speaking is most pronounced in the areas which have received the heaviest Anglo-American influx. “The participation of Spanish-American people in the social life of the Anglos,” writes Carolyn Zeleny, “is even more restricted than that of Anglos in Spanish-American society. During two years’ residence in an Anglo-dominated city, the writer never met a Spanish-American at a social gathering.” Originally most of the mixed marriages were between Anglo men and Spanish-speaking women of the rico class; but nowadays mixed marriages among the working class are not uncommon as shown by the disconcertingly large number of Carmencita O’Briens and Juan O’Rourkes to be found in the schools of the larger towns. While the mores preclude an open discussion of prejudice—both groups being most reluctant to discuss it—there can be no doubt that prejudice serves as a mechanism by which the rise of Spanish-Americans to positions threatening the advantaged place of the Anglo-Americans is retarded if not prevented.6

5. The Buffer State In Arizona relations between Anglos and Hispanos at the outset were, as Dr. Frank C. Lockwood has observed, “on the most friendly terms; neither race scorned the other.” Other long-time residents of Arizona, such as John C. Vosburg, have also commented on the surprising amount of “good will that existed between Americans and Mexicans” who were united, in the pioneer phase, by a mutual fear of the Apaches. In the local annals, one reads of many marriages between Anglo men and Spanish-speaking women. Such wellknown “pioneer” residents of Tucson as L. J. F. Jaeger, Samuel Hughes, Hiram S. Stevens, Solomon Warner, Peter H. Brady, Fritz Contzen, and William S. Curry, all married “Spanish” ladies. It was not long, however, before elements of conflict developed. The rich mining “strikes” around Tombstone in the late eighties, and the campaigns against the Apaches, created a lively demand for beef cattle at fancy prices. As this demand mounted, so-called “Texas cowboys,” more often Texas outlaws, began to raid Mexican ranches along the border. One “outfit” alone is said to have harbored more than three hundred outlaws in the San Pedro and Sulphur Springs valleys. The business of the “outfits,” of course, was to raid Mexican ranches and to drive off the herds which the Mexicans attempted to get through to the mines and army posts. The celebrated Clanton “gang” participated in these raids, often killing, as Paul I. Wellman has written, “a dozen or so Mexicans in a negligent

96  North from Mexico

way.”7 W. M. Breakenridge, author of a spirited history of Tombstone, once said that the hatred of the Texans “was so bitter that they had no compunction about stealing from Mexicans or shooting them and robbing them whenever they got an opportunity.” Later the situation was somewhat reversed and the Mexicans began to do most of the killing and raiding. Throughout the Southwest, the conflict between gringo and greaser often masked a conflict between cattle and sheep interests. “The two callings,” writes Mr. Wellman, “fostered the mutual distrust and dislike which the races held for each other.” Wherever sheep were driven, Mexicans came with them; wherever cattle went, Texans were to be found. Hence the hatred of cattlemen for sheepmen was, in most cases, a hatred of Anglos for Hispanos. During the period that this warfare prevailed, Mexican sheepherders were murdered in every Western state. Teofilo Trujillo, of Taos, who pioneered sheep-raising in southern Colorado, was clubbed to death by “cowboys” in 1884. In the same raid, his home was burned and his herds were driven from the range. Cowboys employed by the Chiricahua Cattle Company rode into the sheep camp of Don Pedro Montano in Arizona one morning in 1889 and murdered three Mexican sheepherders in cold blood. Nearly every sheep-raising section in the West has a record of one or more such “incidents.” Having completed its line to Casa Grande, the Southern Pacific Company discharged some 1,500 Mexican workers in the summer of 1879, most of whom, instead of returning to Mexico, decided to settle in Phoenix. Since Phoenix was a new Anglo-American town, the influx of this number of unemployed Mexican workers was immediately resented and feeling between “greaser” and “gringo” soon reached the boiling point. One day a drunken Mexican, brandishing a saber, dashed on horseback through the streets of Phoenix shouting Cortina’s famous war cry: “Maten los Gringos!” By way of additional provocation, the unemployed Mexican workers gathered in a park where they listened to harangues by their spokesmen against the railroad and the Anglo-Americans. In the meantime, leaders of the Anglo-American elements quietly assembled a lynch mob. Apparently reluctant to launch a frontal attack on the Mexicans, the mob took two Anglo prisoners from the jail, escorted them to the park, and there lynched them in the presence of the Mexicans. After the lynching, a spokesman for the Anglos made a speech in which he warned that a similar fate might await some of the Mexicans if the meeting was not immediately dispersed. Needless to say, the warning was heeded. Like Texas and California, Arizona has a long record of Mexican lynchings. One Mariano Tisnado was lynched in Phoenix on July 3, 1873; Leonardo Cordoba, Clement López, and Jesús Saguaripa were lynched in Tucson on August fourth of the same year, with a coroner’s jury defending the lynchings. Still another Mexican was lynched in Bisbee, a stronghold of anti-Mexican sentiment, on August 11, 1882. So firmly were the Anglo-Americans entrenched

Gringos and Greasers   97

in power that, in the 1890s, one reads of the enactment of ordinances outlawing Mexican fiestas in Arizona. There is also a record of a number of Mexican lynchings in Colorado.8

6. The Lynching of Juanita Mexican miners from Sonora were among the first emigrants to arrive in California after the discovery of gold. Staking out important claims in the “southern” mines—those in Calaveras, Tuolumne, Mariposa, and Stanislaus counties—they had made remarkable progress before the stream of AngloAmerican migration had reached California. Even before the arrival of the Sonorans, several hundred Chilean and Peruvian miners had reached the goldfields by the summer of 1848. Settled by Spanish-speaking miners, the towns of Sonora and Hornitos quickly became populous, world-famous mining camps. The first Anglo-American mining settlements were in the “central” and “northern” districts, along the American, Feather, Bear, and Yuba rivers. It was not long, however, before the Anglo-Americans began to invade the southern mines. “The Mexicans,” writes Walter Noble Burns, “who poured into California during the gold rush, were still inflamed with the anti-American prejudices engendered during the Mexican War. Their attitude towards Americans was hostile from the first and, in return, the Americans regarded them as secret enemies and treated them with frank contempt.” One of the first acts of the California legislature was the adoption of a foreign-miners’ license tax which was aimed specifically at eliminating the competition of Mexican miners. Shortly after this act was passed, a mob of two thousand American miners descended on Sonora, “firing at every Mexican in sight.” The camp was burned to the ground and a hundred or more Mexicans were rounded up and driven into a corral or stockade. During the week that the rioting lasted, scores of Mexicans were lynched and murdered. In the wake of the riots, most of the Mexicans abandoned their claims and fled to the Spanish-speaking counties in the southern part of the state. One of these former miners, Joaquín Murieta, became the leader of a famous band of Mexican outlaws. A year later, on July 5, 1851, a mob of American miners in Downieville lynched a Mexican woman who was three months pregnant. During the excitement of the previous day’s Fourth of July celebration, a drunken miner had broken into a shack in which the Mexican woman, whose name was Juanita, was living with a man who may or may not have been her husband. “In keeping with the characteristics of her race [sic],” writes Owen Cochran Coy.9 “Juanita had a quick passion.” When the miner returned the next day, some say to apologize, a dispute arose which ended with the fatal stabbing of the miner. Stephen J. Field, later a justice of the U.S. Supreme Court, made an eloquent plea to the miners to spare the life of Juanita. The miners heard him out and

98  North from Mexico

then proceeded with the lynching. With incomparable courage, Juanita adjusted the rope with her own hands, smilingly bade the miners “adios,” and swung from the scaffold. The first person to be lynched in California was a Mexican and vast research would be required to arrive at an estimate of the number of Mexican lynchings between 1849 and 1890. In the mining camps, every crime or reported crime was promptly blamed on some Mexican and lynching was the accepted penalty for crimes in which Mexicans were involved. “We can see only indirectly,” wrote Josiah Royce, “through the furious and confused reports of the Americans themselves, how much of organized and coarse brutality these Mexicans suffered from the miners’ meetings.” As violence mounted throughout the mining camps, the Mexicans took refuge in the “cow counties” of Southern California where, in canyon and foothill hideouts, they licked their wounds and plotted their revenge. The gente de razón enjoyed, of course, immunity from the violence which raged in the mining districts. Ties of marriage and bonds of commerce brought them quickly into the American camp. During the first years of the conquest, the Americans were at some pains to distinguish between “native Californians,” meaning gente de razón, and “Mexicans,” meaning “greasers” and cholos. For it was generally recognized that the gente de razón, if properly cultivated, could be of major importance in consolidating American rule in California. When the first constitutional convention was called, seven out of forty-eight delegates were “native Californians.” Needless to say, the cholos were not represented. But by 1876 Walter M. Fisher, the English journalist, could report that “the meanest runaway English sailor, escaped Sydney convict, or American rowdy, despised without distinction the bluest blood of Castile, and the half-breeds descen­­ ded from the Mexican garrison soldiers, habitually designating all who spoke Spanish by the offensive name ‘greasers’ for whom remains only the rust and the dust of a lost power.”10 The ease and swiftness of the victory over Mexico and the conquest of California had bred in the Americans a measureless contempt for all things Mexican. This feeling naturally found violent expression in California for there was really no government in the state from 1846 to 1850. The miners who made up the major element in the Anglo-American population were a tough and hard-bitten lot. Indeed the circumstances were unique for here a large body of restless, adventuresome, single men had been suddenly catapulted into a foreign land, with the excitement of gold in the air, and with no government of any sort to curb their predilection for violence and direct action. “Nowhere else,” wrote Josiah Royce, “were Americans more affected than here, in our lives and conduct, by the feeling that we stood in the position of conquerors in a new land . . . nowhere else were we driven so hastily to improvize [sic] a government for a large body of strangers.” It is not surprising, therefore,

Gringos and Greasers   99

that the manners and actions of the first Anglo-American immigrants to reach the state after 1848 produced a silent bitterness among the Californians which was to last for many years. Crimes of violence had been almost unknown in California prior to the conquest. “Perfect security for the person prevailed in California,” wrote the South American, Don José Arnaz, of his stay in the province in the years from 1840 to 1843. But, after the conquest, the lower classes became extremely dis­­ affected and their unrest often assumed a covert or “criminal” design. It was after 1846, wrote the historian J. M. Guinn, “that a strange metamorphosis took place in the character of the lower classes of the native Californians. . . . Before the conquest by the Americans they were a peaceful and contented people. There were no organized bands of outlaws among them. . . . The Americans not only took possession of their country and its government, but in many cases despoiled them of their ancestral acres and their personal property. Injustice rankles, and they were often treated by the rougher American elements as aliens and intruders, who had no right in the land of their birth.” Such, in general, is the origin of the much discussed “Mexican banditry” of the period. In the ‘fifties, the country between Los Angeles and Fort Miller in the San Joaquin Valley was infested with “Californian and Mexican outlaws” who raided the herds of cattle being driven north to the mines and looted the mining settlements. Two companies of rangers were recruited in Southern California to fight off the raids. Typical of the attitudes of the bandits was the statement of Tiburcio Vásquez, a native Californian, who was executed in 1852. “A spirit of hatred and revenge,” he said, “took possession of me. I had numerous fights in defense of what I believed to be my rights and those of my countrymen. I believed we were being unjustly deprived of the social rights that belonged to us.” The local county histories contain many references to the activities of these bold and daring outlaws: Vásquez, Joaquín Murieta, Louis Bulvia, Antonio Moreno, Procopio, Soto, Manuel García, Juan Flores, Pancho Daniel, and many others. Not a few of these men had fought on the side of Mexico in the war of 1846. Called El Patrio—“The Native”—by the Mexicans, Joaquín Murieta boasted that he could muster two thousand men. Many of the outlaw bands, in fact, contained a hundred or more men and were well organized for guerrilla fighting. “The racial loyalty of the Californians,” to quote from one local history, “not to mention the entanglements of family relationships with the outlaws, plus a tacit policy of non-interference among the old American population, resulted in a negligent tolerance of these evils which within five years after 1849 swept the local situation entirely out of hand.” Mexican banditry gave a color of justification to the practice of lynching Mexicans which soon degenerated from a form of vigilante punishment for crime to an outdoor sport in Southern California. In 1857, four Mexicans were

100  North from Mexico

lynched in El Monte; eleven in Los Angeles. Throughout the 1860s the lynching of Mexicans was such a common occurrence in Los Angeles that the newspapers scarcely bothered to report the details. Horace Bell, who was himself once indicted for killing a Mexican, describes any number of murders and lynchings, in which the victims were Mexicans, in his memoirs: Reminiscences of a Ranger and On the Old West Coast. The last reported lynching of a Mexican occurred in August 1892, when one Francisco Torres was lynched in Santa Ana. A homicide a day was reported in Los Angeles in 1854, with most of the victims being Mexicans and Indians. The previous year California had more murders than the rest of the states combined and Los Angeles had more than occurred elsewhere in California. The subordination of Mexicans in the social structure of California cannot be understood apart from this early-day pattern of violence and intimidation.

7. Birth of a Stereotype From Brownsville to Los Angeles, the first impressions which the AngloAmericans formed of the “native” element were highly unfavorable. Accurate and fair-minded observers like Josiah Gregg in New Mexico, Richard Henry Dana in California, and Frederick Law Olmsted in Texas, all drew essentially the same picture of the “idle, thriftless natives” by contrast with the charm of the ricos and the splendor of their bailes. The poverty and backwardness of the people were consistently stressed without any accompanying explanation of the factors—notably the isolation—which had produced such a “degraded” population. The puritanism of many of the Anglo-American observers was shocked by the “half-naked children” and the “immodesty” of the native women who dressed without benefit of underwear, petticoats, bustles, bodices, or long sleeves. More than one of these observers told, with unconcealed horror, of having seen native women step out of their red skirts and skimpy camisas and go bathing in the streams or of having seen women who drank, gambled, and smoked cigarettes. In these early impressions, carefully embalmed, one can find the outline of the present-day stereotype of the Mexican. “The greater part of them,” wrote Marmaduke, “are the most miserable, wretched poor creatures that I have ever seen, poor, petty, thieving, gambling, bull-baiting .  .  .” Essentially this same impression was formed by a wide variety of observers: men and women; officers, miners, surveyors, trappers, mountainmen, sea captains, and journalists. Passed along to those who were about to leave for the borderlands, repeated by all observers, these same stereotyped impressions were given national currency during the Mexican-American War and the patriotic sanction long continued. It was only natural, as Twitchell explained, that the Missourians who came to New Mexico over the Santa Fe Trail should have carried home

Gringos and Greasers   101

“unfavorable comments upon the character of the people when it was notorious that in the capital itself [Santa Fe] only the children of the dueños or ricos had any opportunity whatever to receive even the most rudimentary schooling. . . . A country without courts, officially bankrupt, and notoriously corrupt in its executive, certainly created anything but a favorable impression upon the minds of the Santa Fe traders.” The trouble was that the tales told by the traders “became the basis of American opinion of Mexican character in toto; a most unfortunate conclusion in its moulding of public opinion of a people who were not themselves properly chargeable with the racial delinquencies so glibly commented upon by the ‘strangers within the gates.’” Above all it is important to remember that Mexicans are a “conquered” people in the Southwest, a people whose culture has been under incessant attack for many years and whose character and achievements, as a people, have been consistently disparaged. Apart from physical violence, conquered and conqueror have continued to be competitors for land and jobs and power, parties to a constant economic conflict which has found expression in litigation, dispossessions, hotly contested elections, and the mutual disparagement which inevitably accompanies a situation of this kind. Throughout this struggle, the Anglo-Americans have possessed every advantage: in numbers and wealth, arms and machines. Having been subjected, first to a brutal physical attack, and then to a long process of economic attrition, it is not surprising that so many Mexicans should show evidences of the spiritual defeatism which so often arises when a cultural minority is annexed to an alien culture and way of life. More is involved, in situations of this kind, than the defeat of individual ambitions, for the victims also suffer from the defeat of their culture and of the society of which they are a part.

8

The Heritage of the Southwest

Unlike the Middle West, there were no rich, fertile valleys in the Spanish borderlands; no plains which invited the plow; no lakes well stocked with fish; no rivers to be used for navigation or harnessed for power; no forests to provide lumber and fuel. The few areas capable of cultivation required irrigation in a land where water was scarce. Learning to survive in this region was a harsh and difficult undertaking. Resources had to be carefully husbanded; communications were hard to establish and difficult to maintain; and isolation magnified every aspect of the problem of settlement. Yet the Spaniards, in a triangular relationship with Mexicans and Indians, succeeded in laying the foundations for the present-day economic structure of the region. Anglo-Americans in the Southwest have been the beneficiaries of three hundred years of experimentation, adaptation, and innovation. If one thinks of the Southwest in terms of mines, sheep, and cattle, and irrigated farming, then it is readily apparent that the underpinnings of the economy are of Spanish origin.

1. Mr. Marshall’s Chispa The lure of gold and silver was, of course, a prime motivation for Spanish explorations in the New World. In the Americas, the ancient mining culture of Spain was fused with elements of Aztec metallurgy to form, what was for the period, an advanced mining technology. Mining is still the most important, as it is the oldest, industry in Mexico. Mexico had its “gold rush” at Zacatecas in 1548, three hundred years before the discovery of gold in California. Out of this experience, the Spaniards and Mexicans had learned a great deal about placer and quartz mining and had made of prospecting a fine art. Although the Spanish had sought gold and silver with the sword and spear, rather than with pick and shovel, they were the first to discover gold in

104  North from Mexico

California. On March 9, 1842, Francisco López, a Mexican herdsman, discovered gold in Santa Feliciana Canyon forty miles from Los Angeles. For a decade prior to James W. Marshall’s famous discovery, Mexicans had work­ ­ed  various “diggings” along the coast range between Los Angeles and Santa Cruz and had found gold in paying quantities. There is no mystery whatever about why they failed to make the big discoveries in California, for the gold of California, as Dr. Rodman W. Paul has pointed out, “was secreted in the interior of the province: precisely the region that the Spanish race had not colonized.”* In one of the most famous and popular scenes in California history, Marshall is supposed to have rushed into Fort Sutter with a nugget in his hand shouting, “Gold! Gold!” Actually, Marshall did not say that he had discovered gold; nor did he use the word “gold” or “nugget.” What he said was that he had discovered a chispa, which is Spanish for “bright speck” or “spangle.” That he should have used this term is some indication of how widely Spanish mining practices, and the Spanish mining vocabulary, had permeated California prior to 1848. The importance of Mexican metallurgy in the Southwest, however, rests on foundations more secure than such circumstantial details. When American mining engineers began to explore the Arizona-Sonora frontier after the Gadsden Purchase, they discovered a long history of prior mining operations in the region. Literally hundreds of mines had been worked by Mexicans in Sonora subsequent to the discovery of the famous bolas de plata at Arizonac in 1763. But most of these mines had been abandoned when Mexican troops were withdrawn from the frontier during the struggle for independence. In the face of incessant Apache raids, most of the equipment had been stored in the tunnels of the mines and the pits were closed in, with the thought of a later reopening once order had been restored. But the confused situation created by the Mexican-American War, the Gadsden Purchase, and the American Civil War had resulted in even more extensive raids by the Apaches. In their hatred of everything American, the Mexicans had mutilated boundary markers and continued to regard Arizona as legitimately a part of Sonora. Instead of reopening the mines, the gambussinos, or professional prospectors, then became freebooters who raided abandoned properties on the American side of the border and carried off equipment to Mexico which was used to smelt ores stolen from American properties. These Sonora gambussinos were among the first outsiders to receive word of the discovery of gold in California. In fact they were first attracted to California

* Reprinted from Rodman Wilson Paul—California Gold: The Beginning of Mining in the Far West. Cambridge, MA: Harvard University Press, 1947.

The Heritage of the Southwest   105

by the secularization of the missions. “They came flocking in,” wrote Hugo Reid, “to assist in the general destruction, lending a hand to kill cattle on shares, which practice, when at last prohibited by government orders, they continued on their private account.” By midsummer of 1848 some five thousand Sonorans had left for the goldfields. Small bands traveled to California from points as distant as Sinaloa, Chihuahua, and Durango. Starting from Tubac, they follo­­ wed the old De Anza trail to Yuma, crossed the Colorado, and then came to Los Angeles by way of San Gorgonio Pass. In California the Sonorans were treated with great contempt. Wearing cotton shirts, white pantaloons, sandals, and huge sombreros, they were known as calzonaires blancos or “white breeches.” In small groups of fifty and a hundred, they started out in the early spring from Mexico, worked in the southern mines in the summer, and returned south in the fall. Not infrequently, their families came along, riding the pack mules and burros. Between 1848 and 1850, ten thousand Sonorans passed through Los Angeles each spring and the processions continued for several years. Katherine M. Bell said that she had seen hundreds of Sonorans in Santa Barbara in 1849 on their way to the mines. Camped in caravans of ten, twenty, and thirty families on the outskirts of the town, they made merry with much singing and dancing to the music of “violins, guitars, and flutes.” The mining camps in the southern district—in Calaveras, Tuolumne, Mariposa, Stanislaus, and San Joaquin counties—were largely made up of Spanish-speaking people: Sonorans, Mexicans from Southern California, Chileans, and Peruvians. The town of Sonora, named after the Sonora miners, was the center of the southern district. As large as Stockton, it was described as being “far ahead of it for gold, gals, music, gambling, and spreeing.” A visitor of 1861 said of Hornitos, still another Mexican mining camp in the district, that “the town is certainly of Spanish origin and . . . there seems to be an omnipresent struggle between the Mexican and American element. . . . This rivalry is visible in everything. . . . Even the very signs seem to fight it out, or compromise. The stage house is the ‘Progresso Restaurant’; the bakery is a ‘panaderia’; the hotels invite both in Spanish and English. . . . In the plaza Brother Jonathan, however, has it pretty much all to himself, and manifest destiny will, undoubtedly, prevail in the end.” True to this prediction, Brother Jonathan had driven the Mexicans from the southern mines by the late sixties; but, before doing so, he had first appropriated their traditional mining lore. “The average American in 1848,” writes Dr. Paul, “was ‘handy’ with a considerable number of trades and occupations, but mining was not one of them.” There had been of course some placer mining in the Carolinas and Georgia prior to 1848 but, by and large, the Anglo-Americans were not a mining people. On the other hand, the Sonorans were experienced miners—the heirs to the great mining tradition of the Spanish people.1

106  North from Mexico

It was the Sonorans who first introduced the batea, a flat-bottomed pan or bowl with gently sloping sides, which was widely used in early creekbed or placer mining in California. The southern mines were so-called “dry diggings,” that is, there was a general shortage of water, except in the rainy season, to wash the dirt from the gold. Mexicans introduced the first successful extractive technique used in the dry diggings. “It was their custom,” writes Paul, “to dry the mixed gold and sand in the sunlight, or over the fire, then to separate the two by blowing upon the dry sands or by tossing them up in the air as one would wheat and chaff.” Known as the “dry-wash” method, this technique was widely used in the southern mines and had an important bearing on the rapid ex­ ploitation of mineral wealth in California. In 1850 the Alta California remarked that “American energy and assiduity, and Mexican skill and experience have together developed the riches of the Southern Placer.” Up to 1860, Mexicans were a majority in perhaps all of the counties in the southern mines; but, by the end of the decade, they had largely disappeared. Even before they had been driven from the district, however, the importance of their contribution to mining technology was widely recognized in California as witness this item from the Stockton Times: The Mexican is of the utmost service in the Southern mines. We ask those who have had actual experience in mining operations in this country, whether the American, with all his impatience of control, his impetuous temperament, his ambitious yearning, will ever be content to deny himself the pleasures of civilized life in the states, and for the sake of from four to eight dollars per day, be content to develop the resources of the dry diggings of the country.**

2. Comstock’s Mistake Most of the gold produced in California from 1848 to 1860 was obtained by working surface placers. As the miners worked up the streams and rivers, however, they soon discovered the ledges or deposits from which the gold had come. Once these deposits were located, the period of quartz mining began. If the Anglo-Americans were novices in placer mining, they knew literally nothing of quartz mining. “The first quartz miners in California,” wrote J. Ross Browne, “were Mexicans, who knew how gold-bearing rocks were reduced in their native country.”2 Thus the Morgan Hill mine in Calaveras County, one of

** On the role that Mexicans played in the development of gold and silver mining in the West, including the inauguration of pack-trains which supplied the mining camps, see The Gold Rushes by W. P. Murrell (1941), pp. 98, 102, 104, 137, 140, 156, 191.

The Heritage of the Southwest   107

the first quartz mines in California, was originally operated by Mexican labor. From this one mine alone over $2,000,000 in gold was taken in 1850. The first quartz ores in California were so rich in gold that the Mexicans treated them by hand mortars, but it was not long before they introduced the arrastre (more often spelled “arrastra”) or “Chili mill.” In building an arrastra, the Mexicans constructed a circular stone pavement in the center of which stood a post. To an arm extending out from the post a mule was hitched, or, in some cases, the arrastra was operated by a water wheel. The mule dragged a heavy piece of granite around the post which pulverized the quartz on the pavement. Once the quartz was pulverized, the gold or silver was then amalgamated by the use of quicksilver. Primitive as it was, the arrastra could be built on the spot, required no manufactured or imported parts, and was simple to operate. Quartz mining might have been retarded for many years in the West had it not been for the Mexicans’ familiarity with the arrastra, its use and construction. In the middle fifties, Ignacio Paredes, a miner from Alamos in Sonora, discovered some valuable ores in Nevada which he tried to work, first by the use of the batea, and, later, by the dry-wash method, but without success. Some years later, Comstock, prospecting in the same area, kept complaining about “base metals” and “blue stuff ” that made it difficult for him to isolate the gold. At this time, Comstock was convinced that he had discovered a gold mine. One day a Mexican miner happened along when Comstock and his partner were rocking gold with a batea. Noticing the heavy stone with bluish cast, he became very excited and started shouting, “Mucha plata! Mucha plata!” It was only then that Comstock realized that he had discovered one of the richest silver mines in the world. Dan De Quille, one of the most accurate of observers, reported that “the business of working silver-mines was then new to our people, and at first they depended much on what was told them by the Mexican silver-miners who flocked to the country.” These miners, he wrote, “were in great demand” and much of what was subsequently learned about quartz mining was based on their experience and knowledge. The arrastra made possible the early development of the Comstock Lode; at one time, some sixty arrastras were in operation at the mine. “The Spaniard of old and his Mexican successor,” wrote the Arizona historian, James H. McClintock, “were the best prospectors and the closest judges of ore ever known . . . the first American mining followed the pathways made by the Spanish.”3 J. Ross Browne, who prepared the first official report on the mineral resources of California, pointed out that “by far the larger portion of the work-people in the California mines are Mexicans who are found to be more adventurous than Cornishmen, and willing oftentimes to undertake jobs which the latter have abandoned.” Quite apart from many technical mining expressions in Spanish which passed into American mining law and the vocabulary of American miners,

108  North from Mexico

dozens of Spanish-Mexican mining terms found wide popular acceptance in the West. Bonanza or rich ore is one such expression; borrasco or barren rock is another. Such terms as placer, xacal (slack), and escoria (slag) are merely a few of many terms that might be cited. In The Big Bonanza, Dan De Quille devoted three pages to a glossary of Mexican mining terms in general use in the Washoe territory. Appropriation of these terms was a necessity since there were, of course, no equivalent expressions in Anglo-American speech.

3. The Vermilion Cave In 1557 Bartolomé de Medina, a miner at Pachuca, Mexico, revolutionized mining technique by the invention of the patio process for separating silver from ore by the use of quicksilver. Thereafter quicksilver became an essential material in quartz mining. Immediately upon the discovery of the patio process, the Spanish reserved the quicksilver monopoly for the famous Almaden mine in Spain by prohibiting quicksilver mining in the Americas. By mono­­ polizing the supply of quicksilver, they hoped to control all mining in the Americas. Most of the quicksilver used in Mexico, Central, and South America, in fact, was imported from the Almaden mine. “This fiat,” writes Gruening, “destroyed a potential industry and greatly hampered an existing one.” In the mountains about twelve miles from San Jose, California, was a cave which contained a bright vermilion clay. For many years, Indians had visited the cave, dabbing their faces with its clay. As early as 1824 two Mexicans, Antonio Sunol and Louis Chaboya, had tried to extract silver from the clay but had failed to do so. In 1845 the Mexican government sent Captain Andrés Castillero, a young cavalry officer, on a military mission to Fort Sutter. When Castillero, who had been trained in metallurgy, heard about the famous vermi­ lion cave from a priest at the San Jose mission, his interest was immediately aroused. Putting some of the cinnabar clay in his gun, he found that drops of quicksilver gathered in the gun-barrel after the gun was fired. On returning to Mexico City he filed a claim on the property but the intervention of the Mexican-American War forced him to assign the claim to a British company. This mine—the famous New Almaden—was the first important quicksilver mine to be discovered in the Western Hemisphere. It was the discovery of the New Almaden that unlocked the gold and silver resources of California and the West. The timing of the discovery and the location of the mine were nothing short of providential. Small quantities of quicksilver were produced late in 1848 and, two years later, the mine was in heavy production. The New Almaden, moreover, was conveniently located with reference to the California mining districts. Prior to its discovery, quicksilver, a monopoly product, had sold on the world market for $99.45 per flask; but once the New Almaden was in production the price fell to $47.83. J. Ross Browne

The Heritage of the Southwest   109

once said that the discovery of gold and silver in California would have meant very little had it not been for the simultaneous discovery of the New Almaden which, as late as 1918, was still producing one million flasks of quicksilver a year. Not only was the New Almaden discovered by a Mexican, but it was deve­ loped by Mexican labor. J. Ross Browne said that five-eighths of the 1,973 miners employed at the property in 1865 were Mexicans.“The laborers,” said another visitor, “are all Mexicans and have generally served a sort of apprenticeship in the silver mines of Spanish-America.”4 Living in a town on the hill near the mine, the Mexicans were divided into two categories: the actual miners or barreteros; and the ore-carriers or tanateros. Starting from the pit of the mine, the ore-carriers would fill a large sack or pannier made of hide with two hundred pounds of ore and then ascend the escalera or ladder-like circular path to the surface. Open at the top, the pannier was flung over the shoulder and supported by a strap passing over the shoulders and around the forehead. Ore-carriers made from twenty to thirty trips a day up the escalera for all the ore was carried to the surface by hand. The escalera was narrow, slippery, and lighted only by a few flickering torches. Visitors told of seeing the tanateros, dressed in pantaloons rolled tight above the knees, and calico shirts, hurrying up the escalera with “straining nerves and quivering muscles.” In the patio near the mine, where the ores were reduced, Americans were employed at wages of from $5 to $7 a day; but the miners and ore-carriers received a wage of from $2 to $3 a day. “In the early years,” writes Dr. Paul, “the Mexicans tended to form a special element in the labor supply, paid at a lower rate than Americans and Europeans.” Living in straw-thatched huts on the hill, the Mexicans were a carefree lot, “spending their money on the visiting señoritas from San Jose,” and celebrating their días de fiesta by sending for girls, guitarristas, and wine. It is also interesting to note that the British and American owners of this fabulously rich mine sent their engineers to Spain to study the processes used in quicksilver production, and later entered into a cartel with the owners of the Almaden to control the world market.

4. Anglo-Saxon Law and Order Prior to the discovery of gold in California there had been so little American mining that the Anglo-Saxon common law had virtually no mining-law precedents. Precedents were never more badly needed than in California, for most of the miners were trespassers on the public domain. With no mining law on the statute books and with no precedents for guidance, the federal government was powerless to bring order out of the chaos that prevailed. Faced with this situation, the miners were forced to develop their own codes and rules which were the only “law” on the subject of mines and mining in effect in the United

110  North from Mexico

States from 1848 to 1866. Later these codes, which form the basis of the present-day “law of mines,” were carried by California miners throughout the West. Wherever these miners traveled, they also carried the knowledge and experience which they had acquired in California. In the history books, the famous miners’ codes are invariably cited as another illustration of “the extraordinary capacity of the Anglo-American for selfgovernment.”5 Bearing in mind that it was the discovery of gold in California that gave birth to a distinctive American mining law—there having been no general mining law in force in the United States prior to 1848—just how was it that these inexperienced Anglo-American miners were able to develop, in such a brief period, a comprehensive system of mining rules and regulations and a law of mines? The Spanish-speaking miners of California had been trained, of course, under the mining ordinances of Spanish America which represented a complete body of mining law tested by experience. “The miners of California,” writes the legal historian, Halleck, “generally adopted, as being the best suited to their peculiar wants, the main principles of the mining laws of Spain and Mexico, by which the right of property in mines is made to depend upon discovery and development,” principles that are still cited as being preeminently Anglo-Saxon in origin. Yale, the outstanding authority on the American law of mines, writes that “most of the rules and customs constituting the codes are easily recognizable by those familiar with the Mexican ordinances. . . . In the earlier days of the placer diggings in California, the large influx of miners from the western coast of Mexico and from South America, necessarily dictated the system of work to the Americans, who were almost entirely inexperienced with this branch of industry. . . . The SpanishAmerican system which had grown up under the practical workings of the mining ordinances of New Spain, was the foundation of the rules and customs adopted.” Although this information has been common knowledge among lawyers for fifty years, there are California historians who still write eloquent chapters in praise of the Anglo-American miner’s “capacity for self-government.” Under the Spanish law, possession of minerals in the subsoil was reserved to the crown. From 1836 to 1883, the State of Texas received five percent of the gross receipts from all mineral concessions, which was used to establish a system of public schools. For this happy prevision, the Texans are indebted to the Spanish law of regalia. When Texas adopted the Anglo-Saxon common law in 1840, the only Spanish statute specifically retained was this doctrine of mine­ral rights.

5. Apaches and Copper In the year 1800 a Spanish colonel, José Carrasco, guided by an Apache Indian discovered the famous Santa Rita silver and copper mine in western New

The Heritage of the Southwest   111

Mexico. A native of Rio Tinto, Carrasco quickly identified the ores of the Santa Rita, for he remembered the appearance of copper ores from his youth. While the nearest smelter was four hundred miles from the mine, poverty had decreed that the peon population of Mexico should have a copper coinage. For many years, Santa Rita ores were carried by pack-trains to smelters in Mexico and sold for sixty-five cents a pound. As early as 1804, the Santa Rita was being operated on a fairly large scale with over six hundred employees living in the community which had grown up about the property. Incidentally, one of the watchtowers built by the Spanish still stands at the mine. The Santa Rita is, perhaps, the most famous mine in Western America for it was here that the techniques of copper-mining were first developed in the Southwest. The Heintzelman mine, thirty miles from Tubac, with its attractive farms and orchards, had also been worked at an early date. Some eight hundred Mexican miners were employed in 1859 at the mine which was then producing $100,000 in silver a year. Along with the Santa Rita and many other mines, the Heintzelman property was abandoned during the Apache raids. At one time, the Arizona Mining Company at Tubac found itself besieged on one side by the Apaches and on the other by a band of enraged Sonorans. Engine boilers weighing six thousand pounds, which had been laboriously freighted in from Lavaca, Texas, a distance of 1,200 miles, were abandoned in 1861 when the owners were forced to move out. Mexican miners from Sonora were employed, from the earliest date, at both the Santa Rita and Heintzelman mines. The prevailing wage of from fifty cents to a dollar a day was paid, according to Mowry, “in large part in merchandise sold at large profits.” Since bullion was too clumsy to handle, wages were paid in company-issued boletas or paper bills with the denominations indicated by the figures of animals—pigs, roosters, cows, and horses. “The only difference between peonage and Negro slavery,” wrote Will H. Robinson, “was that a peon miner could not be sold from one master to another.”6 Visiting the reopened Santa Rita on payday, J. Ross Browne reported that “under every tree sits a group of thriftless vagabonds, conspicuous for their dirty skins and many-colored sarapes, shuffling the inevitable pack of cards or casting their fortunes of greasy ‘hobos’ upon capricious hazards of fortune. The earnings of the month are soon disposed of. The women and children are left dependent upon new advances from the store-houses; the workingmen are stupefied by mescal and many nights of debauch, and when all is over, the fandangos at an end, and the monte tables packed up, every miner is bankrupt.” A curious and accidental by-product of the final “pacification” of the Apaches in Arizona was the discovery of important new copper deposits by cavalry officers. The famous Bisbee mines were discovered around 1875 by cavalrymen in hot pursuit of Apaches. The development of these new properties was largely based upon the early experimental techniques which had been

112  North from Mexico

evolved at the Santa Rita mine and at various mines in Sonora. When Henry Lesinsky began to develop the rich copper deposits at Clifton in 1872, one reads that he went to Juárez to employ Mexican laborers who were “considered very skillful smelter men.” These miners constructed the first furnaces to smelt copper ores in Arizona which were “of the Mexican type, built of adobe,” and fired by charcoal made from mesquite.7 The adobe furnace had a capacity of about two tons of ore per day and its fire was sustained by hand bellows. Mexican miners, using burros, packed the crude ores from the mountains to the smelters. The smelted ore was then packed by ox and mule teams, operated by Mexicans, to Kansas City. Don Antonio, foreman of the Clifton mine, rode throughout Sonora recruiting Mexicans to work in the copper mines. Western mining developed, of course, by a series of “waves”: first gold, then silver, and finally copper. At first only the high-grade copper ores—those that ranged from five to twenty percent copper—were exploited; but a new process was perfected around 1892 for smelting the low-grade ores (the disseminated or porphyry ores). The smelting of these ores involved an enormous capital outlay and brought about a rapid consolidation in ownership. Simultaneously new processes were developed for extracting ores in the underground mines. One of these techniques was the “cave-in” system whereby a whole section of earth would be caved in by a single blast. This system greatly increased the amount of ore that could be produced in a day, but the system was—and still is—extremely dangerous. Experienced miners often refused to work in underground mines where it was used; but Mexican immigrants, excluded from the skilled miner category, were compelled to work in these mines. In this way a rift developed, which has not yet been healed, between Mexican and nonMexican labor in the copper mines. Between 1858 and 1940 the Arizona mines produced three billion dollars’ worth of metal. Copper production increased from 800,000 pounds in 1874 to 830,628,411 pounds in 1929. It was the vast expansion in the electrical industry which enabled copper, “the red metal,” to dethrone its “white rival,” silver. One might say, therefore, that Mexican miners in the copper mines of Arizona, Utah, and Nevada, have played an important role in making possible the illumination of America by electricity. The census of 1930 listed 16,668 Mexicans engaged in the extraction of minerals: 3,880 as “coal-miners,” principally in Colorado and New Mexico; and 12,623 “other operators,” mostly in the copper mines of the Southwest.

6. Homage to the Churro Although Coronado brought the first sheep to the Southwest, the herds that were to constitute the basis of the pastoral economy of New Mexico came north in the famous entrada of Juan de Oñate in 1598. The raising of sheep is preeminently

The Heritage of the Southwest   113

a frontier enterprise. Sheep helped to make the Spanish explorations possible for they were the mobile, marching food supply of the conquistadores. According to Messrs. Towne and Wentworth,8 sheep were an indispensable item in equipping every Spanish expedition to the north. From the founding of New Mexico until the Civil War, sheep fed, clothed, and supported the colonists. The only important source of cash income in the colony, sheep also served as a kind of currency. “Sheep,” writes Winifred Kupper,9 “were the real conquerors of the Southwest.” In 1598 Spain had one of the oldest sheep cultures in the Western world. Its breeds were principally of two types: the beautiful, aristocratic merinos with their fine wool; and the ugly “scrubs” or churros long relegated to the periphery of the Spanish sheep culture. It was the scrub or churro, however, that the Spanish brought to the Southwest: a small, lean, ugly sheep whose wool was coarse and light in weight, seldom averaging more than a pound or a pound and a half at a shearing. But during a long period of adaptation to the semiarid environment of Spain it had learned to hunt food and shelter, to make long marches, to survive in all sorts of weather, and to protect its lambs from wild animals. Its very “scrubbiness” made it ideally adapted to conditions in the Southwest. The Spanish also brought to New Mexico their traditional sheep culture. Without this knowledge, based on six hundred years’ experience under somewhat similar conditions, the value of the churro might have been negligible. There is no doubt whatever that “sheep husbandry in the United States,” to quote Wentworth and Towne, “owes more to Spain than to any other nation on earth.” Long prior to 1598, the Spanish had developed an extensive lore about sheep and had evolved elaborate institutions to protect and to further the sheep industry. In Spain sheep were marched from the lowlands to the highlands and back from the highlands to the lowlands. The privilege of marching sheep in this manner had given rise to the trashumante system under which the rights and privileges of sheepmen were minutely regulated, defined, and safeguarded. To make this system function, an ancient organization of sheepmen had been effected known as the “Honorable Assembly of the Mesta,” which has its almost precise counterpart today in the various “sheepmen’s associations” in the Southwest. In Spain the various “sheep walks” were carefully laid out and defined; and what sheepmen could and could not do, on these marches, was also fixed by custom and ordinance. A somewhat similar system is in use today in the Southwest. In short, Anglo-American sheepmen in the Southwest took over and adapted an already functioning and time-tested pattern of sheep-raising. About all they did was to enlarge the grazing areas by bringing the nomadic Indians under military control (a victory largely made possible by the introduction of the Colt revolver) and improve the breed of sheep. Apart from these contributions, they simply took over the customs, practices, institutions, personnel, and

114  North from Mexico

organization of an existing industry. The system of Spanish and Mexican land grants, based on larger units of land than were to be found in a non-arid environment, made possible the expansion of an industry which involved an extensive use of land resources. Similarly the practice of assigning fixed grazing rights to particular owners was Spanish in origin. Even the breeding of a heavy wool-bearing sheep came about as a result of crossing two Spanish types. For about 1820 some fine Spanish merino sheep were first brought to the eastern seaboard. These sheep reached the Southwest about 1876, with the general westward movement, and were then crossed with the churro to produce a new type ideally adjusted to the environment. Under the Spanish system, sheep-raising was based upon a traditional social structure and a well-defined division of labor. At the base of the pyramid was the pastor or shepherd who was usually assigned a flock of about two thousand sheep. Over each two or three pastores was a vaquero or mounted rider. Supervising the vaqueros was a caporal or range boss and over the caporal was the major-domo or superintendent. Ultimate authority rested, of course, in the owner or patrón. In general, this system of organization was taken over in toto by the Anglo-Americans and still prevails on the large sheep ranches of the Southwest. Between the eastern seaboard and the boundaries of the Southwest, sheep-raising was, and still is, an avocation or sideline business. Once the center of the industry had shifted to the Southwest, which was around 1870, sheep-raising became a specialized business, conducted on a large scale, by men whose sole vocation was sheep-raising. This was the Spanish system and its excellent adaptation to conditions in the Southwest is shown by the phenomenal increase in the wool clip: from 32,000 pounds in 1850, to 493,000 pounds in 1860, to 4,000,000 pounds in 1880. New Mexico was “the ovine nursery of the nation” whose herds provided the foundation stock for the entire West. The Californios had never looked with particular favor on sheep-raising; in fact there were only about seventeen thousand sheep in the province in 1850. But, with the discovery of gold, large herds were driven overland from New Mexico to the mines. It is estimated that, bet­ ween 1850 and 1860, more than five hundred thousand sheep were driven from New Mexico to California. Here, again, the marching qualities of the churro were of considerable importance. From these drives came the herds that were soon grazing in the foothills and valleys of California. It was also in California that the churro was crossed with the merino to produce the present range stock of the Western states. From California large herds were then driven eastward to the Rocky Mountain states and the new and improved breeds made their way back to New Mexico. During the seventies and eighties, large herds were driven eastward every season, grazing as they marched, to the terminal points on the rail lines from which they were then shipped to Middle Western markets. In the process of making these “drives,” such states as

The Heritage of the Southwest   115

Idaho, Utah, Colorado, Wyoming, Nevada, Arizona, and Montana were stocked with sheep. Once this development had taken place, American wool production soared from five million pounds in 1862 to twenty-two million pounds in 1880. Increased wool production in the West meant, of course, increased factory employment in the East. The development of the sheep industry also stimulated another Western industry in which Mexicans have played a key role. For with the establishment of the first Western sugar-beet factories, the modern era of lamb-feeding came into its own. One reason for the rapid growth of the sugar-beet industry in the West was the fact that sheep could be fed and fattened on the by-products of sugar-beet production. Thus one industry neatly supplemented the other. It was the Spaniards, of course, who taught the Indians of the Southwest to weave with wool. From 1800 to 1850 many Navajo women were employed in weaving in New Mexico and these women carried back to the tribes the skills which they had learned. From the Spaniards and Mexicans, also, the Navajo inherited the churro sheep. That they are today largely a pastoral people is to be traced to this early cultural borrowing. In the development of the art of weaving, however, the Spanish borrowed designs and dyes from the Indians. According to Ruth Laughlin, the Spanish brought only two dyes to New Mexico; the rest were all developed from native dyes used by the Indians. Blankets were long an important item in the barter economy of New Mexico. When trade was opened with Los Angeles by way of the Spanish Trail, one reads that New Mexico blankets were exchanged on the Coast for California horses. Blankets were also a principal item in the barter-markets of Taos and Chihuahua.

7. Los Pastores Throughout the Southwest, the term “Mexican sheepherder” is proverbial. In the folklore of the region the solitary, superstitious, patient Mexican sheepherder is supposed to be as witless and moronic as the sheep he herds. But Mexican herdsmen are the carriers of a great tradition and it has been their skill and knowledge which has sustained the sheep industry in the West. Full of incredible lore, they can read the signs of changing weather at a glance; they know the habits of predatory animals; their knowledge of range vegetation is unrivaled; and there is little about the care of sheep that is unknown to them. In the early journals, one reads of how they trained sheep dogs by suckling pups on ewes so that the dog would learn to follow the sheep while they grazed, and return them at night to the corral. Above all, these pastores know how to graze a flock, guiding their movements without driving them, so that the sheep travel slowly and graze contentedly as they travel.

116  North from Mexico

Living on coarse meal, goat’s milk, kid’s flesh, and peppers, New Mexico pastores have tramped over most of the West in all sorts of weather and under the most difficult conditions. Until the 1890s, sheepherding was a hazardous occupation in New Mexico. One reads of Apache raids that netted five thousand sheep and of a raid in 1850 in which forty-seven thousand sheep were stolen. The most famous New Mexico Indian fighters were sheepmen. Colonel Manuel Chavez, and his pastores, fought the Indians for fifty years. Nor were Indians the only hazard. In California the herdsmen lived in mortal terror of bears and slept at night on raised platforms, called tepestras, which were built on poles eight, ten, and twelve feet above the ground. The isolation of the shepherds was even greater than that of the other colonists in New Mexico. Something of the present-day brooding, introspective quality of the Spanish-speaking people of New Mexico can probably be traced to this experience. The notion that sheepherders are a weird lot, often driven crazy by loneliness, may be unfounded; but they are certainly the most taciturn of men. To guard against the hazards of loneliness a state law in New Mexico requires that sheepherders must be employed in pairs. Carrying their giant jews’-harps, called bijuelas, the pastores sang folk songs on the ranges of New Mexico which were of great antiquity when Columbus discovered America. No one knows the precise origin of the ancient folk play, Los Pastores, which has been produced in New Mexico for as long as the colony has existed. Some of the most beautiful New Mexico folk songs are, of course, the songs of the pastores. Lieutenant J. H. Simpson, who traveled west in 1849, reported that the first New Mexican he sighted was “a swarthy, copper-colored young Mexican, of eighteen or twenty years of age, most miserably clad, driving the sheep before him. The morning air was keen and cold, and as he, with brimless straw hat on, a forlorn blanket about his shoulders, and pantaloons which were only an apology for such, hugged his only wrapper, his steps slow and measured, I thought he looked the very personification of patience and resignation.” Whether it was the scrawny character of the sheep or the appearance of the herdsman, somehow the combination of being both a sheepherder and a Mexican came to be synonymous, to most Anglo-Americans, with the lowest possible status. From the 1860s, bands of New Mexican sheep-shearers, each with its capitán, made the great circle of the shearing pens from Texas to California to the Northwest and throughout the Rocky Mountain states. “I remember the Mexican sheep shearers galloping up,” wrote Sarah E. Blanchard of her childhood on a ranch near Santa Paula, “and it was a time of thrilling excitement. Usually an old woman accompanied them to make tortillas and to provide them with Mexican delicacies. They were paid by ticket, so much for each fleece, and at night they gambled with these around the camp fires.”

The Heritage of the Southwest   117

Not infrequently a few Chinese sheep-shearers accompanied these bands. “The shearers would come in,” wrote Sarah Bixby Smith, “a gay band of Mexicans on prancing horses, decked with wonderful silver-trimmed bridles made of rawhide or braided horsehair, and saddles with high horns, sweeping stirrups, and wide expanse of beautiful tooled leather. The men themselves were dressed in black broadcloth, ruffled white shirts, high-heeled boots, and high-crowned, wide sombreros which were trimmed with silver-braided bands, and held securely in place by a cord under the nose. They would come in, fifty or sixty strong, stake out their caballos, put away their finery, and appear in brown overalls, red bandanas on their heads, and live and work on the ranch [in Southern California] for more than a month, so many were the sheep to be sheared.” Shearers were the migratory aristocrats of the industry. They were never herdsmen, for the shearing of sheep was an exclusive vocation. Paid a wage of from five to eight cents a fleece, New Mexicans monopolized sheep-shearing until around 1890 or 1900 when the first power-driven shearing machines were introduced in California. “The New Mexicans,” wrote Twitchell, “were essentially a pastoral people.” Lummis once said that sheep were “the one available utilization of New Mexico” where society was divided into two classes: those who owned sheep and those who tended sheep. Today it is said in New Mexico that those who own sheep are Spanish-Americans; those who herd them are Mexicans. From the earliest date, the great herds of New Mexico were owned by a handful of ricos. According to the New Mexico Guide (WPA), a few large operators still own seventy-five percent of the sheep. The Spanish governor, Bartolomé Baca, once owned two million sheep and employed 2,700 men. “El Guero” Chávez, the first governor under Mexican rule, owned a million head and Don José Leandro Perea of Bernalillo had herds of more than 250,000. In 1880 threefourths of the sheep of New Mexico were owned by about twenty families, sixteen of whom were families native to New Mexico. Sylvestre Mirabal, who owned 250,000 acres of grazing lands at the time of his death some years ago, was descended from a family that had raised sheep in New Mexico since 1600. The dependence of pastor on patrón was complete and absolute. The patrón protected the pastor against the Indians and before the law. In the 1890s New Mexico pastores were paid a wage of from $5 to $8 per month, board included; and as late as 1940 the wage was only $35 a month. Many of the pastores were bound to their patrones by debts inherited from their fathers; even after peonage was abolished, the partido system by which herds were “farmed out” on shares, functioned as a thinly disguised form of peonage. “The social effects of a system of economy,” writes John Russell, “wherein four-fifths of the white male population were employees of a handful of landlords have left their stamp on present-day New Mexico.” Long after 1846, all the patrón needed to do was to say the word and his pastores would vote, as a group, for any

118  North from Mexico

candidate he recommended. “The most paternalistic form of government in the world,” writes Miss Kupper, “is a flock with a sheepherder as dictator” and the relationship between the herder and the flock is essentially that between the patrón and his pastores. For several hundred years, thousands of New Mexico sheep were driven to points as far removed as Veracruz, Guadalajara, and Mexico City, principally for sale at the mines. After Mexico won its independence, the annual sheep drives to Chihuahua became immensely profitable and between 1839 and 1850 about two hundred thousand sheep were driven south every year. For these drives, the ricos would purchase the small herds of the paisanos who were, of course, unable to drive their sheep to market. Profits of from three hundred to four hundred percent were occasionally made on these annual drives to Mexico. With the discovery of gold in California, sheep sold for $16 a head and the drives to the West Coast took the place of the drives to Mexico. Little of this bonanza wealth ever found its way into the pockets of the pastores.

8. From Gregorio de Villalobos The cattle industry began in the mesquitals along the Rio Bravo. j. frank dobie

By a curious cultural transmutation, Anglo-Americans have long claimed credit for the origin and development of the cattle industry. No folk hero in American life has enjoyed anything like the popularity of the American cowboy. Each week millions of Americans see “Western” films and their sons and daughters will probably line up at the box-office years hence to see cowboys ride, rope, and shoot on the screen. Yet with the exception of the capital provided to expand the industry, there seems to have been nothing the American rancher or cowboy contributed to the development of cattle-raising in the Southwest. One Gregorio de Villalobos is supposed to have shipped the first cattle to the New World. From this initial shipment to the West Indies came the stock later used to establish the great herds in Mexico and from these herds, in turn, came the cattle that Coronado drove to the Southwest. Like the lowly churros, the cattle that the Spanish brought to the New World were not much to look at. Light-bodied, long-legged, thin, with elongated heads and muzzles, their widespreading horns often measured five feet from tip to tip. “The general carriage of a Spanish cow,” wrote one early-day historian, “is like that of a wild animal: she is quick, uneasy, restless, frequently on the lookout for danger, snuffing the air, moving with a high and elastic trot, and excited at the sight of a man, particularly if afoot, when she will often attack him.” Such was the parent stock of the American range-cattle industry. Dating from the latter part of the eighteenth century, the cattle industry had its real beginnings in California and Texas.

The Heritage of the Southwest   119

When the San Carlos anchored in San Diego Bay on April 30, 1769, as part of the Serra expedition, some six or seven head of cattle were taken ashore: supposedly the first cattle to appear in California. Somewhat later, small herds were driven overland to California by Rivera and De Anza. So rapidly did cattle multiply in the province that the mission fathers and rancheros could count a million head by the end of the century. In fact cattle came to be regarded as a major nuisance in California. People afoot were forever dodging behind trees or jumping into ditches to escape from the wild charges of Spanish steers, regarded as more dangerous than grizzly bears. Anyone could start a herd in California, for there was no limit to the available pasture. Beef in California, like mutton in New Mexico, became a principal staple in the diet; and the hides, worked up into rawhide, were used for manifold purposes. At the Mission San Jose, a hundred cattle were butchered every Saturday. Cattle horns topped the fences around the wheat fields and the hides of cattle, drying in the sun, were to be seen at all seasons of the year. With the first shipments to South American ports (1810) and the opening of the hide-and-tallow trade with Boston (1822), markets were finally found for the great surplus of cattle in California. The clipper ships, described by Dana as floating department stores, brought merchandise to exchange for the hides and tallow and a flourishing trade developed. Between 1800 and 1848 over five million hides were exported from California. It so happened, also, that the opening of the hide-and-tallow trade coincided with the beginning of the Mexican regime and the secularization of the missions. Under the impetus of this trade, the mission estates were carved up into great ranchos and stocked with cattle often plundered from the missions. By 1860 some three million were grazing on the great unfenced pastures of California. The cattle industry in California, however, had reached its zenith and had begun to decline at about the time that Texas became the cattle nursery of the nation. Large herds were to be found in Texas at an early date but incessant Indian raids kept the industry from developing as rapidly as it did in California. In the chaotic period which followed the Texas rebellion, thousands of cattle roamed wild in the brush country between the Nueces and the Rio Grande. From an estimate of 100,000 in 1830, the number of cattle in the state increased to 382,733 in 1846. Four-fifths or more of this total was made up of so-called “Spanish cattle,” for about the only cattle the Anglo-American settlers brought to Texas were a few milch cows. From these wild herds, the cimarrones of the brush country, came the cattle later driven to the rail terminal points in Kansas for shipment to the stockyards of Kansas City and Chicago and which were used, still later, to stock the ranges of Colorado, Wyoming, and Montana. While the Anglo-Americans may claim credit for the remarkable expansion of the cattle industry and for the conditions which made this expansion possible, the industry is indisputably Spanish in origin.

120  North from Mexico

9. “Ten Gallon Hats” It was the coming together, as J. Frank Dobie puts it, not in blood but in place and occupation, of the Anglo-American, the Spanish owner, and the Mexican vaquero that produced the Texas cowboy—“a blend, a type, new to the world.” The word “cowboy” was unknown prior to 1836.“Cowboy” is the literal American equivalent of vaquero which is derived, of course, from vaca or “cow.” Everything that served to characterize the American cowboy as a type was taken over from the Mexican vaquero: utensils and language, methods and equipment. The Spanish brought the horned saddle, to be distinguished from the English “muley” saddle, to the Southwest. Long before they came to the borderlands, the Spanish had taken this saddle over from the Moors. Along the Rio Grande, Mexican vaqueros made saddle stocks from the soft wood of the giant prickly pear and used a flat-topped silver horn as “big around as a soup plate.” The saddle of the cowboy was merely an adaptation of this Spanish saddle. From the vaquero, the American cowboy took over, and adapted in his own way, the Spanish horned saddle, bridle, bit, and spur. From the vaquero, also, he got his lasso or lariat, cinch, halter, mecate or horsehair rope, “chaps” or chaparejos, “taps” or stirrup tips (tapaderas), the chin-strap for his hat (barboquejo), the feedbag for his horse (morral), and his rope halter or bosal. Even his famous “ten gallon hat” comes from a mistranslation of a phrase in a Spanish-Mexican corrido “su sombrero galoneado” which referred to a festooned or “gallooned” sombrero.10 “The very language of the range,” writes Mr. Dobie, “is Spanish.” Such terms as bronco (from mesteno), mesquite, chaparral, reata, grama, huisache, retama, remuda, cavyard from the Spanish cabal/ada, lariat from la reata, outfit or corrida, lasso from lazo, buckaroo, burro, cinchas, latigo, quirt (from cuerda), stampede (from estampida), hondo or hondoo for loop, calaboose (from calabozo), vamoose, mesa, canyon, barranca (bluff), rodeo, corral, sombrero, loco, all these, and many more, are Spanish-Mexican in origin. From the Spanish-American War, the cowboys of the Southwest brought back the word “hoosegow,” or lockup from the Spanish juzgado. In the borderlands, a ranchero (ranch) was an estate where cattle were raised; while an estate where crops were raised was a hacienda. Among the cowboys with whom I consorted as a youngster in Colorado, nothing was resented more keenly than the suggestion that they worked on a “farm.” The words “farm” and “farmer” were anathema; they were “cowboys” who worked on a “ranch.” The Mexican ranchero loved and understood horses and often had more horses about than he had cattle. Some of the ranches in Texas had as many as a thousand head of horses and these herds, and the wild horses of the range, made the horse market in San Antonio the greatest of its kind in the world. A manada was a unit of horses on the range: one stallion for each twenty-five mares. The bell-mare in the herd was the remudera. The cowboy expression

The Heritage of the Southwest   121

“wind-broken” is from the Spanish. The technique of horse-breaking as practiced by the American cowboy was based directly on the technique of the domador or professional Mexican horsebreaker.11 No language in the world is so rich in hairsplitting terms to distinguish the exact color markings and characteristics of a horse as the “sagebrush” Spanish of the Southwest. The following and many similar terms are really Southwest slang, with the Spanish words being given, by long local usage, a meaning of their own: Alazán tostado, a chestnut sorrel; andaluz, a yellow horse with blond mane and tail; azulero, a dark blue roan; barroso, a smudgy dun-colored horse; canelo, a blue-red roan; cebruno, a dark brown; grullo, a bluish gray; moro, almost blue; tordillo, iron-gray; palomino, a cream-colored horse; roano or ruano, shortened to “roan,” a dapple-colored horse. A “pinto” is, of course, a painted, a piebald horse. An estrello is a horse with a star on its forehead; a cuatralbo is a horse with four white feet; a potro is an unbroken horse.12 The American cowboy’s elaborate lore about the rope and roping techniques was acquired directly from the Mexican vaquero. Roping by the forefeet was based on the mangana technique; while to rope by the hind feet or “to peal” was a feat also learned from the Mexicans. The Mexican expression dale vuelta, meaning to twist a rope about the horn of the saddle, became first “dolly welter” and, later, simply “dolly” on the Anglo-American tongue. The Mexican was an artist with knife and rope both of which he used as weapons. It was only when the Texans got the Colt revolver, about 1838, according to Dr. Webb, that they “became a terror to the Mexicans and all enemies.” At a rodeo in Tucson on May 31, 1939, one José Romero, a Mexican vaquero, roped a fullgrown golden eagle from horseback. It is also quite probable that the famous American cowboy songs are based on the corridos of the vaquero. The great King and Kennedy ranches in Texas still rely upon Mexican vaqueros. The semifeudal organization of these ranches, in fact, is directly patterned after the organization of the large ranchería. The managers are, nowadays, Anglo-Americans; but the “hands” are Mexican—the vaqueros, the caporales or foremen, the pasteros or pasture tenders, and the jinetes or horsebreakers. “The proudest men I ever saw,” is the way George Sessions Perry describes them. They still love goat meat or cabra and brew the tea of the ceniza or sage. Spanishspeaking to a man, they sing the old corridos about the bordertowns, the great cattle drives, the stampedes, and the song of the caballo fragado or broken-down horse. Voting pretty largely as their “boss” tells them, the hands on the King ranch still refer to themselves as Kineños.

10. Cortez Had a Brand Long prior to the appearance of the Anglo-American stockman, the Mexicans had a fully developed system of brands and brand registrations. Their brands

122  North from Mexico

were of three types: the fierro or iron; the señal or earmark; and the venta or sale brand. Like much of the Spanish lore about cattle and horses, brands came to Spain with the Moors. Mexican brands are of great antiquity, some of them being based on the Moorish rúbricas—signs used first as a signet or signature and later added as a flourish when the writers learned to spell their names. Many Mexican brands were also copied from Indian pictographs and from symbols of the sun, the moon, and the stars. The brand used by Cortez—three Christian crosses—is said to have been the first brand used in the Americas. “There were brand books in Spain,” writes Dane Coolidge, “hundreds of years ago.” The Spanish had a system of registering brands which was in use in Mexico as early as 1545. When a horse or cow was sold, the old brand was “vented”—stamped on the shoulder—as a bill of sale; and the new brand was burned below this marking. The American law of brands and the various brand registration systems in use today are based directly on these ancient Spanish-Mexican usages. Once a year in California the rancheros held a general roundup or rodeo which was presided over by one or more Jueces del Campo or Judges of the Plain. These judges settled all disputes over ownership and saw to it that calves were branded with the right brand. “In West Texas, New Mexico, Colorado, and northward,” wrote Charles Howard Shinn, “wherever great cattle ranges are found today, the stockmen, in their round-ups, still follow the ancient Spanish plan; not knowing it is a heritage from a race they despise, they choose ‘cattle judges’ to settle disputes and uphold their decisions as final.” The well-organized and powerful cattlemen’s associations of the West today are based upon the Spanish institution of the alcaldes de la mesta. When Austin drew up his code for the first colony in Texas, the Mexican officials added only two articles: one governing the registration of brands and the other having to do with cimarrones or wild cattle. As with the sheep industry, all the AngloAmericans did was to provide capital for expansion, drive the Indians from the range, and improve the breed of cattle. The same can also be said of horses, goats, and mules. The mule industry of Missouri—once a thriving industry— was Spanish-Mexican in origin. Spanish range laws had an influence even in the Southeastern coastal part of the United States. Many Southern fence laws, range laws, and toll systems in use today are said to have grown out of customs and practices which the Spanish brought from the West Indies.

11. A Drop of Water Just as Anglo-American settlers knew little about mining, sheep, or cattle, so they were almost wholly unfamiliar with irrigated farming. In fact there was little in Anglo-Saxon law or institutions that was applicable in the semiarid environment of the Southwest. The Anglo-Saxon common law, with its

The Heritage of the Southwest   123

doctrine of riparian rights, had been formed in Great Britain where water was not a problem. On the other hand, the Spanish civil law was based on a recognition of the shortage of water and the need for irrigation. The Moors had brought many of their irrigation practices and water-saving institutions to Spain in the tenth century. The similarity in environment made it possible for the Spaniards to carry over into the borderlands practices and institutions which had arisen out of the need for irrigation on the Spanish Peninsula. The oldest irrigation systems now in use in the United States are to be found in the Rio Grande Valley in New Mexico. Here the Spanish were irrigating the bottomlands around Las Cruces when the Mayflower arrived at Plymouth. The Pueblo Indians were irrigating between fifteen thousand and twenty-five thousand acres in the valley when the Spaniards first appeared on the scene. Indeed there are evidences that the Indian irrigation systems of the Southwest are more than nine hundred years old. The Spaniards naturally had a lively interest in and respect for the accomplishments of Indians in the field of irrigation and noted, in their early journals and records, how closely Indian practices resembled those with which they were familiar. While the New Mexico colonists were familiar with irrigation, it is also apparent that they learned a great deal from the Pueblo Indians.13 Irrigation is an art. To prevent wastage of precious water, soils have to be carefully prepared and leveled; and the question of when to irrigate, and to what extent, are matters learned only from long experience. “There are some arts,” writes Edith Nicoll Ellison, “of which a man becomes master in the course of three hundred years or so. Levelling land is one, irrigation is another. In both these arts the Mexican is at his best. . . . With his big hoe and inherited lore, the Mexican is a valuable person.”14 It was from the Mexican and the Indian that the Anglo-Americans learned how to irrigate.15 After carefully leveling the land, the Mexicans blocked out their fields in squares, the sides of which were just high enough to hold the water. When one block was soaked—not flooded—a hole was made in the side wall of earth and the water was permitted to flow into the next square. This manner of irrigating is still known in the Southwest as “the Mexican system.” The first irrigation systems in Texas, Colorado, New Mexico, Arizona, and California were Mexican-Spanish in origin, if the Indian experience is excepted. It was only after the Anglo-Americans had learned to irrigate, after the MexicanIndian manner, that they became successful irrigation farmers. Irrigation has always been a communal enterprise. In New Mexico, the diversion dams, laterals, and canals were always regarded, and are still regarded, as common property. Every spring the villagers elect a mayordomo who has charge of directing work on the irrigation system. A survey made by the government in 1891 revealed that in New Mexico no one was allowed to take waters from the main irrigation ditch unless he had either personally or by

124  North from Mexico

proxy performed the tasks assigned to him by the mayordomo. All work of this kind was performed by the villagers together, as a joint enterprise. The village type of agricultural settlement in both New Mexico and Utah is in part a consequence of the necessity of communal controls in irrigated farming. The pueblo of Los Angeles, at an early date, appointed a zanjero to keep the main ditch or zanja in repair and the office was continued for many years after the American conquest. The word of the zanjero was supreme in all matters relating to water and took precedence over that of the alcalde, the priest, and the military commander. The zanjero was authorized, if necessary, to impose corvées of labor upon the population and to utilize every resource of the community to preserve the water supply. Many of these practices have a striking similarity with those of the Moors and such Spanish words as acequia, zanja, and zanjero are said to be Arabic in origin.16 To the borderlands, also, the Spanish brought a very considerable lore about water wells, both of the handdrawn and water-wheel variety, and of the technique of drilling wells. Above all the Spanish colonists had an inherited social sense of the importance of water which they transmitted wherever colonies were founded. The attempt of the Anglo-Americans to apply the doctrine of riparian rights in the arid Southwest resulted in years of conflict and litigation and retarded the development of the region. In the end most of the states were forced to repudiate the doctrine or to modify it in many important respects. The only state that had little trouble with irrigation law was New Mexico, where water rights were regulated by immemorial custom. At the present time, most of the Western states have adopted the Arid Region Doctrine, or, as it is sometimes called, the Doctrine of Appropriation. In developing this doctrine, AngloAmerican jurists were no doubt influenced by the law of waters in Mexico. Under Mexican law, the government was vested with ownership of all rights in rivers and streams but could grant the use of waters to private owners. This use could be conferred on both riparian and nonriparian properties but it was customarily conferred subject to certain conditions and limitations so as to ensure the maximum utilization of a limited water supply. While Walter Prescott Webb has said that the riparian rights doctrine was abandoned in the Southwest by necessity, rather than through any conscious borrowings from the civil law of Mexico, other students of the problem have shown that Mexican precedents were frequently cited in the decisions of Western courts and must, therefore, have had some influence on the formation of the present-day Doctrine of Appropriation. Under the Spanish scheme of colonization, the pueblos were invested with certain special “pueblo rights” in respect to water. Usually four square leagues of land were set aside as communal lands belonging to the pueblo. Title to the water in streams flowing through these common lands, including the right to the underground flow, was reserved to the pueblo and its inhabitants, not

The Heritage of the Southwest   125

only for domestic use, but for parks, trees, and nonagricultural purposes. Safeguarded by the Treaty of Guadalupe Hidalgo, these “pueblo rights” proved to be of inestimable value to the City of Los Angeles which succeeded to the rights of the former pueblo. In a famous lawsuit between the City of Los Angeles and the landowners of the San Fernando Valley, the Supreme Court of the United States finally ruled that the “pueblo rights” of the city took precedence over the common law rights of the landowners. Thus the city was given a prior claim to all waters originating within the watershed of the Los Angeles River, a claim paramount to that of all appropriators subsequent to 1781 when the pueblo was founded. One could not, therefore, estimate what the City of Los Angeles owes to the lucky circumstance that it was founded by Spanish colonists. In general, the land-use systems developed by the Spaniards and Mexicans were much better adapted to an arid environment than were those long traditional with the Anglo-Americans. A Mexican homestead consisted of 4,470 acres: twenty-eight times the size of a homestead in the Ohio Valley. The land unit which Anglo-Americans found in the Southwest had no counterpart in the East or Middle West. It was a contribution, as Dr. Webb has said, “from Latin-America, and it came by way of Texas into the Great Plains: it was the cattle ranch.” A cattle ranch might comprise two thousand or twenty thousand acres, depending on the circumstances. The Mexican idea was to give the settler some good land along a stream, for farming and considerably more land, back from the stream and not necessarily contiguous, for stock-grazing. Sometimes a three hundred-foot strip along a stream or acequia-madre extended fifteen miles back from the stream. Grants were never made without an adequate porción or portion of water. This land-use system was of major importance in the rapid development of the sheep and cattle industries in the Southwest. In 1839 Texas enacted a homestead law which was directly patterned upon the Mexican homestead. In fact, “the Texas land system,” writes Dr. Webb, “had for its foundation the Mexican and Spanish system.” Property rights as between husband and wife are regulated today in California, Arizona, New Mexico, and Texas, in accordance with the Spanish ganancial system of community property. “Our whole system,” wrote one California jurist, “by which the rights of property between husband and wife are regulated and determined is borrowed from the civil and Spanish law.” When the first state constitution was adopted in California, specific provision was made for the retention of the Spanish law of community property. Of great incidental benefit to the residents of the Southwest, this system has been called “one of the most important landmarks of Spanish civilization in America.” It was certainly a much more equitable system, so far as the wife was concerned, than the Anglo-Spanish common law doctrine which conferred an almost unrestricted control over the wife’s property on the husband and

126  North from Mexico

recognized virtually no right, on her part, to property accumulated during marriage. Here, again, the needs of the West created a predisposition to accept the Spanish practice. For it has been suggested that the unequal ratio between men and women in California, when the first constitution was adopted, was an important factor in the decision to adopt the Spanish law of community property. Any study of cultural borrowings and mutations in the Southwest must recognize that necessity has been as influential as conscious imitation. In the long run, the basic industries of the region, including mining, sheep- and cattle-raising, and irrigated farming, would have developed much as they have developed without the aid of Spanish precept and example. Peoples change their habits and customs in response to the challenge of a new environment; the question, in this instance, is one of the speed and facility with which these changes were made. On this score there can be little doubt but that the Spanish example greatly accelerated the process of cultural adaptation. For instance, studies which have been made of the California system of irrigation districts, widely imitated throughout the West, have traced the beginnings of this system to the mission establishments and their communal utilization of a limited water supply. It is also true that other cases can be cited in which the lineage of present-day institutions is uncertain and can, perhaps, never be determined. But one does not need to accept the “diffusionist” interpretation of cultural history to accord Spanish-Mexican influences due recognition in the heritage of the Southwest.

9

The Borderlands Are Invaded

Captain Luis de Velasco, who accompanied De Oñate on the entrada of 1598, had a most remarkable wardrobe. It consisted, according to Bolton, of one suit of blue Italian velvet trimmed with wide gold passementerie, with green silk stockings, blue garters, and points of gold lace; a suit of rose satin; one of straw-colored satin; another of purple Castilian cloth; another of chestnutcolored cloth; a sixth, and daintier, suit of Chinese flowered silk; two doublets of Castilian dressed kid and one of royal lion skin gold-trimmed; two linen shirts; six linen handkerchiefs; fourteen pairs of Rouen linen breeches; forty pairs of boots, shoes, and gaiters; a raincoat; three hats including one of purple taffeta trimmed with blue, purple, and yellow feathers and a band of gold and silver passementerie. His equipment naturally matched the splendor of his wardrobe: four saddles; three suits of armor, three suits of horse-armor, a silver-handled lance with gold and purple tassels, a sword and gilded dagger, a broadsword, two shields; a bedstead, two mattresses, numerous sheets, pillows and pillowcases; a bevy of servants; thirty horses and mules, and a silken banner. Indeed, the first Spaniards to invade the borderlands did so with plumes waving, banners flying, and armor gleaming. But when Spanish-speaking people reinvaded the borderlands three hundred years later, their leaders were landless peons who forded across the Rio Grande in the dead of night. Their wardrobe—indeed their worldly possessions—consisted of the clothes they wore. No taffeta-trimmed hats for them; no blue, purple, and yellow feathers; no gold and silver ornaments; no mattresses, sheets, and pillowcases. For these latter-day conquistadores were Mexican cholos who came to chop brush, to build railroads, to work in copper mines, and to pick cotton in lands which De Oñate and Juan Baurista de Anza had mapped and charted, explored and colonized. The first entrada was made up of Spanish hidalgos and caballeros; the second of Mexican peons. The first invaders came in search of gold and silver; the second in search of bread and

128  North from Mexico

a job. What the second invasion lacked in color, splendor, and majesty, was more than offset by the capacity of the peons for hard work and endurance. Those colorful murals that one can see nowadays throughout the Southwest in which figures like Captain Luis de Velasco are depicted in all their finery might well be balanced by a few murals showing Mexican migratory workers sweating in desert cement plants, in the copper mines of Morenci, the smelters of El Paso, and the great farm-factories of the San Joaquin Valley. Captain de Velasco and his colleagues may have discovered the borderlands but Spanishspeaking immigrants from Mexico have built the economic empire which exists in the Southwest today. No one knows precisely how many Mexican immigrants came to the United States in the period from 1900 to 1930 but it is generally agreed that the number was in excess of a million. Prior to 1900 there had been a trickle of Mexican immigration to the borderlands: Texas had an immigrant population of 71,062 in 1900; Arizona 14,171; California, 8,086; New Mexico, 6,649. The bulk of these earlier immigrants, of course, were concentrated in the Southwest; and, of the post-1900 immigrants, nine-tenths settled in the borderlands. At first the immigration was largely restricted to Texas, for the use of oriental labor, particularly the Chinese, barred the way to Mexican immigration in California until about 1917. The rapid increase of Mexican immigrants in the border states after 1900, however, can be seen in the following table: 1900

1910

1920

1930

Arizona.....................

14,171

29,987

61,580

114,173

California ...................

8,086

33,694

88,881

368,013

New Mexico ................

6,649

11,918

20,272

59,340

71,062

125,016

251,827

683,681

Texas .......................

Three facts should be noted about the great wave of Mexican immigration which brought to the Southwest after 1900 nearly ten percent of the total population of Mexico: it was overwhelmingly concentrated in the old Spanish borderlands; in point of time it coincided with the birth of the Southwest as an economic empire; and, in each instance, Mexican immigrants labored in the building of industries in which there had been an earlier SpanishMexican cultural contribution. The industries in which Mexicans were concentrated, moreover, were those vital to the economic development of the Southwest. In all essentials, therefore, the story of the invasion of the borderlands can be told in terms of railroads, cotton, sugar beets, and truck or produce farming.

The Borderlands Are Invaded   129

1. Spanish Trails, American Rails The Spanish introduction of the horse, mule, burro, and ox to America marked the longest stride that so many people, in so short a time, have ever taken in the arts of transportation. charles fletcher lummis

The terrain of the Southwest is rugged, mountainous, and semidesert; waterholes are few and far between; and great reaches have remained wholly uninhabited for centuries. To establish a system of transportation in such a region was infinitely more difficult than to build a flatboat at Pittsburgh and float down the Ohio and the Mississippi to New Orleans. Spain is also a country of elevated mountain ranges, of narrow, winding defiles, of rocky trails, and stretches of torrid plain. From the eighth century, when the Spaniards had acquired the aparejo or packsaddle from the Moors, mules and burros had been used as transport. Washington Irving records that in 1486 Isabella organized, equipped, and maintained pack-trains consisting of fourteen thousand mules and burros which were used to supply an army of fifty thousand in the conquest of Granada. This efficient system of transport the Spanish brought to the Americas, where, for three hundred years, the pack-train was the lifeline of their colonies. The trails that later became rail lines and highways in the Southwest were first discovered, charted, and traveled by Spaniards and Mexicans. The historic trail that De Anza blazed from Tubac to San Gabriel might well be regarded as the initial survey for the present-day Southern Pacific line. Long before the rail lines were built, the Spaniards and Mexicans had organized an elaborate system of pack-trains which operated over the endless trails blazed by the conquistadores. In the early days, the “King’s Wagons”—the famous Carros del Rey—made the long journey from Mexico City to Santa Fe, from Santa Fe to Veracruz, carting merchandise, supplies, and silver from the mines. Crisscrossing the deserts and mountain ranges, these pack-trains were the principal means of transportation as late as the 1880s, transporting merchandise to the towns, supplying the army posts, carrying the mail. The tinkling of the pack-train bell was heard throughout the West until the whistle of the locomotive began to echo in the mountain passes and canyons. Mule and burro transport, so well adapted to desert conditions, was a Spanish invention and Southwestern pioneers were in complete agreement that no one could handle burros or mules as skillfully as Mexicans. For efficient service, pack-trains had to be thoroughly organized. The form of organization and the type of equipment were well established when Columbus discovered America. A typical pack-train consisted of the pack-master or patrón; the head-loader or cargador; a blacksmith; a cook; and eight or ten packers or

130  North from Mexico

arrieros. The duties of each man in the pack-train were defined by custom in the most precise detail. Operating from Tucson to San Diego, from Santa Fe to El Paso, from Tucson to Guaymas, pack-trains constituted an extremely efficient and economical mode of transportation for the time and place. Santiago Hubbel, a famous pack-master of New Mexico, freighted heavy mining equipment overland from Lavaca, Texas, to the Heintzelman mine in Arizona—a distance of 1,200 miles—and his pack-trains carried ore to the Missouri River in rawhide bags. Freight rates ranged from thirty to seventy cents per ton per mile and the mules, carrying an average load of 250 pounds, often made thirty miles a day. It was not uncommon for desert pack-trains to travel three hundred miles in four days and, in 1881, a pack-train, loaded two hundred pounds to the pack mule, traveled eighty-five miles in twelve hours. Pack-trains carried supplies to mining camps throughout the West and New Mexico arrieros made their appearance as far north as the Salmon and Frazer rivers. The operation of these pack-trains was based on the most elaborate and intricate lore. In Spain the pack mules had housings, similar to those of the cavalry, of rich cloth embroidered with gold; halters brocaded with silk; and bridles, headpieces, and harnesses that glittered with silver. The arrieros of the Southwest kept this great tradition alive. In odd moments they decorated their equipment with figures of animals and birds, insignia and legend, woven with silken threads of various colors. A pack saddle, stock and all, often cost $100 and was beautifully stamped and engraved by hand, trimmed with Mexican silver dollars, cut and chased in various designs; while the bridles were inlaid with silver and gold. The arrieros dressed like their grandfathers in Spain: with high-heeled top boots and tiny spurs, silken banda or sash wrapped around the waist two or three times, embroidered shirt front, and conical sombrero with a silver snake around the crown, the underside of the brim being trimmed with silver braid. “Such was the holiday costume of the packer of thirty-five or forty years ago,” wrote H. W. Daly in 1910, “when, mounted on his favorite mule, he would sing some Spanish ditty when visiting friends in some nearby hamlet; a man who never turned his back on a foe or forsook a friend in moments of peril, honest and honorable in all his dealings with his fellow men, kind to animals in his care, with a love for his calling and thoroughly imbued with an ‘esprit de corps’ for the pack service.” “I cannot recall,” wrote Charles Fletcher Lummis, “hearing of any arriero that ever robbed his employer.” Used by General Crook in the sixties in campaigns against the Shoshones in Nevada, pack-trains played an important part in his later campaigns against the Apaches in the Southwest; in fact, some historians believe they were the decisive factor. Among the famous muleteers who operated these supply trains for the army were such names as Chileno John, José de Leon, and Lauriano

The Borderlands Are Invaded   131

Gómez. Later the army used this same pack-train system in the SpanishAmerican War, both in Cuba and in the Philippines, and it became a standardized unit in army transport. Just how intricately organized and delicately articulated the whole system was can be seen by reference to a manual of pack transportation which the army issued in 1910. The technique of roping packsaddles, for example, was highly intricate and the most elaborate lore existed about hitches, knots, and splices. Prepared by Chief Packer H. W. Daly, the manual points out that the whole system, in its every detail, was taken over from the Mexicans intact, including its vocabulary. The load was, of course, the cargo, from the Spanish cargar, to load. The pack-train was the atajo; the pack-cushion, or saddle, the aparejo; the sweatcloth, the suadera; the crupper, the grupera; the saddle cloth, the corona; the cover for the harness was the sobre-en-jalma or sobrejalma; the bell, the cencerro; the canteen, the gerafe; the saddlebags, the alforjas; the bell-mare, the acémila; the eye-blind, the tapaojo; the currycomb, the almohaza. Much of the vocabulary of the pack-train was Arabic in origin; but its organization was Andalusian. Estevan Ochoa, whose family had come from Spain to Mexico in the days of Cortez, was born in the cathedral city of Chihuahua in 1831. As a young man, he had journeyed to Independence, Missouri, where he lived for a time, learning the English language. Returning to the Southwest in 1859, by way of the Santa Fe Trail, he established a chain of stores and a well-organized pack-train system that supplied them. When troops of the Confederacy under Captain Hunter seized Tucson during the Civil War, Ochoa was given the alternative of taking an oath of allegiance to the Confederacy or of leaving Tucson. He answered this ultimatum by climbing on his horse and riding out alone through the Apache territory to Mesilla. A partner in the famous early-day freighting firm of Tully and Ochoa, his pack-trains freighted throughout the Southwest and deep into Mexican territory. As a pioneer resident of Tucson, he led the fight that finally resulted in the establishment of the first public school system in the territory. Member of the territorial legislature and mayor of Tucson, Estevan Ochoa was perhaps the first citizen of Tucson in the 1870s. At his beautiful home in Tucson, scores of friends annually assembled from points as distant as El Paso and Guaymas for the ten-day celebration of the festival of Saint Augustine. When the railroad finally reached Tucson on March 25, 1880, Don Estevan Ochoa presented Charles Crocker with the silver spike used in the dedication ceremonies. As a pioneer freighter in the Southwest, Don Estevan was, in effect, presenting his competitor with a spike to be driven into his own coffin. For the coming of the railroads spelled the end of the pack-trains and proved to be the undoing of Ochoa, who died, a few years later, a ruined man.

132  North from Mexico

2. Life in a Boxcar Mexican labor was extensively used in the construction of the Southern Pacific and Santa Fe lines in the Southwest in the eighties, particularly on the desert sections built by the Southern Pacific. From that day to this, Mexicans have repaired and maintained Western rail lines. As watchmen of the rails, section hands must live near their work, while the extra crews literally live on the rails in boxcars which are shunted about the divisions. “Their abode,” as one railroad executive tersely phrased it, “is where these cars are placed.” Hundreds of Mexican families have spent their entire sojourn in the United States bouncing around the Southwest in boxcar homes. Since 1880 Mexicans have made up seventy percent of the section crews and ninety percent of the extra gangs on the principal Western lines which regularly employ between 35,000 and 50,000 workmen in these categories. In 1930 the Santa Fe reported that it was then employing 14,000 Mexicans; the Rock Island 3,000; the Great Northern 1,500; and the Southern Pacific 10,000. According to the census of 1930, 70,799 Mexicans were engaged in “transportation and communication” mostly as common laborers on the western lines and as maintenance workers on the streetcar systems of the Southwest. In Kansas and Nebraska, Mexican settlements will be found to extend along the rail lines while the colonies of Kansas City and Chicago are outgrowths of Mexican railroad labor camps. As late as 1928 the boxcar labor camps of the railroads housed 469 Mexican men, 155 women, and 372 children in Chicago. The principal large-scale importers of Mexican labor, the rail lines of the Southwest constantly fed workers to other industries since so much railroad labor is seasonal in character. Forever losing labor, the railroads kept recruiting additional workers in Mexico. This process was greatly accelerated as increased freight and passenger traffic paralleled the economic development of the region. Railroad employment naturally stimulated migration, since the companies provided transportation to various points along the line. Just how important the railroads were in setting the tide of Mexican immigration in motion can be seen from a statement made by an investigator for the Department of Labor in 1912. Most of the Mexicans then in the United States, he said, had at one time or another worked for the railroads. For years the prevailing wage for section hands in the Southwest was a dollar a day—considerably below the rate paid for similar labor on the middle western and eastern lines. Recruited by labor agents and commissary companies, Mexicans were assembled in El Paso and from there sent out on six-month work-contracts with the Southern Pacific and Santa Fe. In 1908 some sixteen thousand Mexicans were recruited in El Paso for railroad employment. Two years later as many as two thousand Mexicans crossed the border into El Paso in a single month at the instigation of the commissary companies. Starting around 1900, railroad

The Borderlands Are Invaded   133

recruitment reached its peak in 1910 and 1912. Originally recruited by the Southwestern lines, Mexicans were used after 1905 in an ever-widening arc which gradually extended through Colorado, Wyoming, Utah, Montana, Idaho, Oregon, and Washington. As early as 1900 the Southern Pacific was regularly employing 4,500 Mexicans on its lines in California. By 1906 the Southern Pacific and the Santa Fe were importing as many as two and three carloads of cholos a week to Southern California. The rapid extension of the Pacific Electric interurban system in Southern California also greatly stimulated the demand for Mexican labor. Wherever a railroad labor camp was established, a Mexican colonia exists today. For example, the Mexican settlement in Watts—called Tajauta by the Mexicans—dates from the importation of a carload of cholos in 1906. While the lines were being built, the cholos lived in boxcars and tents. Later the company built row-houses on its property and rented these houses to the employees. Thirty or forty such camps are still to be found in Los Angeles County. Around the initial camp site, Mexicans began to buy lots at $1 down and $1 a week and to build the shacks in which their children live today. In the sparsely settled semiarid Southwest, the construction of the rail lines was well in advance of actual settlement. Elsewhere in the West and Middle West, settlers had promoted railroads; but here railroads promoted settlement. The first great land “boom” in Los Angeles, for example, was strictly a railroad promotion. In the economic development of the region, railroads have played an all-important role. Prior to the completion of the Southern Pacific and Santa Fe lines in the eighties, the Southwest was hardly a part of the United States. In every state in the region, the modern phase in its development dates from the arrival of the first passenger or freight train. Largely built by Mexican labor along routes first explored and mapped by Spanish-speaking people, the railroads of the Southwest have been maintained by Mexicans from 1880 to the present time. All the products of the region—copper, cotton, lettuce, produce, wool, beef, and dairy products—move to markets on desert lines dotted at regular intervals by small, isolated clusters of Mexican section-crew shacks lost in time and space.

3. Cotton in Texas As early as the 1890s, Mexican labor from both sides of the Rio Grande was following the cotton harvest on foot into the old cotton-producing sections of east Texas. East Texas was then devoted to cotton; south and west Texas to cattle. Since the Negroes were concentrated in the eastern section, Mexicans remained a secondary source of labor. But from 1890 to 1910 the cattle industry began to retreat before the forces of King Cotton, first in middle Texas and later (1910–1930) in west Texas. In these areas the plantation system had never

134  North from Mexico

been firmly established and large land units and inflated land values demanded a more “efficient” type of labor at lower costs. This was achieved largely through an increasing use of mechanized methods and the substitution of transient Mexican labor for Negro and “poor white” sharecroppers and tenants. Writing in Current History for February 1930, Remsen Crawford observed that “in the unirrigated plains regions of Texas and Oklahoma millions of acres formerly used as ranges for cattle and sheep and goats are being cultivated in cotton, mainly for non-resident landlords, by Mexican tenants, or hired workers, living in miserable shacks. . . . New cotton gin plants have sprung up everywhere. Old towns have greatly increased in size, and new ones have been built. The bulk of the work of making and gathering this cotton is done by Mexicans imported from over the border.” With the coming of cotton, middle and west Texas were inundated with hordes of people who, in the words of one observer, “planted cotton, talked cotton, thought cotton, sold cotton, everything but ate cotton.” It so happened that the expansion of cotton in these new areas coincided with the first rumblings of the social revolution which began in Mexico in 1910. Thus as cotton pushed its way into the Southwest, Mexicans from across the border came to meet it. “It was in Southwest Texas,” writes Dr. Edward Everett Davis, “that fourteenth century feudalism met the southern plantation” and from this meeting came the large-scale cotton farming of Texas based on the use of migratory Mexican labor. As early as 1908, writes Dr. Davis, the Mexican invasion “had driven itself like a wedge into the heart of Texas beyond San Antonio, veering to the south of the Balcones Escarpment and the ranch country, and sticking close to the cotton fields of Comai, Hays, and Caldwell counties.”1 Coming through the ports of Laredo, Eagle Pass, and Brownsville, the Mexicans had concentrated at San Antonio, and that city, “like the small end of a funnel, poured them out into the cotton fields with such speed that by 1920, the greatest density of rural Mexican population in Texas was not along the Rio Grande but in Caldwell County in sight of the dome of the State Capitol.” In the story of Keglar Hill, Dr. Robert H. Montgomery has given a graphic account of what happened in one Texas community when this invasion began. Keglar Hill had moved from cattle to cotton in one generation. As cotton supplanted cattle, the first Mexicans began to arrive in 1887 and 1888. At first they returned to their homes across the Rio Grande after the cotton harvest. But each year a few stayed over at the end of the season and, as the number increased, they gradually began to displace the white tenants, not as tenants, however, but as day laborers. The white tenants were the first to leave, around 1900, but were soon followed to the towns by their former landlords. Since the Mexicans did not speak English and were not in the habit of sending their children to school, the rural schools vanished with the white tenants and landlords. Of some three hundred-odd white settlers who had made up this small

The Borderlands Are Invaded   135

rural community in 1895, writes Dr. Montgomery, not one remained on the land by 1910.2 Obviously it was the joint appearance of cotton and Mexicans that brought about this disintegration of rural life. But the white tenants and sharecroppers, resenting the new dispensation, tended to blame Mexican labor for the transition, whereas the Mexicans were as much victims of the transformation as the tenants they displaced.“Before the incoming hosts of Mexicans,” writes Dr. Davis, “three rural institutions,—the home, the church, and the school,—fell like a trio of staggering tenpins at the end of a bowling race. White tenants could not compete with cheap Mexican labor. Prosperous owners moved to town, leaving the menial work for Mexicans to do. Rural dwellings, orchards, and yard fences went to wreck; deserted country churches made excellent hay barns and tool sheds for absentee landlords; and the large rural schools packed with happy white children dwindled into sickly institutions for a few indifferent Mexican muchachos, as a wilderness of rag-weeds and cockleburs grew on the school grounds. . . . The Mexican did not hit the interior cotton lands with the impact of a hurricane, but seeped in silently and undermined the rural social structures like termites eating out the sills of a wooden house.” Needless to say, I quote this passage for its revealing explanation of the basis of much anti-Mexican sentiment in Texas rather than to describe the transformation that took place. Actually, new modes of production and new social forces had “undermined the rural social structures” of Texas; the appearance of the Mexicans was a symptom, not a cause, of this process. Mexicans did not “seep” into Texas: they were recruited for employment. To charge them with responsibility for the disappearance of rural schools and churches, while exonerating their employers, was certainly to give a vulgar and misleading interpretation to a familiar economic process. By 1940 nearly four hundred thousand workers, two-thirds of whom were Mexicans, were following “the big swing” through the cotton-growing regions of Texas. Starting in the southern part of the state in June, the migratory army sweeps eastward through the coastal counties and then turns west for the later harvest in the central section. It then splits into three units: one moves into east Texas; another proceeds to the Red River country; and a third treks westward to the San Angelo-Lubbock area. Organized in the south, the army gains recruits as it proceeds along the line of march. From the Lower Valley, where the season starts, comes the initial vanguard of about twenty-five thousand Mexican migratory workers. As the army marches through the RobstownCorpus Christi area, an additional twenty-five thousand recruits join the procession. By the time the army has reached central Texas it has probably grown to two hundred thousand workers. During the depression, as many as four hundred thousand Mexicans made this great circle, traveling distances of from 1,800 to 2,000 miles. Over this long and wearisome route, Mexican families travel “like the starlings and the blackbirds.”

136  North from Mexico

The generals of this tatterdemalion army are the Mexican labor contractors and truckers. About sixty percent of the cotton-picking, in fact, is contracted through these jefes or papacitos. The contractor, who usually speaks English, knows the routes, deals with the employer, and organizes the expedition. He is really a capitán or jefe because he happens to own a truck. Paid to transport workers, he is also hired to weigh the cotton, take charge of the commissary, and oversee the work. The trucks are often loaded with fifty or sixty workers as well as quantities of bedding and equipment. Thoroughly mechanized, the cotton-picking army moves on wheels in open or stake trucks. In an accident at McAllen, Texas, in 1940, forty-four Mexicans riding in one of these trucks were injured: twenty-nine were killed and of these, eleven were children under sixteen years of age—a tragedy memorialized in a famous border corrido. In depression years the army, of course, is inflated to fantastic proportions. The growers are anxious to have their crops harvested as rapidly as possible and, with the pay on a piecework basis, total labor costs remain the same regardless of the size of the crew. Here is a picture of what happens, and has been happening for years, on this great circle (from a report by Pauline Kibbe to the coordinator of InterAmerican Affairs, December 29, 1944); On one Saturday afternoon in October, 496 migratory labor trucks were counted on the streets of Lubbock. Lubbock is a city of between 40,000 and 50,000. Each truck carries an average of 15 migrants, of all ages, which means an estimated total of 7,440 migrants who had come to Lubbock to spend the weekend, seek new employment, purchase their groceries, and other supplies, find a little recreation, etc. . . . Suppose each of the 496 trucks in Lubbock spent an average of $25.00. . . . That is a total of $12,400.00 income to the business places of all kinds for one weekend. Yet Lubbock had made no provision whatever for taking care of this influx of people which occurs regularly every fall, and every weekend during each fall. There is no place provided where they may park their trucks, take a bath, change their clothes, even go to the toilet. In Lamesa it was stated . . . that toilet facilities in the City Hall, which the migrants could use most conveniently, were locked up at noon on Saturdays, and filling station facilities were used except where the owners prohibited it because of the objections of customers.

Of 117 complaints of discrimination against Mexicans in public places filed with the Texas Good Neighbor Commission in 1944, two-thirds of the total involved mistreatment in towns of less than five thousand population. These towns were scattered over the general area through which the migratory labor movement passes. It would seem to be obvious, therefore, that a degree of correlation exists between migration and discrimination.

The Borderlands Are Invaded   137

4. Cotton Moves West With the first World War creating a sharp demand for long-staple cotton, the cotton kingdom jumped from west Texas to the Salt River Valley in Arizona. Agents of the Arizona Cotton Growers Association recruited thousands of workers in Mexico for employment on the large-scale irrigated cotton farms. In most cases, the costs of transportation and of subsistence en route were later deducted from earnings. Up to June 30, 1919, 5,824 Mexicans had been imported into Arizona and 7,269 were specifically recruited for cotton-picking, not including those who came without contracts or special inducements. When the cotton boom exploded after the war, thousands of these workers were stranded in Arizona. In the Salt River Valley alone, ten thousand Mexicans were destitute in the winter of 1921 and the Mexican consul had to seek an appropriation of $17,000 to provide temporary relief. When four thousand imported Mexicans went on strike in the cotton fields in 1920, the growers had their leaders deported and arrested scores of the strikers. At about the same time, the development of irrigated cotton in the Mesilla Valley of New Mexico also began to attract Mexican workers from across the border.3 In 1910 the first cotton was planted in the Imperial Valley in California. As the cotton acreage expanded in response to wartime demands, the Imperial Valley growers began to bring in truckloads of Mexicans, in units of 1,500 and 2,000, from such communities as San Felipe and Guaymas. After the war, cotton production declined for a time and then soared to new heights as the center of production shifted to the huge farm-factories of the San Joaquin Valley. With 5,500 acres planted in cotton in 1919, the acreage in the San Joaquin increased to 172,400 acres in 1931. As the cotton acreage expanded, more and more Mexicans were imported, with the year 1920 being referred to in the farm journals as the first “Mexican harvest.” From 1924 to 1930, an average of 58,000 Mexicans trekked into the San Joaquin Valley each year, principally for the cotton harvest. In the middle twenties, over a hundred trucks loaded with Mexicans were counted crossing the Ridge Route in a single day. The townspeople would provide the Mexicans stranded in the valley towns at the end of the season, from public funds, with just enough gasoline to make the trip back over the Ridge Route to Los Angeles, where welfare and charitable agencies took care of them during the winter months. Commanding a premium price, with a yield-per-acre nearly twice the national average, cotton soon became a $40,000,000 crop in California. Largely produced on high-priced irrigated lands which had been capitalized on the basis of five decades of cheap labor, the expansion of cotton in California was premised upon the availability of a large supply of low-cost labor exclusively earmarked for the cotton growers. Consistently opposed to Mexican immigration, the labor unions failed to note that the cotton fields of the Imperial and

138  North from Mexico

San Joaquin valleys were a major factor in the location of large automobile tire factories in Los Angeles which, in turn, stimulated the demand for industrial labor.

5. Vitamins and Mexicans Irrigation equals Mexicans. dr. paul s. taylor

Twice the size of Germany and larger in area than the thirteen original states, the Southwest had a population in 1902 about half the size of the City of Chicago in the same year. Irrigation was the magic key that unlocked the resources of the region. Irrigated farming is intensive farming: with high yields per acre, heavy labor requirements, year-round production, and crop specialization. Small in area, the “winter gardens” of the Southwest have offset, by their exceptional productivity, many of the disadvantages of an arid environment. Throughout the region, the distribution of Mexicans in rural areas is largely determined by the location of irrigated crops. As an economic empire, the Southwest dates from the passage of the Reclamation Act in 1902 which outlined a development policy for the arid West and made possible the use of federal funds in the construction of large-scale irrigation and reclamation projects. Irrigation has had more to do with the economic growth of the Southwest than any single factor. Today a visitor from the North, driving into the Lower Rio Grande Valley in midwinter, would have the illusion, writes Hart Stilwell, “of moving into a modern version of the Garden of Eden. As he drove along the Valley’s ‘Main Street,’ a 65-mile highway, he would see stately palms and green citrus-fruit trees laden with golden orange fruit; bougainvillea vines in full bloom; bright green papaya plants. . . . And when he came to the level, rich fields he would see bronze-skinned people by the thousands harvesting vegetables—red beets with their green tops, white and purple turnips with their green tops, golden carrots with their green tops—everything green. He might see other laborers harvesting cabbage, broccoli, endive, peppers, beans, tomatoes, new potatoes, peas, anise, cauliflower or squash.” In the valley today Mexican laborers—forty thousand of them illegal entrants or “wetbacks”—plant, cultivate, and harvest fruit and vegetable crops worth $100,000,000 a year. Virtually none of this development existed in 1904 when the St. Louis, Brownsville, and Mexico Railway finally completed its line to Brownsville.4 A miraculous transformation, indeed; but just how was it brought about? “The hand of the Mexican laborer,” wrote Dr. Paul S. Taylor in 1927, “is grubbing out the chaparral of south Texas.” Before the lands could be leveled, planted, and irrigated, the brush had to be cut away for this was the brasada or

The Borderlands Are Invaded   139

brush country of Texas. While an occasional tractor was used to pull out small trees, Mexicans grubbed most of the brush by hand with the use of grubbing hoes. “Grubbing brush,” the Texans said, “is a Mexican job.” And properly so, perhaps, for the Mexicans alone knew and understood the brush country. To the Anglo-Americans, writes J. Frank Dobie, the brasada vegetation was just “brush,” an all-encompassing term; but generations of Mexicans had learned to distinguish, and had named, the endless varieties of shrubs and plants to be found in this section. When these varieties are distinguished today, it is by their Spanish-Mexican names: mogotes, or thick patches of evergreen; coma, with its dirk-like thorns; the cejas or thickets; the wand-like retama with its yellow flowers called lluvia de oro (“showers of gold”); grandjero with its yellow berries; the agarita or wild currant; the bitter amargosa; the black chaparral or chaparro prieto; and the tasajillo or rattail cactus. Thousands of acres of this densely thicketed brush had to be grubbed out by hand to make way for the fabulously rich “winter garden” that exists today in the Lower Valley. After 1900 the increasing urbanization of population, the disappearance of the backyard garden, the development of new canning processes, and the introduction of refrigerator cars, brought about an enormous increase in the production of fruits and vegetables on large-scale commercial farms in the irrigated portions of the Southwest. “Citrus fruit output,” writes Dr. Harry Schwartz, “jumped more than five fold in the first decades of this century, while grape tonnage increased four times. Between 1919 and 1939 production of fresh market spinach, lettuce, cauliflower, snap beans, and carrots increased five times or more, while celery output tripled. In 1900 most fruits and vegetables were not considered sufficiently important to justify extensive collection of statistics regarding them. By 1940 they contributed more than a billion dollars to cash farm income, roughly 30% of the total from all crops.”5 (Emphasis added.) It should be emphasized that this increased production represented a net addition to the total American agricultural income.6 Virtually all of this phenomenal increase occurred in the Southwest and was made possible by the use of Mexican labor. In the growth of commercial fruit and vegetable production in the Southwest between 1900 and 1940, there is not a single crop in the production and harvesting of which Mexicans have not played a major role. This fabulous increase in production—which set the Southwest on its feet financially—could never have taken place so rapidly without the use of Mexican labor. To grow and harvest an acre of wheat, during the 1930s, required on the average only about 13 man-hours of labor; but an acre of lettuce required 125 man-hours and an acre of strawberries 500 man-hours. Unorganized Mexican labor in inexhaustible quantities made this production possible.7 It should be noted that the increased production took place on irrigated acreages within the confines of the old Spanish borderlands. It was here that

140  North from Mexico

the Spanish and the Mexicans had first demonstrated the value of irrigation and had developed an understanding of irrigation techniques. Prior to the development of the Southwest, as Ray Stannard Baker pointed out, Anglo-Saxons had never attempted irrigation on a large scale and irrigation “means a complete change of many racial institutions and customs.” Thus the contribution of Mexican laborers related back to a much earlier cultural contribution which had been made by the Spanish and Mexican settlers. Reclaimed and irrigated at enormous costs, much of this land was overcapitalized. Overcapitalization, in turn, had created a terrific pressure for cheap labor. But the years during which the Southwest became an economic empire saw a great increase in living standards for the American working class brought about by the use of more efficient machines. It was precisely in the production of truck crops, fruit, cotton, and sugar beets that the least progress had been made in the introduction of machines. Thus the use of Mexican labor fitted into the economic cogs of the Southwest in perfect fashion.8 To appreciate what Mexican labor meant to the economy of the Southwest, one simple, obvious fact needs to be stressed, namely, the desert or semidesert character of the region. In the San Joaquin, Imperial, Salt River, Mesilla, and Lower Rio Grande valleys, temperatures of 100, 110, and 112 degrees are not uncommon. Those who have never visited the copper mines of Morenci in July or the cotton fields of the San Joaquin Valley in September or the cantaloupe fields of Imperial Valley in June are hardly in a position even to imagine what Mexican workers have endured in these areas. It should be remembered that the development of the Southwest occurred at a time when the living and working conditions of American workmen were undergoing rapid improvement. It was not easy to find in these years a large supply of labor that would brave the desert heat and perform the monotonous stoop-labor, hand-labor tasks which the agriculture of the Southwest demanded. Under the circumstances, the use of Mexican labor was largely noncompetitive and nearly indispensable.

6. Coyotes and Man-Snatchers Spearheaded by the completion of the rail lines, the westward movement of cotton, the spread of “winter garden” fruit and vegetable production, and the phenomenally rapid economic expansion of the Southwest after 1900 created an enormous demand for unskilled labor. Mexicans poured into Texas by the thousands. “Farming is not a profitable industry” said John Nance Garner the Uvalde millionaire, “and in order to make money you have to have cheap labor.” From 70,981 in 1900 the number of Mexicans in Texas shot up to 683,681 in 1930. With enforcement of the contract-labor law being suspended from 1918 to 1921, over 50,000 workers were directly recruited in Mexico for

The Borderlands Are Invaded   141

employment in the United States. By the time the border patrol was established in 1924, and administrative restriction adopted as a policy in March 1929, the great labor pool of Texas had been filled to overflowing with Mexican immigrants. From this reservoir Mexican labor was siphoned off in all directions as the war and the Immigration Act of 1924 drastically reduced the volume of European immigration. The tip of the wave of Mexican immigration reached as far north as Detroit and as far east as Pittsburgh. “Mexicans,” said a California grower in 1927, “scatter like clouds. They are all over America.” From Texas, Mexicans were recruited, in small numbers, for employment on the plantations in the Mississippi Delta; thousands were recruited for employment in the Northern and Western sugar-beet fields; and an entire trainload was, at one time, shipped from Texas to Seattle for employment in the Alaska canneries. Throughout the borderlands prior to 1924 the contract-labor law of 1885 was more often honored in the breach than in its observance. For a quarter of a century, Texas growers had recruited labor in Mexico whenever they needed it. In large measure this traffic had been made possible by the activities of labor smugglers who developed a lucrative racket in Mexicans. The labor smuggler or “coyote” crossed the border, not only to round up crews, but to get workers across the line in violation of the immigration regulations. For a fee of ten or fifteen dollars, the coyote would arrange to get Mexicans across the line, by having them “jump the fence” at La Colorado; or come across concealed in automobiles, carts, or trucks; or by fording the Rio Grande at night. In many cases, forged passports and head-tax receipts were provided. Once across the line, the Mexican was turned over by the coyote to a labor contractor (enganchista), who sold him for a fee of fifty cents to one dollar a head to some agricultural, railroad, or mining employer. Labor agents operating out of Laredo and El Paso had forwarding agents elsewhere in Texas, notably in San Antonio. Charging the employers a fee for supplying the labor, the contractors charged the workers for transportation and subsistence en route. The profits in this racket were really enormous and the smugglers and coyotes and laborcontractors constituted an intimate and powerful alliance from Calexico to Brownsville. Another type of agent, the man-snatcher, also figured in this dubious traffic. The man-snatchers made a business of stealing Mexican labor and selling the same crew to several different employers. Delivering a crew to an employer, they would steal the crew at night and resell it to still another employer. In this manner, the same crew would often be sold to four or five employers in the course of a few days. Frequently the man-snatchers raided crews imported by the labor contractors and made off with them by force of arms. Shipments of workers en route to employers were often kept locked up at night, in barns, warehouses, and corrals, with armed guards posted to prevent their theft.

142  North from Mexico

Crews of imported Mexicans were marched through the streets of San Antonio under armed guard in broad daylight and, in Gonzales County, workers who attempted to breach their contracts were chained to posts and guarded by men with shotguns.9 “Large planters,” wrote James L. Slayden,“welcome the Mexican immigrant as they would welcome fresh arrivals from the Congo, without a thought of the social and political embarrassment to their country.”

7. Los Betabeleros Mexicans have been identified with the sugar-beet industry since its inception; the phrase “Mexican sugar-beet worker” is as common as “Mexican sheepherder.” Prior to the tariff of 1897, the sugar-beet acreage in this country was insignificant; but, by taxing foreign sugar seventy-five percent of its value, the Dingley Tariff immediately created a great demand for sugar beets. States paid bounties, offered tax exemption, and otherwise encouraged the growing of sugar beets as a matter of official policy. From 135,000 acres in 1899, the sugar-beet acreage increased to 376,000 acres in 1906 and, for the last decade, has averaged about 750,000 acres. With each increase in the sugar-beet acreage, of course, the demand for Mexican labor has been stepped up. The production of a large acreage in sugar beets has consistently required, until the last few years, a large amount of labor, for blocking and thinning in the spring and for harvesting in the fall. To induce labor to stay over in the area, so as to be available for the harvest, the sugar-beet companies have always used the device of contract-labor in which workers, more often families, contract to block, thin, and harvest beets for a stipulated sum per acre. For many years, the major companies included in the contract a “hold-back” provision to ensure that the workers would be present in the fall for the harvest. The growing of sugar beets is unique in that it represents “a curious union of family farms and million dollar corporations.” The sugar-beet refineries, rather than the individual farmers, have long assumed responsibility for the recruitment, distribution, and control of the labor supply. The principal growing areas are to be found in California, Michigan, and Colorado. From the beginning of the industry, Mexican workers were used in California. Elsewhere the sugar-beet companies at first experimented with other types of labor: Japanese in Colorado; the so-called “Volga-Germans” in Nebraska and other areas; and Belgians and Poles in Michigan. But these nonMexican groups showed a tendency to aspire to farm ownership and, in some areas, by dint of unbelievably hard effort, succeeded in achieving their goal. It was to curb this tendency that the companies shifted to Mexican labor, particularly after the first World War and the passage of the Immigration Act of 1924. In fact, it was at the insistence of the sugar-beet companies that the contract-labor law was suspended from 1918 to 1920 to permit direct recruitment

The Borderlands Are Invaded   143

in Mexico. By 1927 it was estimated that, of 58,000 sugar-beet workers, 30,000 were Mexicans. Today sixty-six percent of the 100,000 workers in the industry are Mexicans. In states such as Ohio, Michigan, Minnesota, and North Dakota, Mexicans constitute from seventy-five to ninety percent of the labor supply. In Colorado the Great Western Sugar Beet Company began to recruit Mexican labor in 1916, at first from southern Colorado and New Mexico, and, later, from Mexico. In 1920 the company spent $360,000 in the recruitment of Mexican labor. In one year alone, 1921, some ten thousand Mexican workers were imported, most of them from Texas; and, between 1910 and 1930 a total of at least thirty thousand Mexicans were brought to the Colorado sugar-beet areas. Throughout its history, the company has shown a marked preference for the “unspoiled” Mexican, “fresh from Mexico clad in the sombrero, light cotton clothing, and even sandals of the Mexican peon.” Even during the depression years, the company attempted to recruit Mexican labor from outside the state. In an effort to prevent this practice, the governor of Colorado on April 20, 1936, proclaimed a state of martial law and stationed Colorado National Guardsmen along the Colorado-New Mexico border to turn back Mexican workers from New Mexico. At the outset, the sugar-beet companies sought to anchor the imported labor in the sugar-beet areas: first, by some feeble efforts at colonization; then by offering a bonus to those families who agreed to waive the payment of transportation back to the point of recruitment; and, more frequently, by simply stalling on final settlement under the contract at the end of the season. The earnings of the Mexican beet workers were, in not a few cases, so low that they were compelled to stay over during the winter months. Thus wherever Mexican labor has been imported for sugar-beet employment, small Mexican colonies have developed. In the South Platte area of Colorado, the number of stay-overs rose in a six year period from 537 to 2,084. El Paso was the principal place of recruitment for Mexican sugar-beet workers in the Rocky Mountain states, with San Antonio being the center from which they were recruited for the principal Middle Western areas. Over the years, however, Denver has come to occupy, in relation to the sugar-beet areas in Utah, Wyoming, and Montana, much the same relationship that El Paso and San Antonio occupied at an earlier period. There are today some 14,631 Spanish-speaking residents in Denver. Needless to say, the Texas growers became extremely annoyed when the agents for the sugar-beet companies began to tap their great reservoir of Mexican labor. Noting that more and more Mexican labor tended to stay over in the sugar-beet areas, the cotton growers took matters into their own hands. Holes were shot in the tires on the trucks of the sugar-beet agents and Mexicans, in any number of cases, were prevented by force from keeping their rendezvous with the sugar-beet agents. In a final effort to stop this out-of-state recruitment, the cotton growers secured the passage of the Texas Emigrant

144  North from Mexico

Agent Law of 1929 which, in effect, barred outside agents from recruiting labor in Texas. The principal consequence of this law was to make of out-ofstate recruitment a kind of illegal, underground conspiracy. For, despite the law, thousands of Mexicans continued to leave every year for out-of-state employment. For example, it has been estimated that nearly sixty thousand Texas-Mexicans leave the state every year. Originally this movement was organized by the sugar-beet companies through the use of special trains; but, for the last decade, it has been handled by Mexican truckers and contractors. Most of the trucks are open, stake trucks, never intended for passenger transportation. Planks or benches are placed on the truck, which is then loaded with passengers and equipment. Frequently fifty or sixty Mexicans are huddled, like sheep, in these trucks. Once the Mexicans have crowded into the back of the truck, a heavy tarpaulin is thrown over them and fastened down around the edges so that the passengers are concealed. The reason for this conspiratorial atmosphere is, of course, that perhaps two-thirds of the Mexicans who leave the state have been recruited in violation of the Emigrant Agent Law. Outwardly the truck looks as though it were loaded with a cargo of potatoes. Before climbing into the driver’s seat, the trucker tosses a couple of coffee cans into the back of the truck which are used as urinals during the long journey north. Then, usually around midnight, the truck rolls out of San Antonio and heads north. With a relief driver in the cab, the truckers drive straight through to Michigan, and other equally remote destinations, stopping only for gas and oil. By driving night and day, they can make the trip in forty-five to fifty hours. Paid $10 a head to deliver Mexicans in Michigan, which is ultimately charged up by the sugar-beet companies against earnings, the average trucker can make about $3,000 a season. Naturally the truckers are in a hurry; they want to make, if possible, two or three trips. Instead of traveling the main highways, they pursue a crazily zigzag course, making many detours, zooming along country roads and minor highways, in an effort to avoid highway patrolmen. Almost every season since this traffic developed serious accidents have occurred along the line of march. Wherever sugar beets are grown—in Michigan and Minnesota, in Colorado and Montana—the same pattern can be traced. In the sugar-beet area around Findlay, Ohio—dominated by the Great Lakes Sugar Company—nearly three thousand Mexican workers are imported each year from Texas. Telesforo Mandujano, and his six sons, made the trip to Ohio from San Antonio in a truck that carried forty Mexican workers. Here is his abbreviated statement before the Tolan Committee: “The truck did stop a few times for bowel evacuations and eating, when the truck needed gas or oil; but on most occasions cans were used as urinals and dumped out of the truck. Passengers had to stand all the way and one man tied himself upright to a stake so he could not fall out if he should happen to fall asleep.” After deducting all expenses, Telesforo and his

The Borderlands Are Invaded   145

six sons made $200.10 for the season. Catarino Ramirez also made the trip from San Antonio to Findlay in a truck that carried thirty-seven adults and eight children, making the trip in two days and three nights. “No stops were made unless the truckers were forced to,” he testified, “and when such stops were made we ate if we had time.” No aspect of Mexican immigration has been more frequently or more thoroughly investigated than the history of their employment in sugar beets. The documentation is voluminous and covers every aspect of their employment in every sugar-beet area in the United States. It would serve no purpose, here, to attempt a summarization of this data, with its appalling revelation of low earnings, miserable health and housing conditions, child labor, sickness and disease. The data, for those who are interested, is summarized in Part 19 of the Tolan Committee Hearings (the “Detroit Hearings,” September 23, 24, 25, 1941). Suffice it to say, that from 1916 to the present time, between thirty thousand and sixty thousand Mexican workers have been directly dependent upon sugar-beet employment with average annual earnings of from $500 to $600 per family.

8. In Midwest Industries Beginning around 1916, Mexican laborers began to appear in the Chicago industrial area, in Gary, Indiana Harbor, and Calumet. Most of these workers were from such states as Jalisco, Michoacan, and Guanajuato, having “leapfrogged” from the interior of Mexico to the Midwest industrial centers, “literally passing through and beyond,” writes Dr. Paul S. Taylor, “their compatriots of the Mexican northern border states who have made the shorter migration to the adjacent southwestern United States.” Appearing first as track laborers, they were later employed in the steel mills, the packing plants, and the tanneries. From 1920 to 1930 the Mexican population of Chicago increased from 3,854 to 19,362 and is today generally estimated at about 25,000. The small Mexican colony in Detroit had its beginnings in 1918, when several hundred Mexican workers were brought to work in the automobile industry as student-workers. From eight thousand in 1920, the Mexican colony in Detroit rose to a peak of fifteen thousand in 1928, and then declined during the depression years, when many workers returned to the Texas communities from which they had been recruited or were repatriated by welfare agencies to Mexico. Today the colony is estimated to number about 6,515 residents. Throughout the Midwest there are similar colonies, usually quite small, in most of the industrial centers, totaling perhaps seventy thousand for the region. In 1923 the National Tube Company, an affiliate of U.S. Steel, brought 1,300 Mexicans from Texas to work in its plant at Lorain, Ohio. And, in the same year, the Bethlehem Steel Company imported about one thousand to

146  North from Mexico

work in its plant at Bethlehem, Pennsylvania. When the first trainload arrived from San Antonio, police were on hand to escort the workers to their barracks as the “Mexicans in their broad plain sombreros” marched stoically through the streets. The movement to import Mexican labor for industrial employment would have reached large proportions after 1923 had it not been for two factors. Both the sugar-beet companies and the Southwestern agricultural interests were strenuously opposed to any attempt to establish a pattern of Mexican employment in industry and also feared that, if Mexicans spread into new territory, the agitation for a quota on Mexican immigration would immediately assume serious proportions. Hence the passage of the Texas Emigrant Agent Law in 1929, and the depression of that year, brought the movement to a halt. The second factor had to do with immigration policy. The first bill aimed at placing Mexican immigration on a quota basis was introduced in 1926 and was supported by a rapidly growing exclusionist sentiment. Largely for diplomatic reasons, the federal government in March 1929, adopted a policy of “administrative restriction” by simply tightening up the enforcement of existing immigration regulations with the aid and cooperation of the Mexican government. Thus the northbound movement of Mexican immigrants had virtually ceased by 1930; and, in the depression years, some sixty-five thousand Mexican immigrants were repatriated, some voluntarily, some with the aid of the Mexican government, some being summarily shipped back to Mexico by welfare agencies in this country. This pattern of events has a twofold bearing on the so-called “Mexican Problem” in this country. It explains, to some extent, the concentration of Mexicans in the old borderlands region; and it also explains why so many immigrants have been involved in cul-de-sac types of employment. For the doors of Middle Western industrial employment were closed almost as soon as they were opened; and, in the Southwest, employment opportunities were restricted, by custom, by discrimination, and by other factors, to a few limited types of employment. It is not surprising, therefore, that the 1930 census should show that of Mexicans gainfully employed, ten years of age or over, 189,005 should have been employed in agriculture, 150,604 as common laborers.

9. The Balance Sheet Testifying before congressional committees in the twenties, the principal employers of Mexican labor in the Southwest presented facts and figures showing that Mexicans had been a vital factor in the development of agricultural and industrial enterprises valued at $5,000,000,000. Starting with a scant production in 1900, the Southwest was by 1929 producing between 300,000 and 500,000 carloads of vegetables, fruit, and truck crops—forty percent of the

The Borderlands Are Invaded   147

nation’s supply of these products. Most of this development took place in less than two decades and was directly based on the use of Mexican labor which constituted from sixty-five to eighty-five percent of the common labor used in the production of these crops. All the gold and silver that have ever been taken from the mountains of Colorado or that may still be awaiting the touch of pick and drill, cannot compare in value to the wealth already produced by sugar beets.10 From 1900 to 1940, Mexican labor constituted sixty percent of the common labor in the mines of the Southwest and from sixty to ninety percent of the section and extra gangs employed on eighteen western railroads. Obviously the transformation of the Southwest which has occurred in the last forty years was largely made possible by the use of Mexican labor. Conversely, the employment of Mexicans in the Southwest has been of enormous importance to Mexico in this same period. Some gauge of this importance may be found in the fact that from 1917 to 1927, Mexican immigrants sent a yearly average of $10,173,719.31 in remittances to families in Mexico. It was not only the availability of Mexican labor for this period that meant so much to the economy of the Southwest, but its exceptional adaptability to the types of work and to the environmental conditions which prevailed throughout the region. Mexicans were inured to the heat, the aridity, the dust, the isolation that prevailed in most of the areas in which they worked. With Oriental immigration barred and European immigration placed on a rigid quota basis, it is, indeed, questionable if employers could have found other available labor sources. While these employers undoubtedly exaggerated the “indispensability” of the Mexican, it is nevertheless apparent that there was a measure of truth in what they said. For example, in 1920, the Department of Labor appointed a committee to investigate the consequences of Mexican immigration. Two of the members of this committee were old-time officials of the American Federation of Labor. Yet this committee reported that the Mexicans who had been imported during the war years had not come into competition with or displaced “white” labor. In part, the special adaptability of Mexican labor was related to the presence in the Southwest of a large resident Mexican-American population. This circumstance greatly facilitated the use of Mexican labor. From this element came many of the foremen, the straw-bosses, and the contractors who recruited, transported, and supervised Mexican labor. Wherever Mexican immigrants moved in the Southwest, they found colonies of Mexican residents, with Mexican rooming houses, restaurants, barber shops, and stores; and, of course, throughout the entire area they found Spanish-speaking people. Always a buffer group, the native-born Mexican-Americans were the go-betweens, the conduits, which made possible the rapid, large-scale utilization of Mexican labor. Nowadays the migratory phase of Mexican immigration is drawing to a close. By 1937 machines had taken over about ninety percent of the work

148  North from Mexico

involved in preparing, bedding, and cultivating the land in produce or truck crops; and the practice of crating produce in the fields, rather than in the sheds, had brought about a noticeable reduction in the number of workers. Utilizing mobile, mechanized packing sheds in the fields, the shipper-growers of Texas, Arizona, and California are nowadays drastically reducing the number of workers required. Mechanical cotton-pickers are reducing the number of pickers in California at the rate of one machine where forty workers were formerly used and at a picking cost of $5 per bale; and the first successful experiment in the mechanization of the sugar-beet industry occurred in 1936.11 The use of segmented seed, first introduced in 1943, has already greatly reduced the number of workers needed to thin sugar beets. Blocking and thinning machines have been in use since 1940. Experiments have shown that the use of these machines reduces the man-hours of labor required to thin sugar beets from 27.2 man-hours per acre to 2.5 man-hours per acre. The final step in mechanization in sugar beets is, of course, the introduction of the new mechanical harvester which picks, tops, and sorts the beets. One such machine will replace from six to eight men at harvest time. First placed in general use in 1946, this new harvester is destined to have a revolutionary effect on sugarbeet production. Generally speaking, much of the seasonal farm-labor problem of the past two or three decades has resulted from the much more rapid development of machines and techniques for handling the preparation of the soil, the sowing of crops and their cultivation, than of machines for the harvesting of crops. With the perfection of harvesting machines, the rate of technological displacement in agriculture is certain to increase. In social terms this means that the isolation of the Mexican is drawing to a close. Large employers of Mexican labor have consistently pursued a policy of isolating Mexicans as a means of holding them to certain limited categories of work. Systematically discouraging all “outside contacts,” they have kept Mexicans segregated by occupation and by residence. The effect of this policy has been to create several different labor markets, coexisting in time and space, with each group being relatively insulated from the competition of the others by the different attitudes which employers have cultivated toward particular jobs. This policy has directly fostered antagonism against the Mexican and has severely limited his opportunities for acculturation. Technological change, however, is not solely responsible for the gradual abandonment of this policy. The truth is that the Mexicans have forced its abandonment, as the next chapter will explain.

10

The Second Defeat

Having been defeated in their first encounter with Anglo-Americans in the Southwest, the Spanish-speaking people were naturally somewhat reluctant to seek a new accord. Like other minorities under similar circumstances, they sought to minimize contacts with the dominant group by withdrawing into their own world. As time passed, it seemed as though relations between the two groups might be stabilized on the basis of a bicultural accommodation. But the partial accommodation which had been achieved by 1900 was completely disrupted by the avalanche of immigration. As thousands of immigrants streamed across the border at a dozen points, the old conflict of cultures was renewed. The new immigrants were, of course, fitted into the mold of subordination which had previously crystallized in the borderlands. Once they had become conscious, however, of the way in which they were being subordinated in the social structure, the immigrants attempted to rebel. Dating from the late twenties, this rebellion was most decisively crushed. It is this defeat—the second which Spanish-speaking people have suffered in the borderlands—that explains the present-day tendency of the immigrants to retire within the confines of the Mexican colonia and to seek, as their predecessors had sought, a bicultural accommodation. The rebellion of the immigrants found expression in the form of militant trade-unionism; in more than one field, Mexican immigrants have been the pioneers of the trade-union movement in the Southwest.

1. The Myth of Docility When the tide of immigration reached California (the number of immigrants trebled between 1920 and 1930), the growers were at first most eloquent in their praise of Mexican labor and every attempt to restrict Mexican immigration was strenuously resisted. “No labor that has ever come to the United States,” said Dr. George P. Clements of the Los Angeles Chamber of Commerce, “is more satisfactory under righteous treatment. The Mexican as [sic] the result of years of servitude, has always looked upon his employer as his patrón, and

150  North from Mexico

upon himself as part of the establishment.” “The Mexican,” wrote Ralph Taylor, editorial spokesman for the large shipper-grower interests, “has no political ambitions; he does not aspire to dominate the political affairs of the community in which he lives.” Testifying before a congressional committee, a lobbyist for the California growers explained that “the Mexican likes the sunshine against an adobe wall with a few tortillas and in the off time he drifts across the border where he may have these things.” In a speech in 1927, Simon J. Lubin had said that California growers were treating the Mexicans like peons, corralling them in barbed-wire stockades on the ranches. To this charge, the influential Pacific Rural Press replied: “Peon? Isn’t the word peon a little out of character when applied to a Mexican family which buzzes around in its own battered flivver, going from crop to crop, seeing Beautiful California, breathing its air, eating its food, and finally doing the homing pigeon stunt back to Mexico with more money than their neighbors dreamed existed?” It is true that the immigrants were at first pleased with the new opportunities which they found in the border states. But as they came to realize that the occupations assigned them and the conditions under which they worked were regarded by American urban labor as undesirable and substandard, they began to show signs of restiveness. Not only were they set apart as a caste— stereotyped, segregated, and regarded as an inferior “race”—but the discrimination which they encountered in most California communities had the effect of stimulating them to organize in self-protection. When several hundred cowboys had gone on strike in the Panhandle in 1883—the first attempt to form a union of “agricultural” workers in the United States—the strike call was signed by one Juan Gómez. As early as 1903, over a thousand Mexican and Japanese sugar-beet workers went on strike at Ventura, California. The wave of strikes which culminated in the dynamiting of the Los Angeles Times in 1910 was initiated by a strike of Mexican workers on the local street railway. In 1922 Mexican field-workers had sought to establish a union of grape-pickers at Fresno. Most of these early efforts, however, proved to be abortive. The first stable organization of Mexican workers was established in Southern California in 1927, when the CUOM was formed: the Confederación de Uniones Obreras Mexicanas, with three thousand workers organized in twenty locals throughout the region. The first strike called by the union, in Imperial Valley in 1928, was broken by wholesale arrests and deportations. Two years later five thousand Mexican field-workers struck in Imperial Valley for the second time. Caught by surprise, the growers were forced to settle the strike. But a few months later, when the cantaloupe harvest began, the union was viciously attacked before it could call a strike. Over 103 arrests were made and a local newspaper reported that “the county has purchased more teargas guns, shells, and bombs than ever before.”

The Second Defeat   151

In June 1933, seven thousand Mexicans walked out of the berry, onion, and celery fields of Los Angeles County in the largest strike of agricultural workers that had occurred in California up to that time. It was this strike which first aroused acute apprehensions on the part of the growers that the Mexicans might not be quite as docile as they had imagined. “In my opinion,” said Dr. Clements, “this is the most serious outbreak of the Mexican workers here.” Later, in the fall of 1933, the left-wing Cannery and Agricultural Workers Industrial Union called a series of great farm strikes in California. Threefourths of the strikers in the southern San Joaquin Valley were Mexicans who had been drawn to the union, writes Dr. Paul S. Taylor, “by race discrimination, poor housing, and low pay.” Reported at great length and with elaborate documentation in the LaFollette Committee hearings, these strikes were broken by the use of wholesale violence. “We protect our farmers here in Kern County,” said a deputy sheriff.“They are our best people. . . . They keep the county going. . . . But the Mexicans are trash. They have no standard of living. We herd them like pigs.” Later in the year Mexicans struck for the third time in Imperial Valley. On this occasion, they provoked a whirlwind of violence: union meetings were broken up by tear gas and clubs; several labor lawyers and “outside” spectators were kidnapped by the growers and escorted to the county line; and over eighty-six arrests were made. In 1936 Mexican field-workers were involved in two strikes in Southern California which forever ended the infatuation of the growers with Mexican labor and left a heritage of ill-will which still survives in the region. The first of these strikes occurred in the celery fields located, so to speak, in the backyards of Los Angeles County. With two thousand workers on strike, the police marshaled a force of approximately 1,500 armed men to break the strike. When the strikers attempted to march from their headquarters to the fields, Captain William (“Red”) Hynes and his men broke up the procession on three successive days. Strikers were pursued by the police to their homes and, in a number of cases, tear-gas bombs were tossed into shacks where children were playing. One striker was seriously wounded and another badly burned when police fired tear-gas bombs from a distance of five or six feet. In the Domingues hills near San Pedro—on the doorstep of the beautiful Palos Verdes estates—a miniature battle was staged when police converged on an old barn in which the strikers had barricaded themselves. So many arrests were made that neither the police nor the union could keep a tally. Injured strikers had difficulty in securing medical aid at the county hospital and public funds were used to employ field agents who visited the growers and urged them not to settle. The growers alone spent thousands of dollars in the employment of armed guards recruited from a local strikebreaking detective agency. In midsummer 1936 a strike of 2,500 Mexicans tied up for several weeks a $20,000,000 citrus crop in Orange County. During this strike, Orange County

152  North from Mexico

was in a state of virtual siege; even highway traffic was under close police surveillance. Over four hundred special armed guards were recruited. Two hundred arrested strikers were formally arraigned in an outdoor bullpen or stockade which served as a courtroom. Guards with rifles and shotguns patrolled the citrus belt and the sheriff issued “shoot-to-kill” orders. I was in court one day when fifty or more strikers were brought in by guards armed with submachine guns, rifles, and shotguns. The Los Angeles Examiner spoke feelingly of the “quieting effects of the wholesale arrests,” while the Los Angeles Times gave a graphic account of one raid: “suddenly, late in the night, three or four automobiles loaded with grim faced men, appeared out of the darkness surrounding the little settlement. In a few seconds, tear gas bombs hissed into the small building where the asserted strikers were in conclave, the conferees with smarting eyes broke and ran out under cover of darkness and the meeting was at an end.” (My emphasis.) In a front-page story on July seventh, the Times gleefully announced that “old vigilante days were revived in the orchards of Orange County yesterday as one man lay near death and scores nursed injuries.”

2. The Honeymoon Is Over By 1930 the myth of the docility of Mexican labor had been thoroughly exploded and the Mexicans had fallen in grace in the eyes of the Associated Farmers. “The Mexican on relief,” wrote the irrepressible Dr. Clements, “is being unionized and is being used to foment strikes among the few still loyal Mexican workers. The Mexican casual labor is lost to the California farmer unless immediate action is taken to get him off relief.” “Getting the Mexican off relief ” involved large-scale forced repatriations to Mexico. I watched the first shipment of “repatriated” Mexicans leave Los Angeles in February 1931. The loading process began at six o’clock in the morning. Repatriados arrived by the truckload—men, women, and children— with dogs, cats, and goats; half-open suitcases, rolls of bedding, and lunch baskets. It cost the County of Los Angeles $77,249.29 to repatriate one trainload, but the savings in relief amounted to $347,468.41 for this one shipment. In 1932 alone over eleven thousand Mexicans were repatriated from Los Angeles. A few years later Dr. Clements was writing that “of the 175,000 Mexicans who from 1917 to 1930 met the agricultural labor requirements of the whole state, moving from place to place to meet seasonal demands, there were possibly not more than 10% available in 1936.” This background is of major importance to an understanding of “attitudes” in Southern California. For it shows, first of all, that Mexicans were quick to rebel against the subordinate status which had been imposed upon them. “The most effective agricultural labor unions during 1935 and 1936,” writes

The Second Defeat   153

Dr. Stuart Jamieson,1 “were those organized among Mexicans.” Long charged with a lack of “leadership” and talent for organization, they proved all too effectively that neither talent was lacking. The strikes in California in the thirties, moreover, were duplicated wherever Mexicans were employed in agriculture. Mexican field-workers struck in Arizona; in Idaho and Washington; in Colorado; in Michigan; and in the Lower Rio Grande Valley in Texas. When Mexican sheep-shearers went on strike in west Texas in 1934, one of the sheepmen made a speech in which he said: “We are a pretty poor bunch of white men if we are going to sit here and let a bunch of Mexicans tell us what to do.” In 1933 a fifty-car caravan of Mexican workers, members of the Asociación de Jornaleros, paraded the length of the Lower Rio Grande Valley in protest against antiunion activity. In 1934 the Mexican pecan-shellers, employed in the sweatshops of San Antonio, struck against piece-rates of 2¢ and 3¢ a pound for shelling pecans. Over six thousand workers were involved in this protracted strike. Later the introduction of shelling machines deprived most of these workers of their jobs, as the number of pecan-shellers was cut from twelve thousand to two thousand.2 With scarcely an exception, every strike in which Mexicans participated in the borderlands in the thirties was broken by the use of violence and was followed by deportations. In most of these strikes, Mexican workers stood alone, that is, they were not supported by organized labor, for their organizations, for the most part, were affiliated neither with the CIO nor the AFL. “Logic,” writes Dr. R. W. Rosskelly, of the Colorado State Agricultural College, “suggests the impossibility of scoffing at the Mexican culture patterns, of indoctrinating them with those of the Nordics and still expecting them to perform a type of labor and live under conditions which Nordic standards taboo. Neither can it be expected that they will willingly relegate themselves to the status of second-class citizens in a country where equal opportunity, regardless of race, is the symbol of freedom.” Yet this inconsistency in the treatment of Mexican immigrants continues to be the most conspicuous aspect of Anglo-Hispano relations in the United States. Nothing illustrates more clearly the degree to which Mexican immigrants have shown a capacity for “assimilation” than the rebellion which they attempted in the thirties. In two decades, they had learned to protest, in a typically American fashion, against an annual family wage of $600; against poor housing; and above all, against discrimination. In analyzing these strikes, Dr. Stuart Jamieson points out that Mexicans had become dissatisfied with their “distinct status as a lower caste, which they held because of their poverty, color, and cultural attributes. Their position . . . in many ways came to parallel that of Negroes in the Southern States. ‘White Trade Only’ signs appeared in business establishments in some towns, segregation in seating arrangements was imposed in moving picture houses, residential restrictions were applied to real estate, and

154  North from Mexico

a sentiment for segregation in the schools was widespread. Mexicans and Spanish-Americans also faced discrimination before the law.” Once their commendable efforts toward self-organization were crushed—with violence and gross brutality, with mass arrests, deportations, and “repatriations”—the immigrants became demoralized and momentarily abandoned any attempt to establish a rapprochement with the Anglo-Americans. For the suppression of these strikes represented, in effect, a second defeat, a second rebuff.

3. The Gallup Incident In the middle thirties, several thousand Mexican coal miners, employed by the Gallup-American Company in New Mexico (a subsidiary of Kennecott Copper Company), went on strike. Although no violence was reported, the area was placed under martial law for six months. During the strike, three hundred of the miners, most of them immigrants from Chihuahua, settled on a piece of company property which they named Chihuahuaita. Here they built seventyfive or eighty shacks and determined to sit out the strike. Without notifying the occupants, the company sold the property in 1935 to a New Mexico politician. Unable to buy the lots at the prices demanded or to pay the rents which were fixed, they then faced a series of eviction suits. When the first of these eviction suits was brought on for hearing in the local court, a bloody riot ensued. Over a hundred miners were arrested and charged with a variety of offenses. For weeks Gallup was in a state of great excitement as writers and artists flocked in from Santa Fe and Taos to lend the miners a hand.3 During the excitement, Jesús Pallares, a miner from Chihuahua, organized the Liga Obrera de Habla Española which soon claimed eight thousand members. Pressure from this organization forced the authorities to abandon criminal syndicalism proceedings which were then pending against the rioters and finally won relief rights for the strikers. But Jesús Pallares met the fate of many Mexican immigrant labor leaders in the borderlands: he was arrested and deported to Mexico.

4. In the Copper Mines Prior to 1896, the copper miners of Arizona lacked organization. Resembling feudal baronies, the mining camps were an outgrowth of company towns some of which had existed prior to the American conquest. The copper mining companies had inherited a pre-1846 system of peonage in Arizona which they showed little inclination to change or to abolish. In the early seventies, when most of the great copper mines were discovered, common labor was paid 37½¢ a day in the Sonora mines, while furnace-workers received from 50¢ to 75¢ a day. On the American side of the border, according to Sylvester Mowry, the going-rate for Mexican labor was $1 a day, “paid in large part in

The Second Defeat   155

merchandise at large profits.” With a plentiful supply of Mexican labor so near at hand, the American mineowners naturally used Mexican immigrants as common laborers and, in an effort to avert unionization, pitted Mexicans against non-Mexicans. From 1875 to the present time, the copper companies have carried Mexican employees on their payrolls under a special heading of “Mexican Labor” and have paid them at lower rates than those commanded by non-Mexican labor for the same job classifications.4 Uniformly the Mexicans have been segregated in a separate section of the camp known to the AngloAmericans as “frogtown” or “jim-town.” Needless to say, the non-Mexican miners in the skilled labor classifications have always resented the encroachment of peonage which the appearance of Mexican immigrants has symbolized. In their eyes, the Mexicans have been potential scabs and strikebreakers whose employment at substandard wage rates has served as a constant threat to their jobs and wage standards. It was largely for this reason that certain copper-mining districts, particularly those at Bisbee and Prescott, were for many years “white men’s camps” with a tradition that Mexicans were not allowed to stay overnight. Because of this antagonism, the number of Mexicans employed has ranged from twenty percent in the Bisbee mines to fifty-three percent at Douglas, fifty percent at Morenci, and ninety-three percent at Nacozari. In fact, the first strike in the copper mines occurred in 1896, when the Western Federation of Labor struck at the Old Dominion mine against the employment of Mexican labor. Trade-union organization in the copper mines dates from the formation of the first trusts and consolidations. “As the copper mining activities grew in magnitude,” writes Will H. Robinson, “the close relationship between employer and employee that had obtained in the early days largely disappeared.” Troops were called out to suppress one of the first major organizational drives in 1903. In October 1915, three unions of Mexican miners, numbering about five thousand men, went on strike at the Clifton, Morenci, and Metcalf mines. This strike was largely called over the issue of the “Mexican rate” and against the tyrannous conduct of foremen who sold jobs to Mexicans and forced them to buy tickets in raffles as a condition to holding their jobs. The company sealed up the mouth of the mine with cement and told the strikers “to go back to Mexico.” Hundreds of miners were arrested and the National Guard was finally sent in to break the nineteen-week strike.5 On June 27, 1917, the Arizona copper miners, Mexican and non-Mexican, went on strike. After being out for a month, a vigilante mob rounded up 1,186 of the strikers and shipped them, in boxcars, to Columbus, New Mexico. The Columbus officials would not permit them to detrain so they were taken out and dumped in the desert. Investigating the strike some months later for the federal government, Felix Frankfurter reported that “too often there is a glaring inconsistency between our democratic purposes in this war abroad and

156  North from Mexico

the autocratic conduct of some at home.” In the aftermath of the famous “Bisbee deportations,” Walter Douglas, president of the Phelps-Dodge Company, was indicted in the federal court for his part in the deportations; but the charges were later dismissed. In 1944, the International Union of Mine, Mill and Smelter Workers, CIO, hauled three companies—the Miami Copper Company, the Inspiration Consolidated Copper Company, and the International Smelting and Refining Company—before the National War Labor Board on charges of discriminating against Mexican employees. The board found that these companies classified employees as “Anglo-American Males” and “Other Employees.” Included in the latter classification were all females, “Latin-Americans,” Negroes, Filipinos, and Indians. If a Mexican with no experience was hired, he was classified as a “common laborer” and paid $5.21 for a shift; but, if an AngloAmerican with no experience was hired, he was classified as a “helper” and paid $6.36 per shift. The board also found that “there was no apparent relationship between the length of service of an employee and his wage rate.” Workers classified in the “Other Employees” group who had been employed for ten years rarely received more than the starting rate for Mexican labor. “The problem with which the commission is confronted in these cases,” to quote from its decision, “is one which is woven into the fabric of the entire community, indeed of the entire Southwest. Unions and employers alike have had a part, and a significant part, in its creation and continuation. All the forms in which contemporary society is organized are in varying degrees affected.” In its decision, the board ordered the elimination of the discriminatory rates.6 Somewhat later, in a proceeding involving certain refineries of the Shell Oil Company in Texas, the President’s Fair Employment Practice Commission found that the company had two rates of pay for the same work: a “white” rate for Anglo-Americans, and a “non-white” rate, for Negroes and Mexicans. The differential was ten cents an hour. In Region X of the FEPC, embracing New Mexico, Texas, and Louisiana, thirty-seven percent of the complaints received involved discrimination against Mexicans; and in Region XII, comprising California, Nevada, and Arizona, 22.6 percent of the cases involved Mexicans. Following the history-making decision of the National War Labor Board, the copper miners struck throughout the Southwest in March 1946, largely because the companies had been stalling on compliance with the order of the board. “Before the war,” said one of the strikers, Manuel Fraijo, “all good jobs,— hoistman, mechanic, pumpman, and time-keeper,—were held by AngloAmericans. The low-bracket, common jobs were held by Mexicans. . . . When the war came, millions of men were drafted and sent overseas . . . the majority of those drafted from our town were ‘Mexicans.’ Curiously enough, a good percentage of these supposedly stupid Mexicans became pilots, radio operators, and radarmen and many became sergeants and lieutenants. . . . Now that

The Second Defeat   157

we are together again, if only some of these Aryan-minded ‘Anglos’ would discard their white supremacy theories, they would realize that we are just workers kept divided to be used against each other.” One of the leaders of this strike—he had organized Local No. 509 of the International Union of Mine, Mill and Smelter Workers in El Paso in 1940— was Humberto Silex. Entering the United States in 1921, Silex had served a term of enlistment in the army. Married, the father of seven children, he had resided in El Paso since 1931. Employed by the American Smelting and Refining Company, Silex had organized Local 509, which is largely made up of Mexican-Americans, and had been elected its first secretary; later its president. Discharged for union activity in 1945, he had a fist fight with his foreman and was arrested and fined $35. Like most immigrants in El Paso, Silex held a valid resident alien’s border-crossing card and frequently crossed the largely mythical border between El Paso and Juárez. After the strike was called in March 1946, he was arrested on a warrant of deportation based on the charge that he had committed a crime involving “moral turpitude”—namely the $35 fist fight—prior to his last entry into the United States—namely his last pleasure trip to Juárez. One of the outstanding Mexican labor leaders in the Southwest, Silex recently won his fight against deportation in the courts.

5. La Niña de Cabora Teresa de Cabora, as she was later known, was born in Sinaloa in 1872. An illegitimate child, her father was a wealthy Mexican rancher; her mother a Yaqui Indian. At the age of sixteen she went to live with her father, where a servant-woman, whose name was María, taught her to read. One day she went into a trance which, according to local legend, lasted three months and eighteen days. While in this state of trance, she is also supposed to have “cured” María of paralysis by a gentle laying on of hands. Later she effected many similar “cures” and her fame spread throughout Sinaloa and Sonora, where she went to live in the village of Cabora. Saint Teresa did not claim that she possessed the healing power; she merely said that she was willing to pray for the poor and the sick out of her love for them. This strange girl—“the Saint of Cabora” or “La Niña de Cabora” as she was called—was strongly anticlerical and preached against the Catholic Church which she said took money from the poor under false pretenses. With hundreds of Mexicans flocking to Cabora from points as distant as Hermisillo and Tomachic just to see and gaze upon La Niña, the hierarchy in Mexico City issued orders to the priests of Sonora to denounce her from the pulpit as a heretic. Later, at the prompting of the Church, Díaz dispatched a troop of soldiers to arrest La Niña de Cabora. Determined to protect their saint, the Yaqui Indians ambushed the expedition near Cabora. Thus started what was called

158  North from Mexico

the “Tomachic War” or the Revolt of the Yaquis. In the course of this “war,” Teresa was finally arrested and taken to Guaymas where she was imprisoned. To pacify the Yaquis, however, the governor was finally forced to release her. But constant bickering between La Niña and the priests resulted in the issuance of an order in 1893 for her deportation. Hounded out of Tomacacori and El Paso by the malicious denunciations of the Church, she moved to Clifton, Arizona, in 1897. In the eyes of the Mexican copper miners of Clifton, La Niña was a saint whose prayers and intercessions could heal the sick and restore sight to the blind. By them she was loved and worshipped as a miraculous border-counterpart of the Virgin of Guadalupe. When La Niña married a Yaqui Indian, she was separated from her husband by a mob of irate miners who looked upon the marriage as a sacrilege. Good Catholics that they were— and as this action proved—they ignored the interdictions of the priests and never ceased to worship the anticlerical Saint of Cabora. It was in Clifton that Teresa cured the daughter of a local Anglo-American banker who, in wonderment and gratitude, took her to Los Angeles that her healing powers might be more widely known. In the magical atmosphere of Our City the Queen of the Angels her fame, needless to say, was instantaneous and a purse was raised to send her on a tour of the United States and Europe. In 1904 she returned to Clifton and, with the money which she had earned on the tour, built a large two-story building which was used as a hospital. And in Clifton she died, eighteen months after her return, “with a smile on her lips.” Although the doctors said that she had died of tuberculosis, the Mexicans of Clifton insisted that she had worn out her spirit in the service of their people; to them she is still the Saint of Cabora.

6. The Forty Blonde Babies It was in Clifton, also, that the famous affair of the forty blonde babies occurred. The same year that La Niña de Cabora had built a hospital for the Mexicans of Clifton, a new priest, whose name was Father Mandin, came to the local Catholic church. One day the new priest received a routine letter from a New York foundling home, asking if there were any good Catholics in Clifton who might want to adopt children. New to Clifton and unfamiliar with his Mexican parishioners, Father Mandin naively read the letter at the next Sunday service. When asked if any of them desired to adopt children, almost the entire congregation, with audible enthusiasm, lifted their hands. Greatly pleased by this generous response, Father Mandin immediately wrote the hospital to send the children along but specified that only those of fair complexion should be sent since the Mexicans had insisted that they wanted “blonde babies.” On October first, forty children from eighteen months to five years of age arrived in Clifton accompanied by Sister Anna Michella. Each child carried a

The Second Defeat   159

tag which gave its name and birthday and the name of the adopting family. The entire Mexican population of Clifton went with Father Mandin to the station to greet Sister Anna Michella and the forty blonde children. Quite unprepared to cope with the situation, Sister Anna Michella vainly protested that the foster homes would first have to be inspected before she would release the children from her care. But the Mexicans, in a passion of enthusiasm, would not be put off with such technicalities and insisted on immediate delivery of their wonderful blonde wards. After a great deal of argument, Father Mandin upheld the position of his parishioners. Before a child was delivered to its foster parents, however, the parents had first to reimburse Sister Anna Michella for the cost of the clothes in which the children had made the journey and for their rail fare to Clifton. The families immediately paid over the sums stipulated and great enthusiasm prevailed in South Clifton that night as the Mexican community celebrated the arrival of the children. When word got around that the Mexicans had “bought forty blonde babies” a mob of three hundred angry Anglo-Americans assembled in downtown Clifton. Everyone agreed that the iniquity of the situation cried for redress and so a posse was formed which made the round of the Mexican homes collecting the children. When one Mexican miner suggested that a court order would be necessary before the custody of the children could be changed, a leader of the posse shoved a gun in his ribs and said: “Here’s your court order.” Once collected the forty children were taken to the local hotel. “It was raining,” writes the historian of Clifton, “and the crowd had swollen and they were in an angry mood. There was even talk of hanging the priest.” Some of the children were sick, probably from the excitement and the feasts of chili, beans, and tortillas which had been prepared for them. A few, the irate Anglo-Americans noted, even had the odor of beer on their breath. Before the night was over, the children had been parceled out to various residents of Clifton. Later the New York foundling home, bewildered by this strange mix-up in faraway Arizona, brought suit to recapture possession of the children but, in the meantime, adoption proceedings had been perfected, and the Supreme Court of Arizona upheld the adoptions in a decision which was affirmed by the U.S. Supreme Court. Great was the sorrow of the Mexican copper miners of Clifton when the beautiful blonde children, in their handsomely embroidered handstitched garments, were taken from them.7

7. The Battle of Cananea In the early years of the century, Ricardo Flores Magon and his brother Enrique, published a newspaper in Mexico City called Liberación. Sons of an Indian who had held office under Benito Juárez, they were outspoken critics of the Díaz regime. It was as a result of their activities that the First Liberal Congress was called in Mexico. At this meeting, the liberals declared themselves in favor

160  North from Mexico

of land distribution, the eight-hour day, the right to strike, and social insurance, all things for which, presumably, the people had fought a successful revolution. Díaz ignored the first liberal congress, but when the second was called in 1906, federal troops raided the meeting, the Magons were sent into exile, and their newspaper was suppressed. In exile the Magons established their headquarters in St. Louis where they published a newspaper called Regeneración, which called for a revolution against the Díaz regime, and where, on September 28, 1906, they organized the Mexican Liberal Party. From St. Louis copies of the newspaper were smuggled across the border and widely distributed in Mexico. Early in 1906, one of their lieutenants, Enrique Bermúdez, started a newspaper in Douglas, Arizona, which he called El Centenario. From Douglas, copies of both newspapers, and other propaganda issued by the Magons, was circulated among the copper miners in La Cananea, Sonora, where a liberal club was formed. In an effort to keep the peones on the haciendas, Díaz had fixed a maximum rate of pay for all nonagricultural employment. This policy fitted in perfectly, of course, with the policies of the Cananea Consolidated Copper Company which, like the companies with which it was affiliated on the American side of the border, consistently paid a lower wage to Mexican employees. That Mexicans should be subject to this type of discrimination in their own country naturally created a situation which the Magons could exploit to advantage. The American-owned Cananea mines employed between eight and nine thousand workers. On June 1, 1906, the company raised the wages of its Anglo-American employees but refused to raise those of its Mexican employees. The Mexican employees—two thousand in number—promptly struck in protest and before the three-day “Battle of Cananea” was over five Americans and thirty Mexicans were dead and property damage of $250,000 had resulted. Buildings were burned and stores were sacked, as armed Mexican strikers carried on a rifle duel with the company guards. Several hundred Anglo-Americans in Bisbee formed a volunteer company and marched across the border to relieve the beleaguered mine officials while Díaz rushed federal troops up from the south. The strike, of course, was broken and the employees were forced to return to work under the same conditions which had prevailed when the conflict began. The leaders of the strike who survived the battle were promptly sent to San Juan de Ulua, a prison fortress on a tiny island in the harbor of Veracruz. Some of them were still confined in the prison when the Díaz regime was finally overthrown. Until the Cananea strike nothing much in the way of a labor movement had existed in Mexico, for the country was still overwhelmingly agricultural and Díaz had suppressed whatever resistances labor had shown. The present-day Mexican labor movement, therefore, dates from the Battle of Cananea which has been called the first major labor strike in the history of Mexico.8 As a matter of fact, the labor movement in Mexico was largely initiated by returning

The Second Defeat   161

immigrants from the United States who had observed the ways of organized labor in this country.9

8. “The Wearers of the Red” When the Madero revolution was launched in Mexico, the Magons appeared in Chihuahua at the head of a band of a hundred armed men. However they were soon disillusioned with Madero who, as they stated in one of their manifestos, was “a millionaire who has seen his fabulous fortune grow with the sweat and tears of the peons of his hacienda.” On their refusal to accept Madero as the leader of the revolution, Orozco, Madero’s lieutenant, ordered them to leave Chihuahua. On this exile, they came to Los Angeles where they established their headquarters at 519½ East Fourth Street, now the center of a Mexican neighborhood. From this headquarters, they sent out an appeal from the “Organizing Junta of the Mexican Liberal Party of Los Angeles” for funds with which they could equip an army to pursue a truly Socialist revolution in Mexico. Los Angeles was “seething with social discontent,” as Emma Goldman put it, in these years; and a strong Socialist movement existed in the city. Local chapters of the Socialist Party and the I.W.W. helped the Magons raise money, sponsored the Mexican Liberal Party, and sent speakers to their meetings. In fact, the contributions are said to have averaged more than $1,000 a month. While they were raising money, the Magons were also recruiting an army for the invasion of Lower California. For commander-in-chief they selected Rhys Pryce, a British soldier of fortune. Admittedly the recruitment was amateurish. Single rifles were borrowed from their owners with the promise to return them, in good order, “when the revolution is over.” When the invasion was actually launched, the Magon Army had precisely one machine gun. Converging from New Mexico, Texas, Arizona, and California, the Magon forces, made up of Mexican immigrants, met at Calexico, on the American side of the border. Under the command of Pryce and three Mexicans named Leyva, Berthold, and Salinas, they crossed the border on May 8, 1910, took Mexicali without much fighting and then captured Tijuana. By the time they had captured these two towns, however, they had run out of ammunition. Once having possession of the customs fees at Mexicali, they thought they would be able to buy whatever arms were needed. But, alas! they collected only $850. All that Ricardo Magon could buy in the way of arms in Los Angeles, with this meager sum, were fifty rifles and some ammunition, which he rushed to the Magonistas in Tijuana. It wasn’t until the rifles were unloaded that Magon discovered that they were old, condemned U.S. Army Springfields which had been sold as souvenirs. The ammunition which he had purchased, moreover, would not fit the guns. After a few days, Tijuana was

162  North from Mexico

abandoned and the Magonistas began to drift back to Los Angeles and San Diego in small groups. In July of the same year they made another attempt to invade Lower California, but by then Madero had dispatched federal troops to the border and they were easily repulsed. Shortly after the second battle of Tijuana, the U.S. government stepped in. The Magons, together with the rest of the Organizing Junta, were arrested and placed in jail on charges of conspiring to violate the neutrality laws. Being in jail, however, did not dampen their revolutionary enthusiasm. They proceeded to name one Tirza de la Toba as general of the Magonistas and delegated him to carry on the revolution until their release. It was Tirza’s idea to buy arms in the United States and cache them in Mexico. Raising a force of two hundred men, he began to smuggle arms across the border. In the meantime, however, Congress passed an act in March 1912, making it illegal to export arms to Mexico and this gave the authorities a chance to go after De la Toba. Although he was never captured, his force was broken up and a large number of rifles and a large store of ammunition were seized. The trial of the Magons in Los Angeles lasted, in all its phases, from July 1910, to June 1912. At the end of the long trial, the brothers were convicted of breaching the neutrality laws of the United States and were sentenced, along with their coconspirators, to twenty-three months in a federal prison. The trial was a constant source of interest and intense excitement to the Mexican residents of Los Angeles who had known and loyally supported the Magon brothers. Later the prosecuting attorney told a Senate committee that “the courtroom was constantly filled by the followers of the Mexican Liberal Party . . . all Mexicans, the men and the women alike wearing the red badge of the Mexican Liberal Party in the courtroom. Every morning as we began the trial, we would turn around and face a solid phalanx of the wearers of the red.” Ricardo Flores Magon died in Leavenworth Prison. When his body was brought back to Mexico for burial, the Mexican Chamber of Deputies declared a recess in honor of his efforts on behalf of the Mexican labor movement and Mexicans along the border, from San Diego to Brownsville, mourned his passing. According to Walter Prescott Webb, the Texas-Mexicans had also contributed to his revolutionary efforts. Although Mexican immigrants had suffered a second defeat from Anglo-Americans in the borderlands, they had helped launch the movement that was to liberate Mexico.

11

“The Mexican Problem”

In the vast library of books and documents about ethnic and minority problems in the United States, one of the largest sections is devoted to “the Mexican Problem.” There is a curious consistency about the documents in this section. For one thing, the singular is always used. Presumably, also, no problem existed, singular or plural, prior to 1920. Readers’ Guide lists fifty-one articles on “the Mexican Problem” from 1920 to 1930 by comparison with nineteen articles on the same subject for the previous decade. When these articles are examined, it will be found that “the problem” apparently consists in the sum total of the voluminous statistics on Mexican delinquency, poor housing, low wages, illiteracy, and rates of disease. In other words, “the Mexican Problem” has been defined in terms of the social consequences of Mexican immigration. It will also be found that the documents devoted to the problem have been deeply colored by the “social work” approach. With the passage of the 1924 Immigration Act, the immigrant social agencies and Americanization institutes simply had to discover a new “problem” and it was the Mexican’s misfortune to appear on the scene, sombrero and all, concurrently with the impending liquidation of these agencies. As a consequence, he was promptly adopted as America’s No. 1 immigrant problem. The whole apparatus of immigrant-aid social work, with its morose preoccupation with consequences rather than causes, was thereupon transferred to Mexican immigration with little realization that this immigration might not be, in all respects, identical with European immigration. Once assembled and classified, this depressing mass of social data was consistently interpreted in terms of what it revealed about the inadequacies and the weaknesses of the Mexican character. The data “proved” that Mexicans lacked leadership, discipline, and organization; that they segregated themselves; that they were lacking in thrift and enterprise, and so forth. A mountainous collection of masters’ theses “proved” conclusively that Spanish-speaking children were “retarded” because, on the basis of various so-called intelligence tests, they did not measure up to the intellectual caliber of Anglo-American

164  North from Mexico

students. Most of this theorizing was heavily weighted with gratuitous assumptions about Mexicans and Indians. Paradoxically, the more sympathetic the writer, the greater seems to have been the implied condescension. All in all, the conclusion is unavoidable that Mexicans have been regarded as the essence of “the Mexican Problem.” The use of this deceptive, catchall phrase has consistently beclouded the real issues by focusing attention on consequences rather than on causes. Actually the basic issues have always had to do with Anglo-Hispano relations in a particular historical setting as influenced by a specific set of cultural, economic, geographical, and social forces. Once these factors are seen in proper perspective, if only in outline form, the elusive character of “the Mexican Problem” vanishes into thin air.

1. The Structure of the Problem In unraveling the real issues the first question to be raised is: what kind of a minority is the Mexican minority? Here, at the risk of being repetitious, I want to summarize several points. Unlike most European minorities in America, Mexicans have been rooted in space—in a particular region—over a long period of years. One of the important factors in “the problem” has always been their relation to, and their feeling about the region in which they are concentrated. As Dr. Carolyn Zeleny has pointed out, they are more like the typical minority in Europe than like the typical European minority in the United States. Mexicans were annexed by conquest, along with the territory they occupied, and, in effect, their cultural autonomy was guaranteed by a treaty. About the closest parallel that can be found in this hemisphere for the Mexican minority is that of the French-Canadians in Quebec. The parallel would be closer, of course, if the Province of Quebec were part of the United States. Then New Mexico could be regarded as the Quebec of the Mexicans and the million or so French-Canadians in the United States might be compared with the Mexican immigrants outside New Mexico. Like the Mexicans, the French-Canadians were “here first”; hence they have shown much the same tenacity about notre langue, notre foi, nos traditions that Mexicans have shown. With French-Canadians in the United States the question of la survivance is as important as the future of la raza is to most Mexicans. Like the French-Canadians in New England, the Spanish-speaking people know that they are Americans. Yet, as Dr. Campa points out, they never speak of themselves in Spanish as nosotros los americanos any more than they say nosotros los españoles. What a minority is called by others or how it likes to think of itself is less important than the way members of the minority actually speak of themselves in moments of “unbuttoned frankness.” In such moments, Mexican-Americans

“The Mexican Problem”   165

are likely to say nosotros, nuestra gente, la raza, or nosotros los mexicanos. But, as Dr. Campa carefully emphasizes, by mexicanos they do not mean Mexicans; nor can it be translated as such. Like the French-Canadians, and, I suppose, like all annexed or conquered peoples, the Mexicans have been deeply influenced by discrimination. French culture is indigenous to Quebec in much the same sense that Spanish culture is indigenous in New Mexico. Thus there is a time-factor and a space-factor involved in both situations which is not to be found in the usual European immigrant “problem” in America.1 The spatial relation of Mexico to the Southwest, the proximity of the border, the closeness of the parent group, are all important factors in “the Mexican Problem.” It should also be noted that relations between Anglos and Hispanos have been constantly influenced by the state of relations between the United States and Mexico. The assimilation of Italian immigrants might have assumed a different form, for example, if the United States and Italy had been involved in conflict for a hundred years. In the past, the attitude of Mexican consuls in the Southwest has been much more possessive and paternal than that of Italian consuls toward Italian immigrants. Historically the Southwest was once a part of Mexico—an obvious but all-important factor. Geographically the Southwest is one with the northern portions of Mexico and wars do not alter the facts of geography. Thus a specific set of historical and geographical factors are also very much a part of “the Mexican problem.” Furthermore, a unique set of cultural factors has been involved. Three cultures, not two, have fought for supremacy in the Southwest: Anglo, Hispano, and Indian. In fact, the three-sided relationship is so complex, interrelated, and historically interwoven as to defy analysis. Indians were a conquered race despised by Anglo-Americans. Mexicans are related to Indians by race and culture with the Indian part of their cultural and racial inheritance being more important than the Spanish. Mexicans were consistently equated with Indians by the race-conscious Anglo-Americans. Quite apart from the question of how much Indian blood flows in the veins of the Mexican minority, Mexicans are regarded as a racial minority in the Southwest. In the past, Indians exploited every tension between Anglos and Hispanos and each of the latter groups attempted to use the Indians against the other. This conflict has not died out. Native New Mexicans have continued to accuse the federal government of showing more concern for its treaty obligations with the Pueblo Indians than for its obligations under the Treaty of Guadalupe Hidalgo—and candor compels the admission that these complaints have been justified. Indians have recovered lost lands or been compensated for their loss and have received heavy financial subsidies from the federal government not granted Spanish-speaking people. But there is still another facet which distinguishes the Mexican problem from what have appeared to be similar issues.

166  North from Mexico

2. The Buffer Group In the United States a minority has long existed within the Mexican minority: the native-born of native-born parents. The census of 1930 estimated the size of this group as 264,338, although it is easily twice this size or larger. The attitude of the buffer group toward the immigrants and of the immigrants toward them has always been highly ambivalent. To the native-born, the immigrant is a cholo or chicamo; to the immigrant, the native-born is a pocho. The immigrant is likely to be “darker,” more Indian, than the native-born. The immigrant stresses his Mexican-Indian background; the native-born boasts of his “Spanish” inheritance in blood and culture. The immigrant is, also, more likely to be illiterate and to know less English. Despite the division between the two groups, however, Anglo-Americans regard them as one—as Mexican—except for ceremonial occasions when elements of the native-born become “Spanish.” On the other hand, the native-born seek to distinguish their status, in the eyes of the Anglo-Americans, by referring to themselves as Spanish-Colonials, Latin-Americans, Spanish-Americans, “native Californians,” and similar terms. While some of the native-born have “passed” completely into the AngloAmerican world, the majority have not been able to do so nor have they always wished to do so. Constant discrimination, which became more pronounced with the arrival of the immigrants, has complicated their existence and stiffened their resistance to absorption. The Anglo-Americans, in fact, have made it impossible for them to dissociate themselves, as a group, from the immigrants. Noting this fact, the immigrants have taunted the native-born with the mockery of their citizenship. Criticizing the native-born as renegades, they have derided their customs, morals, and affectations. Dr. Gamio quotes a popular corrido of the immigrants called “El Renegado”—“The Renegade,” which he translates as follows: You go along showing off In a big automobile. You call me a pauper And dead with hunger And what you don’t remember is That on my farm You went around almost naked And without sandals. This happens to many That I know here When they learn a little American And dress up like dudes,

“The Mexican Problem”   167

And go to the dance. But he who denies his race Is the most miserable creature. There is nothing in the world So vile as he The mean figure of the renegade. And although far from you, Dear Fatherland, Continual revolutions Have cast me out— A good Mexican Never disowns The dear fatherland Of his affections.*

(This corrido, incidentally, originated in Los Angeles.) Paradoxically, however, both groups regard themselves as members of la raza. They often live in the same districts; speak the same language; attend the same church; and frequently intermarry. Yet the distinction,—the cleavage,— remains. In some respects, the native-born occupy somewhat the same relation to the immigrants that “light” middle-class Negroes occupy to the masses of “dark” Negroes. The relationship between the two groups is, also, somewhat similar to that between German Jews and Russian Jews. It is a truism that the expectations which a dominant group hold forth influence the behavior and attitude of a minority. In this respect, the position of the native-born has been ambiguous. In some circles, they are expected to behave “like Mexicans”; elsewhere this expectation is reversed and the tactful assumption is made that they are “Spanish” or “American.” This ambiguity explains the conflicting attitudes which the native-born have toward “assimilation.” I know a successful young Mexican-American lawyer in Southern California—one of the few in the region—who takes a most extreme view toward “Americanization.” He believes that Mexicans should cut loose entirely from their Mexican background; that they should “mix” more with Anglo-Americans; and that they should, as he says, “quit beefing” about discrimination. But this individual has been highly favored by circumstances, background, and upbringing. It is probably true, as he contends, that he has encountered little discrimination (although he makes this point a little too emphatically); but other Mexican-Americans, “darker” than he is, less favored by circumstance, have * Manuel Gamio, “El Renegado” (The Renegade)” From Mexican Immigration to the United States, University of Chicago Press, 1930. Used by permission of University of Chicago Press.

168  North from Mexico

encountered consistent discrimination and are much less anxious to “mix” with Anglo-Americans. To some extent the two groups are separated by a “third” culture—that of the native-born. “The cultural contacts of the Mexican immigrants in the United States,” writes Dr. Gamio, “are complicated by the fact that besides the modern American civilization there is another and different MexicanAmerican culture, that of the Americans of Mexican origin. This civilization is American nominally, and exhibits the principal material aspects of modern American civilization, but intellectually and emotionally it lives in local Mexican traditions. This element [the American-born] can be said to constitute a peculiar nationality, within the United States. To the immigrant it is a sort of go-between, since these Mexican-Americans do not feel racial prejudice against them. Though a struggle occurs between the purely Mexican culture and this semi-Mexican, in the end it often absorbs the Mexican immigrant. With it there can occur a closer fusion than with the purely American culture, for with the latter it already shares many traits, while the great difference between the purely American and the purely Mexican, together with the fact of race prejudice, makes an intellectual, emotional, and traditional disparity too great to be bridged rapidly and perhaps never completely.”

3. The Conflict in Cultures The central plateau region of Mexico has fed immigrants to Texas; the northern mesa and northwestern coastal sections to California. Immigrants from these sections have had certain distinct handicaps. Most of them have been illiterate; a great many were peons in Mexico; and they have been extremely poor (actually undernourished, according to Dr. Gamio). In the main, also, they have come from a society which, prior to 1910, was calculated to rob individuals of a sense of enterprise, thrift, and initiative. Most Mexican immigrants have come to the United States from a folk culture. A folk culture, writes Dr. Robert Redfield,2 is a small, isolated, nonliterate, homogeneous society. Intimate communication among the members of the society is matched by a lack of communication with the exterior world. It is a society in which people have little access to the thought and experience of the past and in which “oral tradition has no check or competitor.” The people are much alike and have a strong sense of belonging together and the ways by which recurrent problems are solved have been conventionalized. Economi­ cally the folk society is independent of other societies: the people produce what they consume and consume what they produce. There is not much division of labor—one person doing what another does. The tools of production are few and simple. There are no tools to make tools; no rapid, multiple machine manufacture; little use of natural power. “Life,” writes Dr. Redfield, “for the member

“The Mexican Problem”   169

of the folk society, is not one activity and then another and different one; it is one large activity out of which one part may not be separated without affecting the rest.” Since behavior is strongly influenced by convention, there is little disposition to reflect upon traditional acts or to consider them objectively and critically. Behavior is personal, with even nature, the animals, and the environment being personalized and invested with human attributes. Obviously the members of such a society are not prepared for a rapid transition to a society which, at nearly every point, negates the values of their folk culture. In many areas of Mexico, the folk culture centered in the feudalistic hacienda which provided no opportunity for change in status. Ideas of justice were personalized, based on the whims and fancies of the haciendado. Money was meaningless; trade was limited; and the division of labor was simple. Superimposed on this folk society, the ceremonial aspect of the Catholic Church was emphasized somewhat to the detriment of its ethical teachings. Native folk practices were interwoven with church ritual and a “magical mentality” attributed illnesses to los aires or evil spirits. Dr. Gamio gives a long list of herbs which he found on sale in a Chicago drugstore that catered to Mexican immigrants; and one can still see a weird variety of herbs, leeches, and patent medicines on sale in the Mexican drugstores in Los Angeles. Slight wonder, then, that the Mexican peon faltered and became confused and often demoralized when he came in close contact with a highly industrialized, urban society. Uniformly his culturally conditioned traits have been interpreted in the Southwest as racial or biological. The Mexican was “lawless” and “violent” because he had Indian blood; he was “shiftless and improvident” because such was his nature; his excellence as a stoop-laborer consisted precisely in the fact that he did not aspire to land ownership. Point by point, his cultural traits reinforced the earlier stereotype of “the Mexican.” In the Southwest, the immigrant faced a set of formidable handicaps. A strong prejudice had existed in the region against Mexicans for many years; the tradition of dominance was interwoven into the fabric of the community; generations had been steeped in the Mexican stereotype. Almost by instinct, Anglo-Americans equated Mexicans with Indians. The language handicap would have been much less formidable had the immigrant been literate; but learning to read and write in English involved first learning to read and write in Spanish. Unskilled, in the American sense, the immigrant had little acquaintance with trade-unionism. Even his religion, in such muscularly Protestant states as Texas, served to set him apart. But his greatest handicap consisted in the migratory character of his employment. “One assimilates a new culture,” writes Dr. Norman Humphrey, “as one did the old one, largely through perception and imitation of examples.” Traveling over a wide territory, usually in the company of other Spanishspeaking workers, bossed by a Mexican foreman, living in a Mexican labor

170  North from Mexico

camp or shacktown, the immigrant had few chances to learn Anglo-American ways by example or imitation. The presence of a large buffer group of nativeborn Mexicans also retarded assimilation. The Mexican, moreover, was a late immigrant; “the last man in.” “Near the center of a culture,” writes Dr. Humphrey, “are the layers of meaning identified as values”; while, at the periphery, are the utilitarian symbols. When two sharply contrasting cultures come in contact the utilitarian symbols of each are brought into immediate juxtaposition: “utilitarian meaning competes with utilitarian meaning and, in the long run, the meaning having the greater utility supplants that which has the less.” Universally, Mexican immigrants, supposedly “incapable of assimilation,” have rapidly assimilated the utilitarian phases of Anglo-American culture. High on the list of items which 2,104 immigrants brought back to Mexico from the United States, according to Dr. Gamio, were such items as bathtubs; wooden and metal toilets; refrigerators; metal kitchen utensils; washing machines; metal stoves; sewing machines; and automobiles (thirty-seven percent returned with cars). But where the values of the two cultures have been in juxtaposition, the immigrant has been less willing to abandon or to modify the imported cultural pattern. “Spanish speech is retained,” writes Dr. Humphrey, “and la raza is esteemed.” Similarly Dr. Gamio has found that the intellectual culture of Mexico has continued to exert a great influence among Mexican-Americans; that, where values are concerned, they prefer to remain Mexicans. It should be noted, however, that these conclusions were based on studies made in the Southwest. Throughout the Southwest immigrants have been drawn within the folds of existing colonies and opportunities to learn “by perception and imitation,” on an individual basis, have been minimized. Immigrants are more limited in their choice of residence, employment, and associations than in the northern industrial communities where a different pattern of acculturation prevails. Persistent discrimination has repelled the immigrant from the value-side of Anglo-American culture.

4. The Pattern of Employment The basic factor retarding the assimilation of the Mexican immigrant, at all levels, has been the pattern of his employment. A very large proportion of Mexican immigrants were imported, often under contract, by particular employers, for employment in particular industries at particular tasks. With few exceptions, only a particular class of employers has employed Mexican labor in the Southwest: large-scale industrial enterprises; railroads; smelters; copper mines; sugar-beet refineries; farm-factories; large fruit and vegetable exchanges. These concerns have employed many Mexicans, in gangs, crews, and by families as in the sugar-beet industry. It is not the individual who has been

“The Mexican Problem”   171

employed but the group. If a concern employs Mexicans, it will usually be found that they dominate or are used exclusively in specific types of employment rather than being scattered through the plant. The universality of this pattern was clearly established in a study made in California in 1930. In this same study it was found that the jobs for which Mexicans were employed en masse had certain basic characteristics: they were undesirable by location (as section-hand jobs on the desert sections of the rail lines or unskilled labor in desert mines and cement plants); they were often dead-end types of employment; and the employment was often seasonal or casual. Between 1914 and 1919 the number of Mexicans in the citrus industry in California increased from 2,317 to 7,004 (thirty percent of the total); today some 22,000 Mexicans are employed. In effect, Mexicans work, not for individual citrus growers, but for the California Fruit Growers Exchange. The exchange bears about the same relationship to “farming” that the typical industrial plant in which Mexicans are employed bears to “business”: it is highly organized; it represents an enormous capital investment; and it is an enterprise which provides no ladder of advancement for field and packing-house employees. One could count on the fingers of one hand the number of Mexicans who have become owners of citrus groves or who have risen to managerial positions in the exchange. To keep Mexicans earmarked for exclusive employment in a few large-scale industries in the lowest brackets of employment, their employers have set them apart from other employees in separate camps, in company towns, and in segregated colonias. Traditionally, Mexicans have been paid less than AngloAmericans for the same jobs. These invidious distinctions have reinforced the Mexican stereotype and placed a premium on prejudice. By employing large numbers of Mexicans for particular types of work, employers have arbitrarily limited the immigrants’ chance for the type of acculturation that comes from association with other workers on the job. The pattern of employment has, in turn, dictated the type and location of residence. Segregated residential areas have resulted in segregated schools; segregated schools have reinforced the stereotype and limited opportunities for acculturation. In setting this merrygo-round in motion, the pattern of employment has been of crucial importance for it has stamped the Mexican as “inferior” and invested the stereotype with an appearance of reality. “There are people,” writes Bogardus, “who insist on thinking that the Mexican is unable to rise above an unskilled labor level. They cannot visualize a Mexican immigrant on any other plane.” The pattern of employment has consistently fostered prejudice by jeopardizing or appearing to threaten, the standards of the trade-unions. Always opposed to Mexican immigration, the American Federation of Labor has permitted many of its affiliates to bar Mexicans from membership. Exclusion from trade unions has, of course, closed another avenue of escape from the

172  North from Mexico

merry-go-round and provided a further sanction for the stereotype. By keeping Mexicans segregated occupationally, employers have created a situation in which the skilled labor groups have naturally regarded Mexicans as group competitors rather than as individual employees. The nature of the situation has thus inclined such groups as the AFL to take a narrow, particularistic view of Mexican immigration and to regard Mexican labor as “cheap labor.” In some areas, as in west Texas, it is also apparent that the use which has been made of Mexican labor has tended to drive out Anglo-American small farmers and tenants. With the labor of the small farmer and tenant being necessarily in competition with the paid labor of the large-scale farm, cheap agricultural wage rates have been a powerful factor working toward concentration in farm ownership and production. In the various congressional hearings on Mexican immigration, small farmers were invariably lined up with organized labor in opposition to Mexican immigration. While the conflict has always been economic, it has consistently been rationalized as racial or cultural in character. The far-reaching ramifications of the pattern of employment can scarcely be overemphasized. In the citrus belt communities, the California Fruit Growers Exchange has long exercised a decisive influence on local affairs. It has been in a position to influence—and has not hesitated to influence—local school-board policies and to affect the attitude of the police, the courts, and the townspeople. When Mexican workers have gone on strike, the townspeople have generally been arrayed against them. Therefore it is patently nonsensical to regard segregated schools for Mexicans as a more or less “natural” outgrowth of “differences,” racial or cultural in character, between Anglos and Hispanos. A copper town is dominated by the mine ownership and management; a sugar-beet town reflects the attitude and policies of the sugar-beet refineries, etc.

5. The Colonia Complex Scattered throughout Southern California outside Los Angeles are, perhaps, 150,000 or 200,000 Mexicans and Mexican-Americans, for the most part immigrants or the sons and daughters of immigrants. Approximately thirty percent of the total is made up of “aliens” but the alien element is rapidly diminishing. Most of these people—perhaps eighty percent of them—live in “colonies” or colonias which vary in size from a cluster of small homes or shacks to communities of four, five, six, eight, and ten thousand people.3 The history of these settlements is almost uniformly the same. They came into existence some twenty or thirty years ago when the first immigrants began to arrive. Most of them are located in unincorporated areas adjacent to a town or city but invariably on “the other side” of something: a railroad track, a

“The Mexican Problem”   173

bridge, a river, or a highway. Site location has been determined by a combination of factors: low wages, cheap rents, low land values, prejudice, closeness to employment, undesirability of the site, etc. None of the colonies was laid out or planned as a community, although a few are located on the sites of abandoned “boom towns.” Some are outgrowths of labor camps; others have been grafted on a pre-1900 barrio; while a few have come into existence more or less accidentally. For example, the settlement known as Hick’s Camp came into existence thirty-three years ago when a river-bottom camp was washed out by a flood. The health authorities and the Red Cross moved the families to the river bank where a squatter camp grew up because the land was cheap. Nowadays completely surrounded, the colonia in San Gabriel is located near the old Mission—one of the few cases where a Mexican settlement is to be found at the center of an Anglo-American community. North Town, a community near Upland, is a fairly typical colonia. Located on the site of an abandoned subdivision, it is within fifteen minutes’ driving radius of the wineries, packing houses, truck farms, and citrus groves where most of the residents are employed. Here a few Mexican families lived before the great wave of migration began and to these residents the immigrants attached themselves. Today some 1,500 Mexicans live in the six square blocks of North Town surrounded, on all sides, by agricultural land. North Town has a small grocery store; a pool hall; and a motion picture theater. Most of the residents, however, make their purchases in Upland. Two or three blocks from the village is an elementary school in which the enrollment is ninety-five percent Mexican. With as many as three shacks to a lot, the structures are unpainted, weatherbeaten, and dilapidated. The average house consists of two or three rooms and was built of scrap lumber, boxes, and discarded odds-and-ends of material. Ten, twenty, and thirty years old, the houses are extremely clean and neat on the inside and much effort has obviously gone into an effort to give them an attractive appearance. Virtually all the homes lack inside toilets and baths and a large number are without electricity. Almost every family owns an automobile, a radio, and any number of American-made household gadgets of one kind or another. Being unincorporated, almost all forms of municipal service are lacking. Water is purchased from a private owner at rates higher than those paid by the conspicuously successful residents of Upland. North Town is one of dozens of similar colonias scattered all the way from Santa Barbara to San Diego. Occasionally the colonia is part of an incorporated town or city with the Mexican population comprising from twelve to twenty-five percent or more of the total population. It would be misleading, however, to convey the impression that the location of the colonias was accidental or that it has been determined by the natural play of social forces. On the contrary, there is a sense in which it would be

174  North from Mexico

accurate to say that the location of the colonias has been carefully planned. Located at just sufficiently inconvenient distances from the parent community, it naturally became most convenient to establish separate schools and to minimize civic conveniences in the satellite colonia. “Plainly,” writes Fred W. Ross, “it was never intended that the colonias were to be a part of the wider community; rather, it was meant that they were to be apart from it in every way; colonia residents were to live apart, work apart, play apart, worship apart, and unfortunately trade, in some cases, apart.” The physical isolation of the colonias has naturally bred a social and psychological isolation. As more and more barriers were erected, the walls began to grow higher, to thicken, and finally to coalesce on all sides. The building of the walls, as Mr. Ross puts it, “went on concomitantly from without and from within the colonia, layer by layer, tier by tier.” While the walls may have the appearance of being natural growths, they are really man-made. For the relationship that finally emerged between parent and satellite community is the civic counterpart of the relationship between the California Fruit Growers Exchange and its Mexican employees. Living in ramshackle homes in cluttered-up, rundown shacktowns, set apart from their neighbors, denied even the minimum civic services, the residents of the colonia have come to resent the fenced-in character of their existence. They are perfectly well aware of the fact that they are not wanted, for their segregation is enforced by law as well as by custom and opinion. That the colonias lack swimming pools might be explained in terms of the ignorance or indifference of the Anglo-Americans were it not for the revealing circumstance that Mexicans are also denied access to municipal plunges in the parent community. Hence the ostracism of the Mexicans cannot be accounted for in the facile terms in which it is ordinarily rationalized. When public-spirited citizens in the parent community have sought “to do something about the Mexican Problem,” they have generally sought to impose a pattern on the colonia from without. Establishing a clinic or reading-room or social center in the colonia has no doubt been helpful; but it has not changed, in the slightest degree, the relationship between parent and satellite community. In the face of this reality, it is indeed annoying to hear Anglo-Americans expatiate about the Mexicans’ “inferiority complex” and to charge them with being clannish and withdrawn. Friendly, warm-hearted, and generous to a fault, it would be difficult to find a people more readily disposed to mingle with other groups than the Spanish-speaking people of the Southwest. Their “inferiority complex” is really a misnomer for a defeatist attitude arising from their frustration at being unable to break out of the colonia. Resenting the implication of inferiority that attaches to segregated schools and being well aware of economic discrimination, a majority of the youngsters have not bothered to transfer from the segregated elementary school to the

“The Mexican Problem”   175

usually nonsegregated high school. Dropping out of school at the eighth grade level, they have been unable to compete successfully with Anglo-Americans for the more desirable jobs and have fallen back on those for which their fathers were imported. According to the census of 1930, only 5,400 Mexicans were to be found in clerical jobs; 1,092 were teachers; 93 were lawyers and judges; and 165 were physicians and surgeons—this in a population of close to three million people. Once the cycle of employment has been repeated in the second and third generation, writes Mr. Ross, “the insidious process, which began so long ago with low wages and relatively low, dominant group hostility, almost swings full circle.” By the time this has happened, the hostility of the dominant group is fully reciprocated.”4 Hedged in by group hostility, the immigrants long ago lost interest in citizenship. Lack of funds, the language difficulty, and illiteracy were important factors but not nearly as influential as segregation and discrimination. Mexicans have never been encouraged, by prevailing community attitudes, to become citizens. Bogardus, who studied the problem years ago, concluded that in both rural and urban areas segregation was primarily responsible for the lack of interest in citizenship. For the last twenty years, the number of Mexicans who have been naturalized has averaged about a hundred a year. In a Mexican community of fifty thousand in California, Bogardus found only 250 registered voters in 1928, not all of whom were of Mexican descent. In the same year, Charles A. Thompson reported that only two or three naturalization petitions a year were filed in El Paso with a Mexican population of fifty thousand. To some extent, of course, this reluctance to seek naturalization may be traced to the fact that so many Mexican immigrants are in the United States illegally; but this, too, has been a secondary factor. Voluntary disenfranchisement, whatever the cause, has perpetuated the caste-like social structure in which Mexicans are encased. The second generation, however, has begun to show a lively interest in the ballot. Residents of a few citrus belt settlements have, in recent years, elected Mexican-Americans to school boards and city councils and have begun to exercise a measure of their great potential political strength. Wherever they have “come of age” politically, an immediate change has been noted in the attitude of the Anglo-Americans. Anglo-American politicians cannot afford to ignore the needs of Mexican-American communities if the residents will assert their political rights. Acting in liaison with the well-organized Negro community in Los Angeles, Mexicans could easily become a balance-of-power group. While a few political victories have been won, it requires no special insight to foresee that a point will soon be reached when a serious struggle will develop between Anglos and Hispanos. The average Anglo-American community will accept, if somewhat reluctantly, one Mexican-American on the city council or the school board; but there are communities in which Mexican-Americans

176  North from Mexico

could elect a majority of the officeholders. In these communities, resistances will stiffen for the stakes are high. Once this has happened, Mexican-Americans will have to seek out allies in those segments of the Anglo-American community which are now disposed to cooperate with them, namely, in the liberallabor-progressive groups. By comparison with Negroes, Mexicans are still novices in the tactics and strategy of minority-group action and politics.

6. The Northern Settlements In the Middle Western industrial centers, Mexicans have been brought into much sharper and fuller contact with Anglo-American culture than in the Southwest. Here the colony is strikingly similar to that of the typical “foreign” settlement. Much less mobile than their compatriots in the Southwest, Mexicans in Chicago and Detroit work with members of other nationality groups in highly mechanized industries. The boundaries of the colonia are not sharply defined and, in some cases, have already disappeared. Since nearly one-third of the “northern” Mexicans have been solos or single men, the rate of intermarriage has been higher than in the Southwest. Originally concentrated in packing plants, tanneries, steel mills, foundries, and railroad yards, Mexican labor is today more widely and more typically distributed. Generally speaking, Mexicans are less sharply set apart in the Midwest industrial centers than in the Southwest. In Chicago and Detroit, Mexicans are merely another immigrant group; in the Southwest they are an indigenous people. The tendency to regard Mexicans as a “racial minority” is much less pronounced in the Midwest and there is less discrimination. As might be expected, therefore, a much higher proportion have applied for citizenship and English tends to be substituted for Spanish as the language of the home. The lack of cohesion and unity in these colonies is reflected in many ways. For example, Archbishop Mooney in Detroit has strongly discouraged the development of group-consciousness among his Mexican parishioners. Priests have been forbidden to give any encouragement to the idea of a church especially for Mexicans and have been warned that no racial or nationality distinctions, so far as Mexicans are concerned, will be tolerated. Perhaps no one detail points up the contrast between these communities and those in the Southwest more sharply than Dr. Humphrey’s comment that in Detroit Mexicans refer to themselves simply as “Mexicans” and show little sensitivity to the term. The story of the Lorain, Ohio, colony is quite typical of the Midwest settlements which nowadays total around 75,000 Mexicans. In 1923 the National Tube Company, an affiliate of United States Steel Corporation, imported 1,500 Mexicans from Texas to replace an equal number of Negroes (throughout the Midwest, Mexicans have been used to “dilute” or “thin out” Negro labor). From time to time, the colony was augmented by new recruits and by replacements

“The Mexican Problem”   177

drawn to Lorain from the beet fields of the Midwest. At first most of the Mexicans lived in the boxcars in which they had traveled north but most of them have since moved into small homes and apartments. Originally employed by National Tube or the Baltimore and Ohio Railroad, many of them have now secured jobs in restaurants, dry-cleaning shops, trucking firms, and other miscellaneous occupations. The homes in Lorain reflect a striking mixture of the two cultures.“American radios,” writes Robert O. O’Brien, “are covered with zarapes and bits of Indian pottery. Stone metates grind out corn which is cooked on gas and even on electric stoves. American phonographs play South American tangos and Mexican marches. Mexican trunks contain a mixture of objects from Gringo Sunday clothes to old country sombreros. Corona typewriters in vivid colors compete for space with bits of cactus from the Southwest. Bottles of medicine from Lorain doctors vie with patent medicines or Mexican ‘teas’ for position on the bathroom shelf. . . . American ‘canned’ food is supplemented by enchiladas, chili verde, and tamales.” A Lorain merchant sold thirty-six typewriters to Mexican residents in a year, all but two of them being equipped with a Spanish language keyboard. Here the second generation is already far removed from the first and the parents are vainly seeking to arrest the process by attempting to “Mexicanize” their children. It is a foregone conclusion that the northern Mexican settlements will have largely vanished in another generation.

7. Que Maravilla! The oldest settlers in Los Angeles, Mexicans were pushed aside and swept under by the extraordinary velocity and volume of Anglo-American migration after the first great “land booms” in the eighties. Isabel Sherrick, a Middle Western journalist, reported in the 1880s that the Mexicans “little by little are being crowded out and one by one the adobes are falling into ruins or giving way to the thrifty homes of Americans.” Some of the sections in which Mexicans formerly lived are today occupied by factories, terminal facilities, and office buildings. The typical residence of Mexicans in early-day Los Angeles was the “house court” derived from the Mexican vecindad: a sort of tenement made up of a number of one- and two-room dwellings built around a court with a common water supply and outdoor toilets. This same type of settlement, similar to the plaza, is still quite common in Los Angeles, San Antonio, and El Paso. Housecourts multiplied in Los Angeles as the demand for Mexican housing became acute with high land costs and rising rents. In 1916 the city had 1,202 housecourts, occupied by 16,000 people with 298 house-courts being occupied exclusively by Mexicans.5 In some respects, the house-court was not unlike the “bungalow courts” of a later period. The house-court areas quickly became

178  North from Mexico

slums as the city pushed westward from its original center in the old Plaza section. One of the first studies of Mexican housing conditions indicated that some twenty or thirty thousand Mexicans were living in the courts of Old Sonoratown, near the Plaza, in the shacks and houses of Chavez Ravine, and similar areas, and in the railroad labor camps. The houses and courts had dirt floors; wood was used for fuel; there were no bathing facilities; and the outdoor hydrant and toilet, used by a group of families, were universal. Made in 1912, this survey is still up-to-the-minute so far as Mexican housing is concerned, for little improvement has occurred in the last thirty-five years. When the great wave of Mexican immigration reached Los Angeles, an unincorporated section on the “east side” known as Belvedere became the principal area of “first settlement” for most of the immigrant families. “Que Maravilla!” the immigrants exclaimed when they first arrived in Los Angeles: what a marvel! what a wonderful city! Maravilla was their name for Belvedere and Maravilla it still is to thousands of Mexicans. With a Mexican population of fifty thousand in the middle twenties, the Belvedere section has a population today, mostly Mexican, of around 180,000. A city this large in size, it is still governed by remote control as an unincorporated area. Aside from Maravilla, Mexicans are nowadays scattered in “pockets” of settlement in Los Angeles. While they are not segregated as rigidly as Negroes, the various pocket-settlements are almost exclusively Mexican and are, if anything, more severely isolated than the colonias of the outlying sections. The “pockets” are all similar in character—Chavez Ravine, Happy Valley, El Hoyo (The Hollow), and the rest. Chavez Ravine, located in the hills between Elysian Park and North Broadway, is an old Mexican settlement. Shacks cling precariously to the hillsides and are bunched in clusters in the bottom of the ravine. For forty years or more, the section has been without most of the ordinary municipal services. At various points in the ravine, one can still see large boards on which are tacked the rural mailboxes of the residents—as though they were living, not in the heart of a great city, but in some small rural village in the Southwest. Goats, staked out on picket lines, can be seen on the hillsides; and most of the homes have chicken pens and fences. The streets are unpaved; really trails packed hard by years of travel. Garbage is usually collected from a central point, when it is collected, and the service is not equal to that which can be obtained in Anglo districts bordering the ravine. The houses are old shacks, unpainted and weather-beaten. Ancient automobile bodies clutter up the landscape and various “dumps” are scattered about. The atmosphere of the ravine, as of El Hoyo and the other pocket-settlements, is ancient, antiquated, a survival—something pushed backward in time and subordinated. One can make a swift turn off the heavy traffic of North Figueroa or North Broadway and be in Chavez Ravine in a minute’s time. In this socially regressive dead-end, goats bleat and roosters crow and children play in the dirt

“The Mexican Problem”   179

roads. Were it not for the faraway hum of traffic, a visitor might well imagine that he was in some remote village in New Mexico or Arizona. From the City Hall to Chavez Ravine is a five-minute drive by modern traffic-time; sociologically, the two points are separated by a time-span of between fifty and seventy-five years. Today a great modern highway span is being built over the Hollow. Bulldozers have moved in and houses have been jacked-up and lifted out of the way. The shacks not directly in the way of the juggernaut mechanical progress of the city are now left perched on the sides of the Hollow, thirty years old, still badly in need of paint, gradually falling apart. Thousands of motorists will rush over the new span every hour, traveling so fast that they will probably not even notice that they are passing over the remains of what was once a small Mexican village. At 720 San Vicente Boulevard, near the intersection of San Vincente and Santa Monica—on the “west side” of Los Angeles—is an ironic little island of Mexicans completely surrounded by middle-class residences many of which have been built in the so-called “Spanish-Colonial” style with white stucco walls, patios, and red-tiled roofs. This “island” is a thirty-year-old Pacific Electric labor camp where forty Mexican families live as they might live in a village in Jalisco. The company has generously provided four “outside” showers for 120 residents. It has even provided them with “hot water” on Mondays, Wednesdays, and Saturdays! The only facilities for washing clothes or dishes consist of outside sinks, detached from the shacks in the court, and used by all the families. Probably not one percent of the people who live in the surrounding areas know or have ever heard of the camp’s existence. What the Mexican immigrants probably think of Maravilla today is suggested by one of their best-known corridos—El Enganchado* (literally, “the hooked-one”—the labor contractor): I came under contract from Morelia To earn dollars was my dream, I bought shoes and I bought a hat And even put on trousers. For they told me that here the dollars Were scattered about in heaps; That there were girls and theaters And there here everything was good fun. And now I’m overwhelmed— I am a shoemaker by trade * Quoted from Mexican Labor in the United States by Dr. Paul S. Taylor.

180  North from Mexico

But here they say I’m a camel And good only for pick and shovel. What good is it to know my trade If there are manufacturers by the score, And while I make two little shoes They turn out more than a million. Many Mexicans don’t care to speak The language their mothers taught them And go about saying they are Spanish And deny their country’s flag. Some are darker than chapote** But they pretend to be Saxon; They go about powdered to the back of the neck And wear skirts for trousers. The girls go about almost naked And call la tienda “estor” They go around with dirt-streaked legs But with those stockings of chiffon. Even my old woman has changed on me— She wears a bob-tailed dress of silk, Goes about painted like a piñata** And goes at night to the dancing hall. My kids speak perfect English And have no use for our Spanish They call me “fader” and don’t work And are crazy about the Charleston. I’m tired of all this nonsense I’m going back to Michoacan; As a parting memory I leave the old woman To see if someone else wants to burden himself.

** chapote— black tar; piñata—a gaily-colored container.

12

The Pattern of Violence

In March 1942, the Japanese were excluded from the West Coast and the remaining citizens found, rather to their surprise, that this drastic wartime measure had not solved all their social and economic problems as the more rampant West Coast newspapers had led them to believe. Problems which had existed before the Japanese exclusion still existed, intensified by the war activities which involved most of Southern California. In Los Angeles, where fantasy is a way of life, it was a foregone conclusion that Mexicans would be substituted as the major scapegoat group once the Japanese were removed. Thus within a few days after the last Japanese had left, the Los Angeles newspapers, led by the Hearst press, began to play up “Mexican” crime and “Mexican” juvenile delinquency, as though the Mexican element in crime and delinquency could be considered apart from the ordinary crime experienced by a large, congested metropolitan area in wartime. A number of minor incidents in the spring of 1942 enabled the newspapers and the police to build up, within the short period of six months, sufficient anti-Mexican sentiment to prepare the community for a full-scale offensive against the Mexican minority. Once prepared, of course, this sentiment could be expected to assume violent expression with the first major incident. A young Mexican who had been arrested and sentenced to forty-five days in jail for having accosted a woman was, upon his release, taken before the Grand Jury and, if you please, reindicted for rape, on the same offense, and promptly sentenced to prison for twelve years! The case was quickly appealed and, of course, the conviction was reversed. A short time later, a group of Mexican men, celebrating a wedding, were arrested for playing a penny crap game, an offense usually ignored by the police as being inconsequential. But, in this instance, a “conspiracy” indictment was secured from the Grand Jury, thereby neatly converting a petty misdemeanor into a felony charge. On July 13, 1942, the press gave great prominence to a story involving a fight between two groups of Mexican boys, the Belvedere “gang” and the Palo Verde “gang.” In all these preliminary “incidents” pointed mention was made of the “Mexican”

182  North from Mexico

character of the people involved. By these techniques, the groundwork was carefully prepared for the “big incident.”

1. The Case of Sleepy Lagoon On the afternoon of August 1, 1942, Henry Leyvas, a young Mexican-American, had taken his girl for a drive near a little pond in a gravel pit near what was called the Williams Ranch on the east side of Los Angeles. In lieu of other recreational facilities, this abandoned gravel pit had long been used by Mexican youngsters in the neighborhood as a swimming pool. Early that evening, a Saturday night, Leyvas and his girl had been set upon by members of a rival “gang” and a fight had occurred. (Leyvas himself was a member of a group known as the 38th Street “gang.”) Later, the same evening, Leyvas returned to the gravel pit with members of his own gang, in several cars, to look for the troublemakers. Some of the members of this sortie knew that Leyvas intended “to get even,” but others merely went along for the ride and a swim and a general good time. Finding the gravel pit deserted, they discovered that a party was in progress at a nearby house belonging to the Delgadillo family and decided “to crash the gate.” Some fighting and scuffling occurred at the Delgadillo home and the invaders, after a time, left the scene of the party. Early on the morning of August second, the body of young José Díaz was picked up from a dirt road near the Delgadillo house and taken to the General Hospital where Díaz died without ever regaining consciousness. The autopsy showed that he had met his death as the result of a fracture at the base of the skull. He had apparently been in a fight for his hands and face were bruised but there were neither knife nor gun wounds on the body. The road where his body was found was well traveled and the autopsy showed that he was probably drunk at the time of his death. Díaz had left the Delgadillo home with two friends—presumably the last persons to have been with him prior to his death. Never called as witnesses, their version of what happened to Díaz is not known. The autopsy surgeon, it should be noted, testified that Díaz could have met his death by repeated hard falls on the rocky ground of the road and admitted that the injuries at the base of his skull were similar to those seen on the victims of automobile accidents. Such are the facts of the case. With the prior background in mind, it is not surprising that the Los Angeles press welcomed the death of Díaz like manna from the skies. Around the essentially bare facts of the case, they promptly proceeded to weave an enormous web of melodramatic fancy. The old gravel pit was dubbed “The Sleepy Lagoon” by a bright young reporter and the whole case was given an air of sordid mystery. Quick to cooperate, the police rounded up twenty-four youngsters, all alleged to be members of the 38th Street “gang,” and charged them

The Pattern of Violence   183

with the murder of Díaz. Two of them had the wit to demand separate trials and the charges against them were later dropped. But to a fantastic orchestration of “crime” and “mystery” provided by the Los Angeles press, seventeen of the youngsters were convicted in what was, up to that time, the largest mass trial for murder ever held in the county. In the process of “investigating” the case, the police severely beat up two of the boys. While testifying that he had been beaten by the police, one of the boys was shown a photograph by the prosecution. This photograph had been taken of him, purportedly, just prior to his entering the Grand Jury room, and indicated that, at that time, he was unmarked and unbeaten. The boy then pulled from his pocket a photograph which had been taken by a newspaper photographer just as he was leaving the Grand Jury room. This untouched photograph showed him with severe bruises about the head and face. Anna Zacsek, attorney for Leyvas, testified that she had walked into a room at the police station where her client, handcuffed to a chair, was being beaten by the police, and that she found him barely conscious, smeared with his own blood. Held incommunicado while they were being “worked over” by the police, the defendants were then marched, en masse, to the Grand Jury which proceeded to indict the lot of them for first-degree murder. When they appeared before the Grand Jury they were dirty, haggard, bruised—a thoroughly disreputableappearing group of youngsters completely terrified by the treatment they had just received. Who were these “criminals”—these hardened “gangsters”? Henry Leyvas, twenty, worked on his father’s ranch. Chepe Ruiz, eighteen, a fine amateur athlete, wanted to play big league baseball. In May 1942, his head had been cracked open by the butt of a policeman’s gun when he had been arrested on “suspicion of robbery,” although he was later found not guilty of the charge. In San Quentin Prison, where he and the others were sent after their conviction in the Sleepy Lagoon case, Ruiz won the admiration of the warden, the prison staff, and the inmates when he continued on in a boxing match, after several of his ribs had been broken. Robert Telles, eighteen, was working in a defense plant at the time of his arrest. Like many Mexican youngsters on the east side, he had remarkable skill as a caricaturist and amused his codefendants during the trial by drawing caricatures of the judge, the jury, and the prosecutor. Manuel Reyes, seventeen, had joined the navy in July 1942, and was waiting induction when arrested. Angel Padilla, one of the defendants most severely beaten by the police, was a furniture-worker. Henry Ynostrosa, eighteen, was married and the father of a year-old girl. He had supported his mother and two sisters since he was fifteen. Manuel Delgado, nineteen, also a woodworker, was married and the father of two children, one born on the day he entered San Quentin Prison. Gus Zamora, twenty-one, was also a furnitureworker. Victor Rodman Thompson, twenty-one, was an Anglo youngster who, by long association with the Mexican boys in his neighborhood, had become

184  North from Mexico

completely Mexicanized. Jack Melendez, twenty-one, had been sworn into the navy before he was arrested. When a dishonorable discharge came through after his conviction, he said it was “like kicking a guy when he’s down.” John Matuz, twenty, had worked in Alaska with the U.S. Engineers. These, then, were the “criminals,” the “baby gangsters,” the “murderers” who provided Los Angeles with a Roman holiday of sensationalism, crimemongering, and Mexican-baiting. From the very outset, a “gang” was on trial. For years, Mexicans had been pushed around by the Los Angeles police and given a very rough time in the courts, but the Sleepy Lagoon prosecution capped the climax. It took place before a biased and prejudiced judge (found to be such by an appellate court); it was conducted by a prosecutor who pointed to the clothes and the style of haircut of the defendants as evidence of guilt; and was staged in an atmosphere of intense community-wide prejudice which had been whipped up and artfully sustained by the entire press of Los Angeles. From the beginning the proceedings savored more of a ceremonial lynching than a trial in a court of justice. The defendants were not allowed to sit with their counsel—there were seven defense attorneys—and were only permitted to communicate with them during recesses and after adjournment. For the first weeks of the trial, the defendants were not permitted to get haircuts and packages of clean clothes were intercepted by the jailer on orders of the prosecutor. As a consequence of this prejudicial order, the defendants came trouping into the courtroom every day looking like so many unkempt vagabonds. Following a trial that lasted several months and filled six thousand pages of transcript, they were convicted on January 13, 1943: nine were convicted of second-degree murder plus two counts of assault and were sentenced to San Quentin Prison; others were convicted of lesser offenses; and five were convicted of assault and sentenced to the county jail. Following the conviction, the Sleepy Lagoon Defense Committee was formed which raised a large fund to provide new counsel and to appeal the case. I served as chairman of this committee and Harry Braverman, a member of the Grand Jury who had tried to stop the indictment, served as its treasurer. On October 4, 1944, the District Court of Appeals, in a unanimous decision, reversed the conviction of all the defendants and the case was later dismissed “for lack of evidence.” In its decision, the court sustained all but two of the contentions which our defense committee had raised, castigated the trial judge for his conduct of the trial, and scored the methods by which the prosecution had secured a conviction. On October twenty-fourth, when the charges were finally dismissed after the defendants had served nearly two years in San Quentin Prison (we had been unable to provide bonds during the appeal), hundreds of Mexicans crowded the corridors of the Hall of Justice to greet the boys. “Hysterical screams and shrieks,” reported the Los Angeles Times,

The Pattern of Violence   185

“laughter and cries of jubilation welled from the crowd. The atmosphere was electric with excitement as the liberated men were besieged by well-wishers who enthusiastically pumped their hands and slapped their backs. Tears flowed unashamedly.” For the first time in the history of Los Angeles, Mexicans had won an organized victory in the courts and, on this day, bailiffs and deputy sheriffs and court attachés were looking rather embarrassed in the presence of Mexicans. The work of the Sleepy Lagoon Defense Committee received nationwide attention and was hailed as an important contribution to the war effort by exPresident Cardenas of Mexico and by the Mexican consul-general. In Mexico City, the magazine Hoy devoted a three-page spread in its issue of September 30, 1944, to the work of the defense committee. During the time the committee was in existence, we received hundreds of letters from GIs, from posts in Guam, New Guinea, Hawaii, the Fiji Islands, the Aleutians; in fact, from all over the world. Soldiers with names like Livenson, Hart, Shanahan, Hecht, Chavez, Scott, Bristol, Cavouti, and Burnham, enclosed dimes, quarters, and dollars for the work of the committee. Marine Corps Captain M. A. Cavouti wrote us from New Guinea: “This war is being fought for the maintenance and broadening of our democratic beliefs and I am heartily in accord with any effort to apply these principles by assisting in obtaining a review of this case. Please accept my modest contribution.” From Hawaii, Corporal Samuel J. Foreman, a Negro, wrote: “I saw in the Pittsburgh Courier that you were leading the fight for victims of aggression. We members of the colored race are sympathetic to your worthwhile and moral fight to free these Mexican boys.” Dozens of letters came from Mexican-Americans in the service. Everyone liked what we had done except, of course, the dominant cliques in Los Angeles. Since the initial suggestion for the formation of the committee had come from LaRue McCormick, a member of the Communist Party, we were systematically red-baited. The press accused us of “inciting racial prejudice,” scoffed at the charge of bias during the trial, and lauded the trial judge and the prosecutor. Even the unanimous decision of the District Court of Appeals, sustaining the charges we had made, failed to bring so much as a mumbled retraction of the accusations that had been made against the boys or so much as a grudging acknowledgment that we had been right. While the case was pending on appeal, several members of the committee, including myself, were subpoenaed by the Committee on Un-American Activities in California, headed by Senator Jack Tenney, and grilled at great length. Naturally these various grillings were reported in the press in a manner calculated to make it most difficult for us to raise money for the appeal. The assistant district attorney, who conducted the prosecution, threatened the First Unitarian Church of Los Angeles with the removal of its tax-exempt status if it permitted the committee to hold a meeting on its premises. In fact, permission

186  North from Mexico

to hold the meeting, was, at the last minute, revoked by the church in response to this pressure. That I had expressed opposition to segregation and had testified that I was opposed on principle to miscegenation statutes was actually cited by the Tenney Committee on page 232 of its report as proof (!) of Communistic inclinations! As a postscript to this section, I should add that not long after his release from prison Henry Leyvas was convicted of a criminal offense after receiving a fair trial. So far as Leyvas was concerned, he had been convicted of being a Mexican long years ago and the damage was done. Needless to say, his general morale and attitude were not improved by his experiences in the Sleepy Lagoon case.

2. Captain Ayres: Anthropologist To appreciate the social significance of the Sleepy Lagoon case, it is necessary to have a picture of the concurrent events. The anti-Mexican press campaign which had been whipped up through the spring and early summer of 1942 finally brought recognition, from the officials, of the existence of an “awful” situation in reference to “Mexican juvenile delinquency.” A special committee of the Grand Jury, shortly after the death of José Díaz, was appointed to investigate “the problem.” It was before this committee, within two weeks after the arrest of the defendants in the Sleepy Lagoon case, that Captain E. Duran Ayres, chief of the “Foreign Relations Bureau” of the Los Angeles sheriff ’s office, presented a report presumably prepared under the instructions of his superiors. “Mexicans as a whole, in this county,” reads the report, “are restricted in the main only to certain kinds of labor, and that being the lowest paid. It must be admitted that they are discriminated against and have been heretofore practically barred from learning trades. . . . This has been very much in evidence in our defense plants, in spite of President Roosevelt’s instructions to the contrary. . . . Discrimination and segregation, as evidenced by public signs and rules, such as appear in certain restaurants, public swimming plunges, public parks, theaters, and even in schools, cause resentment among the Mexican people. . . . There are certain parks in this state in which a Mexican may not appear, or else only on a certain day of the week. There are certain plunges where they are not allowed to swim, or else only on one day of the week, and it is made evident by signs reading to that effect, for instance, ‘Tuesdays reserved for Negroes and Mexicans.’ . . . Certain theaters in certain towns either do not allow Mexicans to enter, or else segregate them in a certain section. Some restaurants absolutely refuse to serve them a meal and so state by public signs. . . . All this applies to both the foreign and American-born Mexicans.” So far, in the report, Captain Ayres was simply drawing a true picture of conditions in Los Angeles County. But, since his real purpose was “to explain”

The Pattern of Violence   187

the causes of Mexican juvenile delinquency, he soon began to draw some extraordinary conclusions. “The Caucasian,” he went on to report, “especially the Anglo-Saxon, when engaged in fighting, particularly among youths, resort to fisticuffs and may at times kick each other, which is considered unsportive: but this Mexican element considers all that to be a sign of weakness, and all he knows and feels is a desire to use a knife or some lethal weapon. In other words, his desire is to kill, or at least let blood. That is why it is difficult for the Anglo-Saxon to understand the psychology of the Indian or even the Latin, and it is just as difficult for the Indian or the Latin to understand the psychology of the Anglo-Saxon or those from northern Europe. When there is added to this inborn characteristic that has come down through the ages, the use of liquor, then we certainly have crimes of violence.” (Emphasis added.) This passage should, perhaps, be compared with similar conclusions drawn by another amateur anthropologist. “Race,” wrote Adolf Hitler, “does not lie in the language but exclusively in the blood. A man may change his language without any trouble but . . . his inner nature is not changed.” The close agreement between these two experts was shown after the publication of the Ayres Report when Radio Berlin, Radio Tokyo, and Radio Madrid quoted passages from the report to show that Americans actually shared the same doctrines as those advocated by Hitler. The Los Angeles sheriff, who had previously made much fuss over his “Latin blood” and his “early California background,” was sufficiently embarrassed by these broadcasts to suggest to a reporter from the New York Daily News that the Japanese, upon being evacuated, had incited the Mexican population of Los Angeles to violence. Thus the sheriff, who had always identified himself with the Mexican population on Cinco de Mayo and the Sixteenth of September, inferentially charged that the Mexicans, his own people, had become agents of the Japanese government! In considering the subsequent pattern of events, it is important to remember that the Ayres Report had been formally presented to the Grand Jury by the sheriff and had presumably represented the official views, candidly expressed, of law enforcement officers in Los Angeles. Thus the chief law enforcement agency in the county had given voice to the view that the Mexican minority possessed an inborn tendency to criminal behavior and to crimes of violence. Being primarily men of action, the law enforcement officials proceeded to act in accordance with this belief. Essentially, therefore, there is nothing incredible about their subsequent behavior and conduct.

3. Plotting a Riot If one spreads out the span of one’s right hand and puts the palm down on the center of a map of Los Angeles County with the thumb pointing north, at the tip of each finger will be found a community where the population is

188  North from Mexico

predominantly Mexican. In each of these neighborhoods, moreover, a majority of the juveniles living in the area will be found to be first-generation MexicanAmericans, sons and daughters of the Mexican immigrants who came to Southern California during the 1920s. Now, if one believes that Mexicans have an inherent desire to commit crimes of violence, the logical first step, in a crime prevention program, is to arrest all the people living in these areas. Unfortunately for the practice of this cozy little theory, there are well over a hundred thousand people living in these areas who are of Mexican descent. The maximum capacity of the Los Angeles jails being somewhat under this figure, it therefore becomes necessary to proceed on a more selective basis. If one group of Mexicans, say, the young people, could be selected for token treatment, and if sufficient arrests could be made from this group, perhaps this would serve as an example to all Mexicans to restrain their inborn criminal desires. . . . If this sounds a bit fantastic, consider the following letter which Captain Joseph Reed sent to his superior on August 12, 1942: C. B. Horrall, Chief of Police. Sir: The Los Angeles Police Department in conjunction with the Sheriff, California Highway Patrol, the Monterey, Montebello, and Alhambra Police Departments, conducted a drive on Mexican gangs throughout Los Angeles County on the nights of August 10th and 11th. All persons suspected of gang activities were stopped. Approximately 600 persons were brought in. There were approximately 175 arrested for having knives, guns, chains, dirks, daggers, or any other implement that might have been used in assault cases. . . . Present plans call for drastic action . . . . Respectfully, Joseph f. Reed Administrative Assistant (Emphasis added.)

On the nights in question, August 10 and 11, 1942, the police selected the neighborhoods which lay at our fingertips on the maps and then blockaded the main streets running through these neighborhoods. All cars containing Mexican occupants, entering or leaving the neighborhoods, were stopped. The occupants were then ordered to the sidewalks where they were searched. With the occupants removed, other officers searched the cars for weapons or other illicit goods.

The Pattern of Violence   189

On the face of it, the great raid was successful, for six hundred people were arrested. The charges? Suspicion of assault, suspicion of robbery, suspicion of auto thefts, suspicion of this, suspicion of that. Of the six hundred taken into custody, about 175 were held on various charges, principally for the possession of “knives, guns, chains, dirks, daggers, or any other implement that might have been used in assault cases.” This is a broad statement, indeed, but it is thoroughly in keeping with the rest of this deadly serious farce. For these “other” implements consisted, of course, of hammers, tire irons, jack handles, wrenches, and other tools found in the cars. In fact, the arrests seem to have been predicated on the assumption that all law-abiding citizens belong to one or another of the various automobile clubs and, therefore, do not need to carry their own tools and accessories. As for those arrested, taking the names in order, we have, among those first listed, Tovar, Marquez, Perez, Villegas, Tovar, Querrero, Holguín, Rochas, Aguilera, Ornelas, Atilano, Estrella, Saldana, and so on. Every name on the long list was obviously either Mexican or Spanish and therefore, according to the Ayres Report, the name of a potential criminal. The whole procedure, in fact, was entirely logical and consistent once the assumptions in the report were taken as true. Harry Braverman, a member of the Grand Jury who had opposed returning the indictment in the Sleepy Lagoon case, was greatly disturbed by these mass dragnet raids and by the manner in which the Grand Jury was being used as a sounding board to air the curious views of Captain Ayres. Accordingly, he arranged for an open Grand Jury hearing on October 8, 1942, at which some of the damage caused by the Ayres Report might, if possible, be corrected. At this hearing, Dr. Harry Hoijer of the University of California; Guy T. Nunn of the War Manpower Commission (who later wrote, on his return from a German prison camp, a fine novel about Mexican-Americans called White Shadows); Manuel Aguilar of the Mexican consulate; Oscar R. Fuss of the CIO; Walter H. Laves of the Office of the Coordinator of Inter-American Affairs and myself all endeavored to create in the minds of the Grand Jurors at least a doubt that everything that Captain Ayres had said was true. To appreciate the incomparable irony of this situation, suffice it to say that here we were having to defend “the biological character” of the Mexican people months after Mexico had declared war on Germany, Italy, and Japan on May 22, 1942; after the first shipment of 1,500 Mexican workers—the vanguard of an army of 100,000 workers that Mexico sent to this country during the war—had arrived in California on September 29, 1942; and after Henry Wallace, then vice-president of the United States, had declared to a great Sixteenth of September celebration in Los Angeles that “California has become a fusion ground for the two cultures of the Americas. . . .” On the occasion of this hearing, representatives of the coordinator of InterAmerican Affairs made the rounds of the newspapers, calling attention to the

190  North from Mexico

serious harm being done the war effort and the Good Neighbor Policy by the newspaper campaign against resident Mexicans. In the interest of winning the war, these officials had suggested, there might well be some abatement in this campaign: we were fighting the Germans and the Japanese, not the Mexicans. With stated reluctance, and obvious misgivings, the newspapers promised to behave and, from October to December 1942, the great hue and cry either disappeared from the press or was conducted sotto voce. That the campaign had seriously interfered with the war effort, there can be no doubt. When the Sleepy Lagoon defendants were convicted, for example, the Axis radio beamed the following message in Spanish to the people of Latin America: In Los Angeles, California, the so-called City of the Angels, twelve Mexican boys were found guilty today of a single murder and five others were convicted of assault growing out of the same case. The 360,000 Mexicans of Los Angeles are reported up in arms over this Yankee persecution. The concentration camps of Los Angeles are said to be overflowing with members of this persecuted minority. This is justice for you, as practiced by the “Good Neighbor,” Uncle Sam, a justice that demands seventeen victims for one crime. (Axis broadcast, January 13, 1943)

The representatives of the Coordinator’s Office urged the newspapers in particular to cease featuring the word “Mexican” in stories of crime. The press agreed, but, true to form, quickly devised a still better technique for baiting Mexicans. “Zoot suit” and “Pachuco” began to appear in the newspapers with such regularity that, within a few months, they had completely replaced the word “Mexican.” Any doubts the public may have harbored concerning the meaning and application of these terms were removed after January 13, 1943, for they were consistently applied, and only applied, to Mexicans. Every Mexican youngster arrested, no matter how trivial the offense and regardless of his ultimate guilt or innocence, was photographed with some such caption as “Pachuco Gangster” or “Zoot-suit Hoodlum.” At the Grand Jury hearing on October 8, 1942, some of us had warned the community that, if this press campaign continued, it would ultimately lead to mass violence. Bur these warnings were ignored. After the jury had returned its verdict in the Sleepy Lagoon case and Mr. Rockefeller’s emissaries had left Los Angeles, the campaign, once again, began to be stepped up. On the eve of the Zoot Suit Riots in Los Angeles, therefore, the following elements were involved: first, the much-publicized “gangs,” composed of youths of Mexican descent, rarely over eighteen years of age; second, the police, overwhelmingly non-Mexican in descent, acting in reliance on the theories of Captain Ayres; third, the newspapers, caught in a dull period when there was only a major war going on, hell-bent to find a local scapegoat, “an

The Pattern of Violence   191

internal enemy,” on which the accumulated frustrations of a population in wartime could be vented; fourth, the people of Los Angeles, Mexican and nonMexican, largely unaware that they were sponsoring, by their credulity and indifference, a private war; and, fifth, the men of the armed services stationed in or about the city, strangers to Los Angeles, bored, getting the attitudes of the city from its flamboyant press. They entered the plot, however, only at the climax. Knowing already of the attitude of the police and of the press, let’s examine the Mexican “gang.”

4. The Origin of Pachuquismo In Los Angeles, in 1942, if a boy wished to become known as a “gangster” he had a choice of two methods. The first, and by far the more difficult, was to commit a crime and be convicted. The second method was easier, although it was largely restricted to a particular group. If you were born of Mexican parents financially unable to move out of certain specific slum areas, you could be a gangster from birth without having to go to all the trouble of committing a crime. For Los Angeles had revised the old saying that “boys will be boys” to read “boys, if Mexican, will be gangsters.” The only reservation to be noted, of course, consists in the definition of a “gang.” Adolescent boys in the United States are among the most gregarious groups in our society. American boys traditionally “hang out with the gang.” Their association is based, of course, on common interests. The boys in the “gang” may go to the same school, live in the same neighborhood, or have the same hobbies. There is, however, a difference in the degree to which the members of various “gangs” feel a sense of solidarity. A boy who belongs to a club for those who make model airplanes may have little loyalty toward the club. It serves a particular interest and beyond this interest he must have other associations. But a “gang” of Mexican boys in Los Angeles is held together by a set of associations so strong that they outweigh, or often outweigh, such influences as the home, the school, and the church. The various teenage clubs in the better parts of Los Angeles often get together and spend an evening dancing in Hollywood. But the respectable places of entertainment will often refuse to admit Mexicans. The boys and girls who belong to the “Y” often make up theater parties. But the “best” theaters in Los Angeles have been known to refuse admission to Mexicans. Many youngsters like to go roller-skating or ice-skating; but the skating rink is likely to have a sign reading “Wednesdays reserved for Negroes and Mexicans.” Wherever the Mexicans go, outside their own districts, there are signs, prohibitions, taboos, restrictions. Learning of this “iron curtain” is part of the education of every Mexican-American boy in Los Angeles. Naturally it hits them hardest at the time when they are trying to cope with the already tremendous problems of

192  North from Mexico

normal adolescence. The first chapters are learned almost on the day they enter school, and, as time passes and the world enlarges, they learn other chapters in this bitter and peremptory lesson. Most of the boys are born and grow up in neighborhoods which are almost entirely Mexican in composition and so it is not until they reach school age that they become aware of the social status of Mexicans. Prior to entering school, they are aware, to a limited extent, of differences in background. They know that there are other groups who speak English and that they will some day have to learn it, too. But it is at school that they first learn the differences in social rank and discover that they are at the bottom of the scale. Teachers in the “Mexican” schools are often unhappy about their personal situation. They would much rather be teaching in the sacrosanct halls of some Beverly Hills or Hollywood school. Assignment to a school in a Mexican district is commonly regarded, in Los Angeles, as the equivalent of exile. Plagued by teachers who present “personality problems,” school administrators have been known to “solve” the problem by assigning the teacher to “Siberia.” Neither in personnel nor equipment are these schools what they should be, although a definite attempt to improve them is now under way. Discovering that his status approximates the second-rate school has the effect of instilling in the Mexican boy a resentment directed against the school, and all it stands for. At the same time, it robs him of a desire to turn back to his home. For the home which he knew prior to entering school no longer exists. All of the attitudes he has learned at school now poison his attitude toward the home. Turning away from home and school, the Mexican boy has only one place where he can find security and status. This is the gang made up of boys exactly like himself, who live in the same neighborhood, and who are going through precisely the same distressing process at precisely the same time. Such is the origin of the juvenile gangs about which the police and the press of Los Angeles were so frenetically concerned. Gangs of this character are familiar phenomena in any large city. In Los Angeles, twenty years ago, similar gangs were made up of the sons of Russian Molokan immigrants. They have existed in Los Angeles since the city really began to grow, around 1900, and they will continue to exist as long as society creates them. Thus “the genesis of pachuquismo,” as Dr. George Sanchez has pointed out, “is an open book to those who care to look into the situations facing Spanish-speaking people” in the Southwest. In fact, they were pointed out over a decade ago in an article which Dr. Sanchez wrote for the Journal of Applied Psychology.1 The pachuco gang differs from some other city gangs only in the degree to which it constitutes a more tightly knit group. There is more to the pachuco gang than just having a good time together. The pachucos suffer discrimination together and nothing makes for cohesiveness more effectively than a commonly shared hostility. Knowing that both as individuals and as a group they

The Pattern of Violence   193

are not welcome in many parts of the city, they create their own world and try to make it as self-sufficient as possible. While the fancier “palladiums” have been known to refuse them, even when they have had the price of admission, there are other dance halls, not nearly so fancy, that make a business of catering to their needs. It should be noted, however, that Mexican boys have never willingly accepted these inferior accommodations and the inferior status they connote. Before they have visited the “joints” on Skid Row, they have first tried to pass through the palatial foyers on Sunset Boulevard. When they finally give up, they have few illusions left about their native land. It should also be remembered that pachuquismo followed a decade of important social change for Mexicans in Los Angeles. During the depression years, thousands of Mexicans had been repatriated and those remaining began to adjust to a new mode of existence. The residence of those who had been migratory workers tended to become stabilized, for residence was a condition to obtaining relief. Thousands of Mexicans were replaced, during these same years, by so-called Okies and Arkies in the migratory labor movement. A greater stability of residence implied more regular schooling, better opportunities to explore the intricacies of urban life, and, above all, it created a situation in which the Mexican communities began to impinge on the larger Anglo-American community. During the depression years, one could watch the gradual encroachment of Mexicans upon downtown Los Angeles. Stores and shops catering to Mexican trade crossed First Street, moving out from the old Plaza district and gradually infiltrated as far south as Third or Fourth streets. The motion picture theaters in this neighborhood, by far the oldest in the city, began to “go Mexican” as did the ten-cent stores, the shops, and the small retail stores. Nowadays the old Mason Opera House, in this district, has become a Mexican theater. Being strangers to an urban environment, the first generation had tended to respect the boundaries of the Mexican communities. But the second generation was lured far beyond these boundaries into the downtown shopping districts, to the beaches, and above all, to the “glamour” of Hollywood. It was this generation of Mexicans, the pachuco generation, that first came to the general notice and attention of the Anglo-American population. Thus concurrently with the growth of the gangs there developed a new stereotype of the Mexican as the “pachuco gangster” the “zoot-suiter.” Many theories have been advanced and reams of paper wasted in an attempt to define the origin of the word “pachuco.” Some say that the expression originally came from Mexico and denoted resemblance to the gaily costumed people living in a town of this name; others have said that it was first applied to border bandits in the vicinity of El Paso. Regardless of the origin of the word, the pachuco stereotype was born in Los Angeles. It was essentially an easy task to fix this

194  North from Mexico

stereotype on Mexican youngsters. Their skin was enough darker to set them apart from the average Angeleno. Basically bilingual, they spoke both Spanish and English with an accent that could be mimicked by either or both groups. Also there was an age-old heritage of ill-will to be exploited and a social atmosphere in which Mexicans, as Mexicans, had long been stereotyped. The pachuco also had a uniform—the zoot suit—which served to make him conspicuous. Mexican-American boys never use the term “zoot suit,” preferring the word “drapes” in speaking of their clothes. “Drapes” began to appear in the late thirties and early forties. In general appearance, “drapes” resemble the zoot suits worn by Negro youngsters in Harlem, although the initiated point out differences in detail and design. Called “drapes” or “zoot suit,” the costume is certainly one of the most functional ever designed. It is worn by boys who engage in a specific type of activity, namely, a style of dancing which means disaster to the average suit. The trouser cuffs are tight around the ankles in order not to catch on the heels of the boy’s quickly moving feet. The shoulders of the coat are wide, giving plenty of room for strenuous arm movements; and the shoes are heavy, serving to anchor the boy to the dance floor as he spins his partner around. There is nothing esoteric about these “sharp” sartorial getups in underprivileged groups, quite apart from their functional aspect. They are often used as a badge of defiance by the rejected against the outside world and, at the same time, as a symbol of belonging to the inner group. It is at once a sign of rebellion and a mark of belonging. It carries prestige.2 For the boys, peg-topped pants with pleats, high waists up under the armpits, the long loose-backed coat, thick-soled bluchers, and the duck-tailed haircut; for the girls, black huaraches, short black skirt, long black stockings, sweater, and high pompadour. Many of the boys saved their money for months to buy one of these getups. The length of the coat and the width of the shoulders became as much a mark of prestige as the merit badges of the Boy Scout. But, it should be noted, that the zoot suit was not universal among Mexican boys. Some never adopted it, while others never adopted it completely. There were all varieties of acceptance. The newspapers, of course, promptly seized upon the zoot suit as “a badge of crime.” But as one zoot-suited boy said to me, with infallible logic, “If I were a gangster, would I wear a zoot-suit so that everyone would know I was a gangster? No, I’d maybe dress like a priest or like everyone else; but no zoot-suit.” With the backdrops all in place, the curtain now rolls up on an interesting tableau in Our City the Queen of the Angels which was founded in the year 1781 by Mexican pobladores under the direction of Spanish officers who wore costumes far more outlandish than those worn by the most flamboyant pachucos.

13

Blood on the Pavements

On Thursday evening, June 3, 1943, the Alpine Club—made up of youngsters of Mexican descent—held a meeting in a police substation in Los Angeles. Usually these meetings were held in a nearby public school but, since the school was closed, the boys had accepted the invitation of a police captain to meet in the substation. The principal business of the meeting, conducted in the presence of the police captain, consisted in a discussion of how gang-strife could best be avoided in the neighborhood. After the meeting had adjourned, the boys were taken in squad cars to the street corner nearest the neighborhood in which most of them lived. The squad cars were scarcely out of sight, when the boys were assaulted, not by a rival “gang” or “club,” but by hoodlum elements in the neighborhood. Of one thing the boys were sure: their assailants were not of Mexican descent. Earlier the same evening a group of eleven sailors, on leave from their station in Los Angeles, were walking along the 1700 block on North Main Street in the center of one of the city’s worst slum areas. The surrounding neighborhood is predominantly Mexican. On one side of the street the dirty brick front of a large brewery hides from view a collection of ramshackle Mexican homes. The other side of the street consists of a series of small bars, boarded-up storefronts, and small shops. The area is well off the beaten paths and few servicemen found their way this far north on Main Street. As they were walking along the street, so they later stated, the sailors were set upon by a gang of Mexican boys. One of the sailors was badly hurt; the others suffered minor cuts and bruises. According to their story, the sailors were outnumbered about three to one. When the attack was reported to the nearest substation, the police adopted a curious attitude. Instead of attempting to find and arrest the assailants, fourteen policemen remained at the station after their regular duty was over for the night. Then, under the command of a detective lieutenant, the “Vengeance Squad,” as they called themselves, set out “to clean up” the gang that had attacked the sailors. But—miracle of miracles!—when they arrived at the scene

196  North from Mexico

of the attack they could find no one to arrest—not a single Mexican—on their favorite charge of “suspicion of assault.” In itself this curious inability to find anyone to arrest—so strikingly at variance with what usually happened on raids of this sort—raises an inference that a larger strategy was involved. For the raid accomplished nothing except to get the names of the raiding officers in the newspapers and to whip up the anger of the community against the Mexican population, which may, perhaps, have been the reason for the raid. . . . Thus began the so-called “Zoot Suit Race Riots” which were to last, in one form or another, for a week in Los Angeles.

1. The Taxicab Brigade Taking the police raid as an official cue—a signal for action—about two hundred sailors decided to take the law into their own hands on the following night. Coming down into the center of Los Angeles from the Naval Armory in Chavez Ravine (near the “Chinatown” area), they hired a fleet of twenty taxicabs. Once assembled, the “task force” proceeded to cruise straight through the center of town en route to the east side of Los Angeles where the bulk of the Mexicans reside. Soon the sailors in the lead-car sighted a Mexican boy in a zoot suit walking along the street. The “task force” immediately stopped and, in a few moments, the boy was lying on the pavement, badly beaten and bleeding. The sailors then piled back into the cabs and the caravan resumed its way until the next zoot-suiter was sighted, whereupon the same procedure was repeated. In these attacks, of course, the odds were pretty uneven: two hundred sailors to one Mexican boy. Four times this same treatment was meted out and four “gangsters”—two seventeen-year-old youngsters, one nineteen, and one twenty-three—were left lying on the pavements for the ambulances to pick up. It is indeed curious that in a city like Los Angeles, which boasts that it has more police cars equipped with two-way radio than any other city in the world (Los Angeles Times, September 2, 1947), the police were apparently unable to intercept a caravan of twenty taxicabs, loaded with two hundred uniformed, yelling, bawdy sailors, as it cruised through the downtown and east-side sections of the city. At one point the police did happen to cross the trail of the caravan and the officers were apparently somewhat embarrassed over the meeting. For only nine of the sailors were taken into custody and the rest were permitted to continue on their merry way. No charges, however, were ever preferred against the nine. Their evening’s entertainment over, the sailors returned to the foot of Chavez Ravine. There they were met by the police and the Shore Patrol. The Shore Patrol took seventeen of the sailors into custody and sent the rest up to the ravine to the Naval Armory. The petty officer who had led the expedition,

Blood on the Pavements   197

and who was not among those arrested, gave the police a frank statement of things to come. “We’re out to do what the police have failed to do,” he said; “we’re going to clean up this situation. . . . Tonight [by then it was the morning of June fifth] the sailors may have the marines along.”1 The next day the Los Angeles press pushed the war news from the front page as it proceeded to play up the pavement war in Los Angeles in screaming headlines. “Wild Night in L.A.—Sailor Zooter Clash” was the headline in the Daily News. “Sailor Task Force Hits L.A. Zooters” bellowed the Herald-Express. A suburban newspaper gleefully reported that “zoot-suited roughnecks fled to cover before a task force of twenty taxicabs.” None of these stories, however, reported the slightest resistance, up to this point, on the part of the Mexicans. True to their promise, the sailors were joined that night, June fifth, by scores of soldiers and marines. Squads of servicemen, arms linked, paraded through downtown Los Angeles four abreast, stopping anyone wearing zoot suits and ordering these individuals to put away their “drapes” by the following night or suffer the consequences. Aside from a few half-hearted admonitions, the police made no effort whatever to interfere with these heralds of disorder. However, twenty-seven Mexican boys, gathered on a street corner, were arrested and jailed that evening. While these boys were being booked “on suspicion” of various offenses, a mob of several hundred servicemen roamed the downtown section of a great city threatening members of the Mexican minority without hindrance or interference from the police, the Shore Patrol, or the Military Police. On this same evening, a squad of sailors invaded a bar on the east side and carefully examined the clothes of the patrons. Two zoot-suit customers, drinking beer at a table, were peremptorily ordered to remove their clothes. One of them was beaten and his clothes were torn from his back when he refused to comply with the order. The other—they were both Mexicans—doffed his “drapes” which were promptly ripped to shreds. Similar occurrences in several parts of the city that evening were sufficiently alarming to have warranted some precautionary measures or to have justified an “out-of-bounds” order. All that the police officials did, however, was to call up some additional reserves and announce that any Mexicans involved in the rioting would be promptly arrested. That there had been no counterattacks by the Mexicans up to this point apparently did not enter into the police officers’ appraisal of the situation. One thing must be said for the Los Angeles police: it is above all consistent. When it is wrong, it is consistently wrong; when it makes a mistake, it will be repeated. By the night of June sixth the police had worked out a simple formula for action. Knowing that wherever the sailors went there would be trouble, the police simply followed the sailors at a conveniently spaced interval. Six carloads of sailors cruised down Brooklyn Avenue that evening. At Ramona

198  North from Mexico

Boulevard, they stopped and beat up eight teenage Mexicans. Failing to find any Mexican zoot-suiters in a bar on Indiana Street, they were so annoyed that they proceeded to wreck the establishment. In due course, the police made a leisurely appearance at the scene of the wreckage but could find no one to arrest. Carefully following the sailors, the police arrested eleven boys who had been beaten up on Carmelita Street; six more victims were arrested a few blocks further on, seven at Ford Boulevard, six at Gifford Street—and so on straight through the Mexican east-side settlements. Behind them came the police, stopping at the same street corners “to mop up” by arresting the injured victims of the mob. By morning, some forty-four Mexican boys, all severely beaten, were under arrest.

2. Operation “Dixie” The stage was now set for the really serious rioting of June seventh and eighth. Having featured the preliminary rioting as an offensive launched by sailors, soldiers, and marines, the press now whipped public opinion into a frenzy by dire warnings that Mexican zoot-suiters planned mass retaliations. To ensure a riot, the precise street corners were named at which retaliatory action was expected and the time of the anticipated action was carefully specified. In effect these stories announced a riot and invited public participation. “Zooters Planning to Attack More Servicemen,” headlined the Daily News; “Would jab broken bottlenecks in the faces of their victims, . . . Beating sailors’ brains out with hammers also on the program.” Concerned for the safety of the Army, the Navy, and the Marine Corps, the Herald-Express warned that “Zooters . . . would mass 500 strong.” By way of explaining the action of the police throughout the subsequent rioting, it should be pointed out that, in June 1943, the police were on a bad spot. A man by the name of Beebe, arrested on a drunk charge, had been kicked to death in the Central Jail by police officers. Through the excellent work of an alert police commissioner, the case had finally been broken and, at the time of the riots, a police officer by the name of Compton Dixon was on trial in the courts. While charges of police brutality had been bandied about for years, this was the first time that a seemingly airtight case had been prepared. Shortly after the riots, a Hollywood police captain told a motion picture director that the police had touched off the riots “in order to give Dixie (Dixon) a break.” By staging a fake demonstration of the alleged necessity for harsh police methods, it was hoped that the jury would acquit Dixon. As a matter of fact, the jury did disagree and on July 2, 1943, the charges against Dixon were dismissed. On Monday evening, June seventh, thousands of Angelenos, in response to twelve hours’ advance notice in the press, turned out for a mass lynching. Marching through the streets of downtown Los Angeles, a mob of several

Blood on the Pavements   199

thousand soldiers, sailors, and civilians, proceeded to beat up every zoot-suiter they could find. Pushing its way into the important motion picture theaters, the mob ordered the management to turn on the house lights and then ranged up and down the aisles dragging Mexicans out of their seats. Street cars were halted while Mexicans, and some Filipinos and Negroes, were jerked out of their seats, pushed into the streets, and beaten with sadistic frenzy. If the victims wore zoot suits, they were stripped of their clothing and left naked or half-naked on the streets, bleeding and bruised. Proceeding down Main Street from First to Twelfth, the mob stopped on the edge of the Negro district. Learning that the Negroes planned a warm reception for them, the mobsters turned back and marched through the Mexican east side spreading panic and terror. Here is one of numerous eyewitness accounts written by Al Waxman, editor of The Eastside Journal: At Twelfth and Central I came upon a scene that will long live in my memory. Police were swinging clubs and servicemen were fighting with civilians. Wholesale arrests were being made by the officers. Four boys came out of a pool hall. They were wearing the zoot-suits that have become the symbol of a fighting flag. Police ordered them into arrest cars. One refused. He asked: “Why am I being arrested?” The police officer answered with three swift blows of the night-stick across the boy’s head and he went down. As he sprawled, he was kicked in the face. Police had difficulty loading his body into the vehicle because he was one-legged and wore a wooden limb. Maybe the officer didn’t know he was attacking a cripple. At the next corner a Mexican mother cried out, “Don’t take my boy, he did nothing. He’s only fifteen years old. Don’t take him.” She was struck across the jaw with a night-stick and almost dropped the two and a half year old baby that was clinging in her arms. . . . Rushing back to the east side to make sure that things were quiet here, I came upon a band of servicemen making a systematic tour of East First Street. They had just come out of a cocktail bar where four men were nursing bruises. Three autos loaded with Los Angeles policemen were on the scene but the soldiers were not molested. Farther down the street the men stopped a streetcar, forcing the motorman to open the door and proceeded to inspect the clothing of the male passengers. “We’re looking for zoot-suits to burn,” they shouted. Again the police did not interfere. . . .Half a block away . . . I pleaded with the men of the local police substation to put a stop to these activities. “It is a matter for the military police,” they said.

Throughout the night the Mexican communities were in the wildest possible turmoil. Scores of Mexican mothers were trying to locate their youngsters

200  North from Mexico

and several hundred Mexicans milled around each of the police substations and the Central Jail trying to get word of missing members of their families. Boys came into the police stations saying: “Charge me with vagrancy or anything, but don’t send me out there!” pointing to the streets where other boys, as young as twelve and thirteen years of age, were being beaten and stripped of their clothes. From affidavits which I helped prepare at the time, I should say that not more than half of the victims were actually wearing zoot-suits. A Negro defense worker, wearing a defense-plant identification badge on his workclothes, was taken from a street car and one of his eyes was gouged out with a knife. Huge half-page photographs, showing Mexican boys stripped of their clothes, cowering on the pavements, often bleeding profusely, surrounded by jeering mobs of men and women, appeared in all the Los Angeles newspapers. As Al Waxman most truthfully reported, blood had been “spilled on the streets of the city.” At midnight on June seventh, the military authorities decided that the local police were completely unable or unwilling to handle the situation, despite the fact that a thousand reserve officers had been called up. The entire downtown area of Los Angeles was then declared “out of bounds” for military personnel. This order immediately slowed down the pace of the rioting. The moment the Military Police and Shore Patrol went into action, the rioting quieted down. On June eighth the city officials brought their heads up out of the sand, took a look around, and began issuing statements. The district attorney, Fred N. Howser, announced that the “situation is getting entirely out of hand,” while Mayor Fletcher Bowron thought that “sooner or later it will blow over.” The chief of police, taking a count of the Mexicans in jail, cheerfully proclaimed that “the situation has now cleared up.” All agreed, however, that it was quite “a situation.” Unfortunately “the situation” had not cleared up; nor did it blow over. It began to spread to the suburbs where the rioting continued for two more days. When it finally stopped, the Eagle Rock Advertiser mournfully editorialized: “It is too bad the servicemen were called off before they were able to complete the job. . . . Most of the citizens of the city have been delighted with what has been going on.” County Supervisor Roger Jessup told the newsmen: “All that is needed to end lawlessness is more of the same action as is being exercised by the servicemen!” While the district attorney of Ventura, an outlying county, jumped on the bandwagon with a statement to the effect that “zoot suits are an open indication of subversive character.” This was also the opinion of the Los Angeles City Council which adopted a resolution making the wearing of zootsuits a misdemeanor! On June eleventh, hundreds of handbills were distributed to students and posted on bulletin boards in a high school attended by many Negroes and Mexicans which read: “Big Sale. Second-Hand Zoot Suits.

Blood on the Pavements   201

Slightly Damaged. Apply at Nearest U.S. Naval Station. While they last we have your Size.”

3. When The Devil Is Sick . . . Egging on the mob to attack Mexicans in the most indiscriminate manner, the press developed a fine technique in reporting the riots. “44 Zooters Jailed in Attacks on Sailors” was the chief headline in the Daily News of June seventh; “Zoot Suit Chiefs Girding for War on Navy” was the headline in the same paper on the following day. The moralistic tone of this reporting is illustrated by a smug headline in the Los Angeles Times of June seventh: “Zoot Suiters Learn Lesson in Fight with Servicemen.” The riots, according to the same paper, were having “a cleansing effect.” An editorial in the Herald-Express said that the riots “promise to rid the community of . . . those zoot-suited miscreants.” While Mr. Manchester Boddy, in a signed editorial in the Daily News of June ninth, excitedly announced that “the time for temporizing is past. . . . The time has come to serve notice that the City of Los Angeles will no longer be terrorized by a relatively small handful of morons parading as zoot suit hoodlums. To delay action now means to court disaster later on.” As though there had been any “temporizing,” in this sense, for the prior two years! But once the Navy had declared the downtown section of Los Angeles “out of bounds,” once the Mexican ambassador in Washington had addressed a formal inquiry to Secretary of State Hull, and once official Washington began to advise the local minions of the press of the utterly disastrous international effects of the riots, in short when the local press realized the consequences of its own lawless action, a great thunderous cry for “unity,” and “peace,” and “order” went forth. One after the other, the editors began to disclaim all responsibility for the riots which, two days before, had been hailed for their “salutary” and “cleansing” effect. Thus on June eleventh the Los Angeles Times, in a pious mood, wrote that, at the outset, zoot-suiters were limited to no specific race; they were AngloSaxon, Latin and Negro. The fact that later on their numbers seemed to be predominantly Latin was in itself no indictment of that race at all. No responsible person at any time condemned Latin-Americans as such.

Feeling a twinge of conscience, Mr. Boddy wrote that “only a ridiculously small percentage of the local Mexican population is involved in the so-called gang demonstrations. Every true Californian has an affection for his fellow citizens of Mexican ancestry that is as deep rooted as the Mexican culture that influences our way of living, our architecture, our music, our language, and

202  North from Mexico

even our food.” This belated discovery of the Spanish-Mexican cultural heritage of California was, needless to say, rather ironic in view of the fact that the ink was not yet dry on Mr. Boddy’s earlier editorial in which he had castigated the Mexican minority as “morons.” To appreciate the ironic aspects of “the situation,” the same newspapers that had been baiting Mexicans for nearly two years now began to extol them.2 As might have been expected, this post-mortem mood of penitence and contrition survived just long enough for some of the international repercussions of the riots to quiet down. Within a year, the press and the police were back in the same old groove. On July 16, 1944, the Los Angeles Times gave front-page prominence to a curious story under the heading: “Youthful Gang Secrets Exposed.” Indicating no source, identifying no spokesman, the story went on to say that “authorities of the Superior Court” had unearthed a dreadful “situation” among juvenile delinquents. Juveniles were using narcotics, marijuana, and smoking “reefers.” Compelled to accept drug addiction, “unwilling neophytes” were dragooned into committing robberies and other crimes. Young girls were tattooed with various “secret cabalistic symbols” of gang membership. The high pompadours affected by the cholitas, it was said, were used to conceal knives and other “weapons.” Two theories were advanced in the story by way of “explaining” the existence of these dangerous gangs: first, that “subversive groups” in Los Angeles had organized them; and, second, that “the gangs are the result of mollycoddling of racial groups.” In view of the record, one is moved to inquire, what mollycoddling? by the police? by the juvenile authorities? by the courts? Backing up the news story, an editorial appeared in the Times on July eighteenth entitled: “It’s Not a Nice Job But It Has to Be Done.” Lashing out at “any maudlin and misguided sympathy for the ‘poor juveniles,’” the editorial went on to say that “stern punishment is what is needed: stern and sure punishment. The police and the Sheriff ’s men should be given every encouragement to go after these young gangsters” (emphasis mine). Coincident with the appearance of the foregoing news story and editorial, the Juvenile Court of Los Angeles entered a most remarkable order in its minutes on July 31, 1944. The order outlined a plan by which Mexican wards of the Juvenile Court, over sixteen years of age, might be turned over to the Atchison, Topeka, and Santa Fe Railroad for a type of contract-employment. A form of contract, between the parents of the youngsters and the railroad, was attached to the order. The contract provided that the ward was to work “as a track laborer” at 58½¢ per hour; that $1.03 per day was to be deducted for board, $2.50 per month for dues in a hospital association, and 10¢ a day for laundry. It was also provided that one-half of the pay was to be turned over to the probation officers to be held in trust for the ward. That this order was specifically aimed at Mexican juveniles is clearly shown by the circumstance that the court,

Blood on the Pavements   203

prior to approving the arrangement, had first secured its approval by a committee of “representative” leaders of the Mexican-American community.

4. The Strange Case of the Silk Panties All of this, one will say—the Sleepy Lagoon case, the riots, etc.—belongs to the past. But does it? On the morning of July 21, 1946, a thirteen-year-old Mexican boy, Eugene Chavez Montenegro, Jr., was shot and killed by a deputy sheriff in Montebello Park on the east side of Los Angeles. The deputy sheriff later testified that he had been called to the area by reports of a prowler. On arriving at the scene, he had stationed himself near a window of the house in question and had played his flashlight on the window. A little later, he testified, “a man” lifted the screen on the window, crawled out, and ran past him. When the “man” failed to halt on order, he had shot him in the back. At the coroner’s inquest, the same deputy also testified that he had seen another officer remove a pair of “silk panties” from the dead boy’s pocket and that the boy was armed with “a Boy Scout’s knife.” While incidents of this kind have been common occurrences in Los Angeles for twenty years, in this case the officers had shot the wrong boy. For it turned out that young Montenegro was an honor student at St. Alphonsus parochial school; that his parents were a highly respectable middle-class couple; and that the neighbors, Anglo-Americans as well as Mexicans, all testified that the boy had an excellent reputation. Accepting the officers’ version of the facts, it was still difficult to explain why they had made no effort to halt the boy, who was five feet three inches tall, when he ran directly past them within arms’ reach. Before the hearings were over, the “silk panties” story was exposed as a complete fake. Despite a gallant fight waged by Mr. and Mrs. Montenegro to vindicate the reputation of their son, nothing came of the investigation. “Raging Mother Attacks Deputy Who Slew Son” was the Daily News headline on the story of the investigation. . . .On January 23, 1947 the attorney general of California ordered the removal of two police officers for the brutal beating of four Mexican nationals who, with eight hundred of their countrymen, had been brought to Oxnard to harvest the crops. . . . On March 30, 1946, a private detective killed Tiofilo Pelagio, a Mexican national, in a café argument. . . . On the same day affidavits were presented to the authorities that confessions from four Mexican boys, all minors, had been obtained by force and violence. . . . Esther Armenta, sixteen years of age, complained to her mother that she was being mistreated by Anglo-American classmates in a Los Angeles junior high school. “They would spit on her,” said Mrs. Catalina Armenta, the mother, “and call her a ‘dirty Mex.’ Esther would come home in tears and beg me to get her transferred.” A few weeks later the girl was in juvenile court charged with the use of “bad

204  North from Mexico

language.” She was then sent to the Ventura School for Girls, a so-called “correctional” institution. When Mrs. Armenta finally got permission to visit her daughter, in the presence of a matron, the girl had “black and blue marks on her arm” and complained that she had been whipped by one of the matrons. . . .  On April 10, 1946, Mrs. Michael Gonzales complained to the Federation of Spanish-American Voters that her daughter had been placed in the Ventura School without her knowledge or consent and that when she had protested this action she had been threatened with deportation by an official of the juvenile court. . . . On the basis of a stack of affidavits, the San Fernando Valley Council on Race Relations charged on May 16, 1947 that the police had broken into Mexican homes without search warrants; that they had beaten, threatened, and intimidated Mexican juveniles; and that they were in the habit of making “wholesale roundups and arrests of Mexican-American boys without previous inquiry as to the arrested boys’ connection—if any—with the crime in question.” . . . In 1946 a prominent official of the Los Angeles schools told me that she had been horrified to discover that, in the Belvedere district, Mexican-American girls, stripped of their clothing, were forced to parade back and forth, in the presence of other girls in the “gym,” as a disciplinary measure . . . *

5. The Politics of Prejudice I reported the zoot-suit riots in Los Angeles for PM and The New Republic and had a hand in some of the hectic events of that memorable week. Following the June seventh rioting, I chaired a meeting of a hundred or more citizens at which an emergency committee was formed to bring about, if possible, a return to sanity in Los Angeles. That same evening we communicated with Attorney General Robert W. Kenny in San Francisco by telephone and urged him to induce Governor Earl Warren to appoint an official committee of inquiry. The next day the governor appointed a committee of five which included four names from a panel which I had submitted. The fifth member was the governor’s own selection: Mr. Leo Carrillo. Mr. Carrillo, like the sheriff of Los Angeles, is a descendant of “an early California family.” The committee immediately assembled in Los Angeles where Mr. Kenny presented to them a proposed report, with findings and recommendations, which I had prepared at his request. With some modifications, this report was adopted by the committee and submitted to the governor. Out of the work of our emergency committee there finally emerged, after a year of negotiation, the present-day Council of Civic Unity. * For a detailed account of still another “incident,” see Justice for Salcido by Guy Endore, published by the Civil Rights Congress of Los Angeles, July 1948.

Blood on the Pavements   205

Praising the report of the governor’s committee—which I had prepared— the Los Angeles Times devoted several harsh editorials to certain “reckless” individuals, myself included, who had suggested that “racial prejudice” might have had something to do with the riots! “When trouble arose,” said the Times in an editorial of June 15, 1943, “through the depredations of the young gangs attired in zoot-suits, it was their weird dress and not their race which resulted in difficulties. That is a simple truth which no amount of propaganda will change.” In the same editorial, the charges of unfairness which I had raised in connection with the Sleepy Lagoon case were branded as “distortions,” “wild charges,” and “inflammatory accusations” (charges later confirmed in minute detail by the District Court of Appeals). When Mrs. Eleanor Roosevelt innocently remarked in her column that the zoot-suit riots were “in the nature of race riots,” she was severely taken to task by the Times in an editorial of June eighteenth under the caption: “Mrs. Roosevelt Blindly Stirs Race Discord.” Even the president of the Los Angeles Chamber of Commerce felt compelled to reply to Mrs. Roosevelt. “These socalled ‘zoot-suit’ riots,” he said, “have never been and are not now in the nature of race riots. . . . At no time has the issue of race entered into consideration. . . . Instead of discriminating against Mexicans, California has always treated them with the utmost consideration.”3 The zoot-suit riots in Los Angeles were the spark that touched off a chainreaction of riots across the country in midsummer 1943. Similar “zootsuit” disturbances were reported in San Diego on June ninth; in Philadelphia on June tenth; in Chicago on June fifteenth; and in Evansville, Indiana, on June twenty-seventh. Between June sixteenth and August first, large-scale race riots occurred in Beaumont, Texas, in Detroit, and in Harlem. The Detroit riots of June 20–21 were the most disastrous riots in a quarter of a century. The swift, crazy violence of the Harlem riot resulted, in a few hours’ time, in property damage totaling nearly a million dollars. The rapid succession of these violent and destructive riots seriously interfered with the war effort and had the most adverse international repercussions. The spark that ignited these explosions occurred in El Pueblo de Nuestra Señora La Reina de Los Angeles de Porciúncula, founded by Felipe de Neve in 1781, settled by Mexican pobladores. None of these disturbances had more serious international consequences than the zoot-suit riots. On April 20, 1943, President Roosevelt had held his historic meeting with President Camacho on the soil of Mexico. At the time the riots occurred, Mexico was our ally in the war against Germany, Italy, and Japan. Large-scale shipments of Mexican nationals had just begun to arrive in the United States to relieve the critical manpower shortage. “Our two countries,” President Roosevelt had said, “owe their independence to the fact that your ancestors and mine held the same truths to be worth fighting for and

206  North from Mexico

dying for. Hidalgo and Juárez were men of the same stamp as Washington and Jefferson.” President Camacho, replying to this toast, had said that “the negative memories” of the past were forgotten in the accord of today. And then in the largest city in the old Spanish borderland had come this explosion of hatred and prejudice against Spanish-speaking people. In response to a request from the Mexican ambassador, Secretary of State Hull had asked Mayor Fletcher Bowron for an official explanation. With a perfectly straight face, the mayor replied that the riots were devoid of any element of prejudice against persons of Mexican descent! The same edition of the newspapers that carried this statement also carried another statement by the mayor under a headline which read: “Mayor Pledges 2-Fisted Action, No Wrist Slap”—a reference to police action contemplated against the Mexican minority. On June ninth Mr. Churchill Murray, local representative of the coordinator of Inter-American Affairs, wired Mr. Rockefeller that the riots were “non-racial.” “The frequency of Mexican names among the victims,” he said, “was without actual significance.” If all this were true, asked Dan G. Acosta in a letter to the Los Angeles press, “Why are we consistently called hoodlums? Why is mob action encouraged by the newspapers? Why did the city police stand around saying very nonchalantly that they could not intervene and even hurrahed the soldiers for their ‘brave’ action? Not until these questions are answered, will the Mexican population feel at ease.” What the riots did, of course, was to expose the rotten foundations upon which the City of Los Angeles had built a papier-mâché façade of “InterAmerican Good Will” made up of fine-sounding Cinco de Mayo proclamations. During the riots, the press, the police, the officialdom, and the dominant control groups of Los Angeles were caught with the bombs of prejudice in their hands. One year before the riots occurred, they had been warned of the danger of an explosion. The riots were not an unexpected rupture in AngloHispano relations but the logical end-product of a hundred years of neglect and discrimination. The riots left a residue of resentment and hatred in the minds and hearts of thousands of young Mexican-Americans in Los Angeles. During the rioting, one Los Angeles newspaper had published a story to the effect that the cholitas and pachucas were merely cheap prostitutes, infected with venereal disease and addicted to the use of marijuana. Eighteen Mexican-American girls promptly replied in a letter which the metropolitan press refused to publish: “The girls in this meeting room consist of young girls who graduated from high school as honor students, of girls who are now working in defense plants because we want to help win the war, and of girls who have brothers, cousins, relatives and sweethearts in all branches of the American armed forces. We have not been able to have our side of the story told.” The letter, with a picture of the girls, was published in Al Waxman’s Eastside Journal on June 16, 1943.

Blood on the Pavements   207

Still another group of Mexican-American girls—real pachucas these—bitterly protested the story in another letter which the metropolitan press did not publish. These girls insisted that they should be examined, as a group, by an officially appointed board of physicians so that they could prove that they were virgins. Long after the riots, I have seen Mexican-American boys pull creased and wrinkled newspaper clippings from their wallets and exhibit this slanderous story with the greatest indignation. Four years have now passed since the riots, but the blood has not yet been washed from the pavements of Los Angeles.

14

The War Years

World War II has had, of course, a profound effect on Anglo-Hispano relations in the Southwest. The Spanish-speaking people had demonstrated their loyalty and patriotism during the Spanish-American War and World War I, but the enthusiasm with which they participated in the second World War had a special motivation. Prior to Pearl Harbor, the 200th and 515th Coast Artillery units of the New Mexico National Guard had been sent to the Philippines, largely because the officers and troops in these units spoke Spanish. That so large a percentage of the American troops captured or killed at Bataan were Mexican-Americans merely served to stress the intensity with which the Spanish-speaking identified themselves with the Allied cause. While the precise number of Mexican-Americans who served in the armed forces is not known, the available estimates range from 375,000 to 500,000. That the figure was actually somewhat disproportionate to the size of the Spanish-speaking minority can be assumed from the large number of MexicanAmericans of draft age and the fact that so few Mexican-Americans served on selective service boards. Throughout the war long lists of Mexican-American casualties appeared in the newspapers of the Southwest, usually accompanied by stories of Mexican-Americans who had won special citations for gallantry. Long before the war was over, the cumulative effect of the casualty lists and the stories of Mexican-American gallantry had left a noticeable impress on the Anglo-American conscience.

1. Joe Martinez and Company One of the most impressive stories of Mexican-American gallantry in the war was that of Joe Martinez. Born in Taos, Martinez was working in the sugarbeet fields of Colorado when he enlisted in the army. For exceptional gallantry in the Aleutians, where he was killed, Martinez was posthumously awarded the Congressional Medal of Honor. A chapter of the American Veterans Committee in Colorado has been named in his honor. Five Texas-Mexicans also received

210  North from Mexico

the Congressional Medal of Honor, as did Sylvester Herreras, of Phoenix, who lost both legs in battle, and Pfe. Manuel Perez, of Oklahoma City, who was killed in the Battle of Luzon. Indeed the list of Mexican-American war heroes is a long one and includes names from every state in the Southwest. “As I read the casualty lists from my own state,” said Congressman Jerry Voorhis, “I find anywhere from one-fourth to one-third of those names are names such as Gonzales or Sanchez, names indicating that the very lifeblood of our citizens of Latin-American descent in the uniform of the armed forces of the United States is being poured out to win victory in the war. We ought not to forget that. We ought to resolve that in the future every single one of these citizens shall have the fullest and freest opportunity which this country is capable of giving him, to advance to such positions of influence and eminence as their own personal capacities make possible.”1 Born on the rancho Los Potreros in the Lower Rio Grande Valley, Ricardo Noyola, like his father, could not speak English. Having worked as a farmhand since the age of thirteen, he had had little opportunity for schooling. There were fifty-five or sixty such boys at Camp Robinson, some from Texas, some from New Mexico, some from Colorado. Not knowing what to do with them, the post commander finally put them in a special platoon under the command of an officer who spoke Spanish. Within thirteen weeks, the members of the unit had not only mastered the technique of soldiering but had acquired a conversational knowledge of English sufficient to enable them to serve in mixed units. Several of these soldiers, including Noyola, won special citations for gallantry. For such men wartime service was a real opportunity for acculturation, perhaps the first such opportunity they had ever had. In every phase of the war, including the defense plants and the training schools as well as the armed services, similar opportunities opened up for thousands of Mexican-Americans: to learn new skills, to acquire new experiences, to come in contact with entirely new currents of thought and opinion. In more than one community, joint service in various civilian defense agencies had a marked tendency to break down the barriers which had so long separated the Spanish-speaking from the rest of the population. Out of this wartime experience, as might have been expected, came a new pride in citizenship and a growing resentment of all forms of discrimination. Sergeant Macario García was one of the five Texas-Mexicans who received the Congressional Medal of Honor. One day while home on furlough he dropped into the Oasis Café in Sugarland, Texas, for a cup of coffee. Informed that the Oasis Café did not serve Mexicans, he demanded service in no uncertain terms. Two sailors came to his aid in the fight which ensued when the proprietor attempted to eject him. A deputy sheriff, summoned to the café, broke up the fight and told the participants to “forget it.” The story naturally aroused a great furor in Mexico and Walter Winchell brought the facts to the

The War Years   211

attention of his radio listeners. Once Sugarland had received this unwelcome notoriety, the county authorities felt compelled to vindicate the honor of the community: García was then arrested on a charge of “aggravated assault!” Sergeant José Mendoza López, of Brownsville, another winner of the Congressional Medal of Honor, was denied service in a restaurant in a small town in the Rio Grande Valley, under similar circumstances, following his return from a goodwill tour of Mexico which had been arranged by the army.2 Needless to say, Sergeant López protested the incident with a vehemence which must have come as quite a surprise to the Anglo-American residents. On February 25, 1946, Pfe. Daniel S. Elizalde, while on leave in Los Angeles, was killed by a special night watchman under circumstances that might easily have resulted in a murder prosecution had the victim not been a MexicanAmerican. The failure of the authorities to prosecute, despite the most insistent pressure from the Mexican community, resulted in the formation of the Elizalde Anti-Discrimination Committee made up of Mexican-American veterans. “Mexican-American soldiers,” said Marine Corps veteran Balton Llanes, “shed at least a quarter of the blood spilled at Bataan. . . . What they want now is a decent job, a decent home, and a chance to live peacefully in the community. They don’t want to be shot at in the dark.” On the evening of April 23, 1947, Charles White, a Mexican-American war veteran who had been awarded the Silver Star, the Purple Heart, the Infantry Badge, and a Presidential Citation, demanded service at the Silver Slipper Nite Club near La Junta, Colorado, after two other Mexican-Americans had been denied service. In a fight with the proprietor, Sergeant White was killed. Petitions signed by three thousand Mexican-Americans demanded the indictment of the night-club owner but the authorities ignored the petitions and even refused to revoke the liquor license for the club. Out of this incident came a committee which has been conducting a militant campaign in Colorado against all forms of discrimination. In fact, wherever incidents of this sort have occurred, similar committees have been formed, usually spearheaded by Mexican-American veterans.

2. A Tear for José Davilla During the war an increasing number of Anglo-Americans began to protest acts of discrimination against Mexican-Americans. While working in the cherry orchards near Hart, Michigan, José Davilla, nineteen years of age, became acquainted with an Anglo-American girl. They played together; worked together; and, on several occasions, went out together. One day, as a silly prank, young Davilla took the girl’s glasses and refused to return them. Hearing of this affront to Anglo-American womanhood, the sheriff of Oceana County

212  North from Mexico

sought out Davilla in the main street of Hart and attempted to arrest him without a warrant. For half an hour the man and the boy fought in the street. When the sheriff was unable to subdue the boy, even with the aid of an eighteen-inch blackjack, he shot and killed him. Swift Lathers, editor of the local Mears News,—“the smallest newspaper in the world”—promptly wrote an editorial accusing the sheriff of murder. When the sheriff had Lathers arrested for criminal libel, he replied: Do I stand alone facing the sullen crowd? I have stood there before. I am that way. I would rather stand up against the whole world to defend the underdog than to sit on the plush chairs of the aggressors. I know that somewhere there is a tear for José Davilla. Short days ago he worked and sang among us, felt dawn and saw sunsets glow. A few more days and his lingering countrymen will go back to the Border. But that pool of blood on the sidewalk of Hart will not wash away.

With such newspapers as the Detroit Free Press rushing to his defense, Lathers was finally acquitted. “The streets of Hart are wet today,” wrote James S. Pooler in the Free Press (November 1, 1944), “and golden leaves are plastered over the spot where José Davilla died. José hasn’t been dead a month. In fact it was only a week ago that his family took their son’s body home to Texas. José has no tomb or tombstone here but he has an epitaph, written by Swift Lathers.” Nor was Swift Lathers the only editor who wrote an epitaph for a MexicanAmerican during the war. Led by the St. Louis Star-Times, the newspapers of St. Louis devoted five hundred thousand words of copy to their contention, finally vindicated, that Edward Melendes, a Mexican-American, had been kicked to death by policemen in a St. Louis jail on July 2, 1942. “If Melendes can die in a St. Louis police cell,” wrote the Star-Times, “as the result of an inhuman beating, and the perpetrators go unpunished, the painfully established liberties of all men have been whittled away. Human beings in a democracy cannot be divided into two classes, those who may safely be beaten and left to die in police stations and those who may not. That is why the Star-Times will continue to fight to learn and print the truth, and continue to ask, ‘Who Killed Edward Melendes?’”

3. Across the Border During the war, also, Mexican-Americans began to receive some extremely effective assistance from their brothers across the border. Anxious to see the Good Neighbor Policy realized, many groups, individuals, and organizations in Mexico began to give a new emphasis to the treatment of Spanish-speaking people in the United States. One of the most effective of these groups was the

The War Years   213

Comité Mexicano Contra el Racismo, formed in August 1944. In its publication Fraternidad appeared a regular column entitled “Texas, Buen Vecino?” (“Texas, Good Neighbor?”). In issue after issue, the column kept listing acts of discrimination against Mexicans in Texas. Giving names of establishments, and the dates of “incidents,” Fraternidad documented a long list of discriminatory practices which had occurred in approximately 150 Texas communities. This constant needling soon began to have its effects on both sides of the border. Mexican officials were compelled to bring pressure on both the State Department and the Texas authorities to correct these conditions, while, on this side of the border, Texans began to make “goodwill” tours in Mexico and to play up the Good Neighbor Policy. At the same time, the issue of discrimination was raised at a series of official or semiofficial meetings in Mexico City: at the meeting of the Inter-American Bar Association in August 1944; at the conference of the International Labor Organization in April 1945; and at the Inter-American Conference on Problems of War and Peace at which the Act of Chapultepec was adopted on March 6, 1946. In a resolution adopted at this conference all of the nations of the Western Hemisphere entered into a pledge “to make every effort to prevent in their respective countries all acts which may provoke discrimination among individuals because of race or religion.” These international pronouncements, coupled with the insistent pressure of Mexican officials, brought about a new recognition of the importance of equal treatment of Mexicans, aliens and citizens, in this country. At the same time, social tensions in the borderlands were consistently aggravated, during the war years, by agents of the Sinarquista movement, founded in Leon, Mexico, in 1937, by Oskar Hellmuth Schreiter, a German Nazi, José Antonio Urquiza, and Trueba Olivares, both with close Falangist connections. Adopting a Fascist ideology, Point 13 in the movement’s “16 Principles” stated that Mexico’s true sons must be “worthy of their fatherland and reclaim as well as honor and respect its lands,” a clear reference to the recapture of the borderlands taken from Mexico in 1848.3 The first regional Sinarquista Committee was organized in Los Angeles on November 1, 1937, shortly after the movement was founded in Mexico. By 1942 the movement boasted of two thousand members in the United States and “cells” were known to exist in such Southern California communities as Pacoima, San Fernando, San Bernardino, La Verne, Ontario, Watts, El Monte, Oxnard, Pomona, and Azusa; and, in Texas, at El Paso, McAllen, Mission, and Laredo. The first meeting of Sinarquista chiefs in the United States took place in El Paso on September 27, 1942. Stories in El Sinarquista tell, in elaborate detail, of how local committees were formed in the Belvedere section of Los Angeles. In April 1943—and the date is important both in relation to the meeting of President Roosevelt and President Camacho in Mexico and to the Zoot

214  North from Mexico

Suit Riots in Los Angeles—four Mexican “students”—all men over thirty years of age—toured the United States lecturing on the principles of Sinarquismo to Mexican-American audiences. Two of these men were high-ranking officials of the Sinarquista movement in Mexico. In the latter part of 1942 and throughout 1943 local Sinarquista committees were seeking to capitalize upon discrimination against Mexican-Americans in the borderlands to interfere, if possible, with the war effort. The Mexican government regarded the agitation with sufficient seriousness to send Ernesto Felix Díaz to Los Angeles, as its official representative, to alert the Mexican residents to the dangers of Sinarquista propaganda.4 Testifying before a legislative committee, Pedro Villasenor, president of the Southern California branch of the movement, said that the local branches had five hundred members; but the Department of Justice reported that the membership was nearer two thousand. With the local press charging that the Zoot Suit Riots had been fomented by subversive left-wing elements, it is ironic to note that the Sinarquista movement was completely whitewashed in this investigation which ended up by being yet another investigation of “Communism.”5 While it would be highly inaccurate to say that the Sinarquista movement had a direct responsibility for the Zoot Suit Riots, it is true that its propaganda had a most disquieting effect on Mexican opinion and, to this extent, was a factor in the riots. By harping on the theme that the United States was fighting another war of “Yankee imperialism” and aggression, Sinarquista had inflamed a small section of Mexican opinion and had treated the issue of discrimination in the most demagogic manner.6

4. Los Braceros With World War II the same clamorous demand for Mexican labor of 1918 was repeated. Concerned over the Good Neighbor Policy, however, the government did not capitulate quite so easily to these demands as it had in World War I. For one thing, the Mexican government looked with some considerable disfavor upon the proposal. But following Mexico’s declaration of war on Germany, Italy, and Japan, on May 22, 1942, negotiations were renewed, and, in August, an agreement was entered into between the two nations setting forth the conditions on which Mexican labor might be recruited for wartime employment. This agreement stipulated that imported workers were to be assured of free transportation to and from their homes; that they were to be provided subsistence en route; that they were not to be used to displace other workers or to reduce wage rates; and that certain minimum guarantees, governing wages, and working conditions, would have to be observed. Both the idea of such an agreement and its form can be traced back to proposals which Dr. Manuel

The War Years   215

Gamio and Mr. Ernesto Galarza, of the Pan-American Union, had previously advanced. On September 29, 1942, the first shipment of 1,500 Mexican braceros arrived in Stockton, California, with the slogan “De Las Democracias Será La Victoria” scribbled in chalk on the Pullman cars. Previously the Farm Security Administration had worked out the various forms and had defined the various relationships to be used in carrying the agreement into effect. Individual agreements were entered into with each worker recruited, which embodied the guarantees of the master agreement. The FSA then entered into agreements with various farm organizations which contained similar guarantees. To police the agreements, Mexican consuls and a limited number of Mexican labor officials were authorized to make inspections and to investigate complaints and grievances. The FSA officials were also careful to provide a friendly reception for the workers. Reception committees were on hand at the station, flags waved, bands played, and many speeches were made. Excellent recreational and educational programs had been worked out with the thought in mind of using the braceros as future ambassadors of goodwill in Mexico. As long as the FSA was in charge of the program, the agreement was carried out to the letter; but, on July 1, 1943, the War Food Administration was substituted for the FSA as the enforcing agency—a change which was tantamount to turning the whole program over to the farm associations. Once they were in control of the program, the new arrangement could not have been improved upon from the growers’ point of view. With the government paying all transportation and administration expenses, they were spared even the trouble of recruiting labor. Assured an unlimited market and a high level of prices, the large-scale employers of farm labor made fabulous wartime profits. From 1943 through 1947, the federal government appro­­ priated $120,000,000 for the labor importation program—every penny of which should be regarded as a direct subsidy to the large-scale employers of farm labor in a period of unprecedented prosperity. While the War Food Administration did insist upon livable camps, the wage guarantees were farcical: braceros were assured $33 for each two weeks of employment. In an effort to ensure payment of prevailing wage rates, hearings were held for each major crop in advance of the season. But these “hearings” were ludicrous. For example, on January 27, 1946, officials of the War Food Administration refused to hear the testimony of Isabel Gónzales, David Braco, and Vincent G. Vigil who had come to Salt Lake from Denver to present evidence that MexicanAmerican sugar-beet workers were earning, on an average, not more than $550 per year. To appreciate what a bonanza this program was to the large farm-factories, it should be pointed out that the braceros were limited to agricultural employment. If any worker accepted a job in industry, he was subject to immediate

216  North from Mexico

deportation. The effect of this provision was to remove the farm-labor market from competition with industrial wage rates. Theoretically the employment of the braceros was not supposed to reduce wage rates; actually, with no fault on their part, it had this effect. Nevertheless the agreement represented a notable advance over the 1918 experience and it also demonstrated that a migratory labor movement can be planned and rationalized. Workers were paid transportation and subsistence en route; they were provided with better camps, medical care, accident insurance, and minimum earnings were guaranteed. Ten percent of these earnings were deducted by the government and transmitted to Mexico City for the account of the workers: in the nature of a compulsory savings fund. From 1943 to 1947, the braceros transmitted a huge sum to their families and dependents in Mexico. Thousands of Mexican workers were eager to enlist and even paid sizable sums, in the way of bribes, to be enrolled. Three members of the Chamber of Deputies in Mexico City, as well as a number of minor officials, were indicted for having solicited these bribes. The number of workers recruited for agricultural employment were as follows: 1942—4,203; 1943—52,098; 1944—62,170; 1945—120,000; 1946— 82,000; 1947—55,000. These are the totals for “foreign labor” and include a few thousand Puerto Ricans and other workers from the West Indies; but most of those imported were from Mexico. Each year, of course, workers returned to Mexico at the end of the season, with the number employed throughout the year not being in excess of twenty-two thousand. Hence it is impossible to give the actual total of workers recruited for agriculture during the war but it would be in excess of one hundred thousand. These workers helped produce and harvest practically every major crop: sugar beets, grapes, tomatoes, apricots, peaches, prunes, cotton, and many others. In the Rocky Mountain states, they constituted half the labor supply used in harvesting 354,000 acres of sugar beets. Throughout the Middle West, they helped cultivate and harvest sugar beets, vegetables, orchard crops, and hay and grain. Working in twenty-one states, they harvested crops the value of which was estimated in 1944 at $432,010,000. In addition to those employed in agriculture, eighty thousand Mexicans were recruited and brought to this country for employment as section hands and maintenance workers on the railroads at a minimum hourly rate of 57¢ with no guarantee of minimum earnings. Employed by thirty-two rail lines, these workers performed an indispensable service in keeping the Western lines in repair during a period of exceptionally heavy freight and passenger traffic. It has been estimated that the railroad workers received $63,000,000 in wages in 1944, a large portion of which was remitted to Mexico in the form of money orders.

The War Years   217

The farm-labor importation program came to an end on December 31, 1947, but the large growers have kept up an incessant clamor for its renewal. Should the agreement be renewed, it is apparent that the planned migration of the war years can serve as an extremely important precedent. Although the wartime agreement was frequently violated it did provide a measure of protection against the hazards and rigors of migratory employment. Over the years the use of Mexican labor to relieve acute manpower shortages in the United States has proved to be of benefit to both nations. The issue has always turned on the choice between planned migration and unplanned immigration. For it is extremely debatable whether, under any circumstances, Mexican workers can be kept from crossing the border. Given the attraction of industrial employment in the United States and the ease with which the border can be crossed, Mexicans will continue to follow the old, familiar paths which lead north from Mexico. At the present time hundreds of them are paying as much as $150 to be smuggled into the United States in trucks and airplanes and recently one Mexican was so anxious to return to California that he rode a log upstream on the Colorado and then walked a hundred miles through the desert to a ranch in Imperial Valley. In 1946, alone, sixty-six thousand “wetbacks” were apprehended by the Immigration Service along the far western section of the border.7

5. The Counterpoint of Migration One of the conspicuous advantages of the farm-labor importation agreement of 1942 was that it gave the Mexican government a firm basis on which to protest acts of discrimination against Mexicans in the borderlands and also provided a means by which these protests could be backed up. After the Farm Security Administration had been relieved of responsibility for enforcing the agreement, complaints began to multiply and repeated charges were made that the growers were chiseling on guaranteed wages and working conditions. On July 30, 1945, Ernesto Galarza prepared a fourteen-page memorandum setting forth in detail the various ways by which American employers in the Southwest were undermining the agreement. More important than any specific violations of the agreement, however, were the incidents of discrimination involving both Mexican nationals and Mexican-Americans, particularly in Texas. In Snyder, Texas, an Anglo-American dentist refused to treat an American soldier of Mexican descent and in Melvin permission was refused a MexicanAmerican PTA group to use a community center built by the National Youth Administration. On October 15, 1943, the Mexican government formally protested against the segregation of children of Mexican descent in certain Texas schools. This protest was filed after a large number of Mexican-American

218  North from Mexico

families in Bolling and Goose Creek, Texas, had refused to send their children to the jim crow school. Visiting New Gulf, Texas, to participate in the celebration of the Sixteenth of September, Adolfo G. Gominguez, Mexican consul at Houston, was denied service in the Blue Moon Café. Alejandro Carillo, a member of the Chamber of Deputies, immediately brought the facts of this incident to the attention of President Ávila Camacho. Three days before Governor Coke Stevenson proclaimed the Sixteenth of September an official holiday in Texas, Sergeant Macario García had been arrested in Sugarland. In October 1943, a Mexican boy, seventeen years of age, entered a café in Levelland, Texas, to buy a package of cigarettes. The usual altercation developed in the course of which the owner of the café hit the boy over the head with a coke bottle. Within a few moments, a thousand or more Mexicans gathered outside the café and the sheriff arrived just in time to prevent a riot. In December 1944, some three hundred AFL members went on strike at Huron, South Dakota, against the employment of Mexican braceros. A storm of protest immediately appeared in the Mexican press and the State Department received a call from the Mexican ambassador. Incidents of this sort had been occurring for a hundred years but never, prior to the war, had such significance attached to them. Conditions reached such a point in Texas that Roberto Medellin of the Mexican Ministry of Labor announced in 1943 that no more braceros would be sent to work in the state “because of the number of cases of extreme, intolerable racial discrimination.” In an effort to induce the Mexican government to lift this ban, Governor Stevenson made a “goodwill” tour of Mexico and on his return appointed a Good Neighbor Commission. On June 25, 1943, the governor issued a formal proclamation calling upon all tried-and-true Texans to adhere to the Good Neighbor Policy. A few days later, the Commissioners’ Court in Cameron County adopted a resolution “condemning any discriminations against our fellow citizens of Latin-American extraction” and expressing regret that “embarrassing occurrences have injured the feelings of our said fellow citizens and neighbors.” These feeble gestures of “goodwill,” however, were largely offset by the tone of the press which proceeded to add insult to injury by explaining that the only reason Mexicans were discriminated against in Texas was because they were “dirty”!—. It is interesting to note, however, that Resolution No. 105, in which Governor Stevenson proclaimed the Good Neighbor Policy in Texas, merely called upon the citizens of the state to adopt a nondiscriminatory policy as to “all persons of the Caucasian race,” thereby attempting to deny long-resident Negro citizens a status sought to be conferred on Mexican nationals. It should also be noted that the three “Mexicans” named to the commission were upper-class, oldresident Tejanos. For these and other reasons, the Good Neighbor Commission has not been particularly effective although the mere appointment of such a

The War Years   219

commission, of course, is a significant gesture. Long after the proclamation creating the commission was issued, many Texas jails continued to exhibit signs which specified a special visiting day for “Negroes and Mexicans.” Noting the reluctance of the Texans to abandon their folkways, the Mexican weekly Mañana said that “The Nazis of Texas are not political partners of the Fuhrer of Germany but indeed they are slaves to the same prejudices and superstitions.” During the visit of American dignitaries to Mexico City for the Independence Day celebration of 1943, Francisco de P. Jiménez, a member of the Chamber of Deputies, made a speech in which he bitterly denounced the mistreatment of Mexicans in the borderlands. A former member of the Mexican consular corps in Texas, he called attention to the case of one Antonio Rangel who had been murdered by an Anglo-American. The defendant was released, shortly after his arrest, on $25 bail! He also mentioned the case of one José Aguilar who, in the course of an assault, had been seriously injured by an Anglo-American. In this case, the aggressor was exonerated upon payment of a $50 fine. In the same speech, Deputy Jiménez quoted a pastoral letter in which Archbishop Lucey of San Antonio had denounced, by name, a certain Catholic church which had openly exhibited a sign reading: “no se admite a Mexicanos.” Concluding a long recital of similar discriminatory acts, he demanded the appointment of a commission to meet with Governor Stevenson and to investigate conditions in Texas. A resolution to this effect was unanimously adopted by the Chamber of Deputies. After appointing the Good Neighbor Commission, Governor Stevenson again appealed to Foreign Minister Padilla to lift the ban against sending braceros to Texas. The foreign minister replied that the appointment of the commission had not brought about a satisfactory correction of conditions in Texas. “In many parts of Texas,” he said, “Mexicans cannot attend public gatherings without being subject to vexations, complaints and protests. There are towns where my fellow countrymen are forced to live in separate districts. Just a week ago the daughter of a Mexican consul was refused service in a public establishment.” The ban would not be lifted, he went on to say, until Texas had passed a law prohibiting such practices. A bill to this effect was introduced in the 1945 session of the Texas legislature but, needless to say, it did not pass. As a consequence, Texas still remains on Mexico’s blacklist. The boycott, however, has not seriously cramped the Texas growers for they still have a large pool of Mexican labor, including some forty thousand illegal entrants or “wetbacks.”

6. Good Neighbors and Band Music During the war years the circumstance that Mexico was our ally against an enemy that preached the doctrine of a master race, the magnificent record

220  North from Mexico

made by Mexican-Americans in the service, the emphasis on the Good Neighbor Policy, and our dependence on Mexican labor, all served to bring out and to emphasize the mistreatment of persons of Mexican descent in the United States. For the first time since 1848, the full pattern of this treatment was brought to light and glaringly exposed. In a carefully documented article in Common Ground,8 Dr. George I. Sanchez, of the University of Texas, former president of the League of United LatinAmerican Citizens, called attention to the pattern of discrimination in employment, in the schools, and in civic life throughout the borderlands. In the course of a hike, a Scoutmaster and his troop of Boy Scouts, all in uniform, were ordered out of a public park, where they had stopped to rest, because they were “Mexicans.” Texas churches posted signs reading “For Colored and Mexicans” and refused Mexicans permission to attend the “white churches” on Sundays. “In many cemeteries, whether owned by county authorities, by private individuals or corporations, or by religious organizations . . . the bodies of ‘Mexicans’ are denied the right of burial. . . . In those cemeteries where such bodies are received they are assigned a separate plot of land, far enough from the plot destined for the so-called ‘whites’ so as to be sure that the bodies of the ‘whites’ will not be contaminated by the presence of the bodies of the Mexicans.” Toilets in many Texas courthouses have signs which read: “For Whites—Mexicans Keep Out.” Mexicans are segregated with Negroes in the balconies of many motion picture theaters in the Southwest. Certain subdivisions in Southern California are restricted against Mexican occupancy, although in at least one case such restrictions have been ruled illegal. Although the pattern of discrimination against Mexicans is “spotty” and less rigid than against Negroes, it is nevertheless true that Mexicans are generally assigned a second-class status throughout the borderlands today. A careful study of the status of minority groups in Los Angeles has shown that, by reference to a number of conventional indices of status, Mexicans occupy a lower position in the community than that occupied by Negroes. In an article in The Texas Spectator for October 11, 1946, Hart Stilwell, the author of a fine novel about Mexicans in a Texas bordertown, placed his finger on the real crux of “The Mexican Problem” in the Southwest: An Anglo-American was tried in a criminal district court in a small Texas town recently on a charge of murder. The man who was killed was a Texas-Mexican, a Latin-American if you prefer that term. The Anglo-American was acquitted. The trial attracted small attention in Texas. It was not even reported fully in the newspapers printed in the town where it took place. If it made the wire services, I failed to see it in any of the larger Texas papers.

The War Years   221

I happened to be familiar with some of the details of this case. And from what I know of it, I make the following observations: if the man who was killed had been an Anglo-American and the man who did the killing had been a Latin-American, I believe the verdict would have been different. And if both men had been Latin Americans, I believe the verdict would have been different. What happened in this case is typical of what has been happening in Texas courts for a hundred years or more—twenty-five years to my personal knowledge. I have been a newspaper man in Texas for twenty-five years and I have carefully watched criminal cases in which members of the two races were involved. And if an Anglo-American has served one day in the penitentiary for the killing of a Latin-American during that period of time, I have not heard of it. . . . We can bring ten thousand Tipica Orchestras to Texas and send five thousand Rotary Clubs and Kiwanis Clubs and other goodwill delegations into Mexico, yet so long as the Mexican knows that he may be killed with impunity by any American who chooses to kill him, then all our talk about being good neighbors is merely paying lip service to a friendship we both know is a joke.

In a broadcast on May 10, 1947, Antonio Espiñosa de los Monteros, Mexico’s ambassador to the United States, called upon the people of the United States to make “a sincere, determined effort to do away with racial prejudices” against persons of Mexican descent. “The citizens of those sections,” he said (referring to the borderlands), “should realize that the day has come when it is absolutely necessary to give the Mexican absolute equality of opportunity.” There can be no doubt but that this issue is today the most sensitive test of good neighborly relations between Mexico and the United States. “Discriminations of this character,” said Sumner Welles, “inevitably cut deep. They create lasting resentments, which no eloquent speeches by government officials, nor governmental policies, however wise, can ever hope to remove. . . . So long as they continue anywhere in the United States they are bound to undermine the foundations which the two governments have laid for those cooperative ties which are so greatly to the interests of both countries, and they will, in the wider sense, impair that inter-American relationship which is today more necessary than ever before. Unless these discriminations are obliterated, and obliterated soon, the term ‘good neighbor policy’ will lose much of its real meaning.”9

15

After a Hundred Years

During the century that has elapsed since the signing of the Treaty of Guadalupe Hidalgo, the borderlands have been twice invaded: by AngloAmericans from the north and east; and, in our time, by Mexican immigrants from the south. The first invasion took place under the shadow of an approaching war between Mexico and the United States; the second invasion culminated at a time when a state of undeclared war existed between the two nations. Throughout this period of a hundred years, relations between Anglos and Hispanos in the Southwest have been affected by the state of relations between the United States and Mexico. In fact, a prime condition to an improvement in Anglo-Hispano relations has always consisted in a clarification of relations between the two nations. After a hundred years, this clarification of relations has finally been achieved—in broad outline, in first principles. As relations between the United States and Mexico have been stabilized, on the basis of mutual dependency and respect, significant steps have been taken toward an improvement of Anglo-Hispano relations in the borderlands. It all began on a cold, raw day in March 1933, when Franklin D. Roosevelt, in his first inaugural address, pledged the United States to the policy of the Good Neighbor.

1. A Beginning Is Made While the Office of the Coordinator of Inter-American Affairs was apparently established without any thought of using it to improve Anglo-Hispano relations in the Southwest, Mr. Nelson Rockefeller was soon deluged with suggestions that it be used in this manner. To many people living in the Southwest it seemed obvious that here was the logical place to invest the Good Neighbor Policy with real meaning and content. On October 15, 1941, as commissioner of immigration and housing in California, I submitted to Mr. Rockefeller a plan for the improvement of Anglo-Hispano relations and similar suggestions were transmitted by Dr. Joaquín Ortega, of the University of New Mexico; by Dr. George Sanchez of the University of Texas; by Dr. W. Lewis Abbott, of

224  North from Mexico

Colorado College; by Dr. Ben Cherrington, of the University of Denver; and by Mr. C. J. Carreon, a Mexican-American member of the Arizona legislature. After some initial hesitation, the coordinator sent Mr. David Saposs to make a survey of conditions in the Southwest. Upon receipt of his report on April 3, 1942, the Spanish-Speaking People’s Division was established as part of the Office of Inter-American Affairs to stimulate and coordinate public and private rehabilitation programs aimed at preparing the Spanish-speaking to participate more actively in American life and to educate the English-speaking to the necessity of eliminating discriminatory practices injurious to the war effort and to our relations with Spanish America.1 Unfortunately the coordinator had great difficulty in making up his mind about the real function of the new division. Limited funds were wasted in trifling ballyhoo campaigns of one kind or another and too often the division functioned as though its prime objective were to induce Anglo-American clubwomen to sponsor Latin-American “fiestas.” In many ways, the division acted as though it wanted to frustrate any real efforts on the part of Spanish-speaking people to improve their lot. Some of the field representatives seemed to be actually afraid of Mexican-Americans, for they insisted on working with the least representative elements in the various Spanish-speaking communities. On the board of the Southern California Council on Inter-American Affairs, subsidized by the CIAA, not a single Spanish-speaking person appeared although a local Mexican Affairs Coordinating Committee was set up to advise the council. This cleavage perpetuated, of course, the basic fault in Anglo-Hispano relations. A sample of items taken from the bulletin of the council will show how public funds were used to promote the Good Neighbor Policy: a cocktail party for Alfred Ramos Martínez, the Mexican painter, at the Hatfield Dalzell galleries; a breakfast for the Latin-American consular corps; a Pan-American “fiesta” at a local high school featuring José Arias and his Latin-American Troubadores; a cocktail party for the Latin-American consular corps at the Jonathan Club; an exhibit of New Mexico santos and bultos at the Southwest Museum; an institute on community relations which, the bulletin proudly notes, was reported “in a series of colorful and gossipy stories” by Princess Conchita Pignatelli in the Los Angeles Examiner; a series of lectures, illustrated “with many unusual colored slides” entitled “Travelling South Through Colombia, Ecuador, Peru, Bolivia, and Chile”; and an endless series of “goodwill” proclamations and radio programs stressing the necessity of Latin-American support in the war effort. Feeble as these efforts were, the establishment of the Spanish-speaking section of the CIAA must be regarded as a landmark in Anglo-Hispano relations in the Southwest for it constituted a recognition, however belated, that the United States had not fulfilled its obligations under the Treaty of Guadalupe Hidalgo. Also, thanks largely to the work of Mrs. Jane W. Pijoan, not all of the

After a Hundred Years   225

division’s funds were wasted on cocktail parties, receptions, and window displays of the flags of Latin-American countries. Field representatives were stationed in Los Angeles and Austin; grants-in-aid were made to a number of established institutions in California, Texas, New Mexico, Colorado, and Michigan; in cooperation with the Institute of International Education some ten or fifteen fellowships were provided for Spanish-speaking students; and a series of important and useful conferences were organized in the Southwest. Patterned on the Northern Colorado Conference on the Problems of the Spanish-Speaking People held in Greeley on May 10, 1942, the Coordinator’s Office organized a larger conference in Denver in June and July 1943; in Santa Fe in August 1943; and at the Arizona State Teachers College in Temple in June of the same year. Effectively organized and well-attended, these conferences initiated action programs which have had a wide influence throughout the region. At the University of New Mexico, four one-week institutes were organized for the training of Spanish-speaking rural leaders and teacher “workshops” were promoted in many cities. With funds provided by the Coordinator’s Office, the National Catholic Welfare Conference organized its first seminar on “The Spanish-Speaking People of the Southwest and West,” in San Antonio, in July 1943, which was followed by similar conferences in Denver, Santa Fe, and Los Angeles. These conferences represent, to the best of my knowledge, the first region-wide effort by the Catholic Church to concern itself with the problems of the Spanishspeaking people. While individual Catholic leaders, like Archbishop Lucey in San Antonio, have long been concerned about the Spanish-speaking people, the same cannot be said of the Church as a whole. The principal institutional influence in the life of the Mexican-American, its main contribution to a solution of “the Mexican Problem” has been a policy of religious nationalism and exclusiveness which has further isolated the Mexican from the general community with which he must some day make his adjustment.2 Several points should be noted about all these conferences and institutes. Beginning about 1943, that hardy perennial “the Mexican Problem” began to give way to a discussion of “The Spanish-Speaking People of the Southwest.” Whereas communities had formerly been preoccupied with some situation in their backyards, they now began to realize, for the first time, that the phase of the problem with which they were concerned was related to a much larger situation throughout the Southwest. The particularistic view, in which the camera-eye was focused on some specific Mexican shacktown, began to give way to the generic view, in which the camera swept the whole panorama of the Southwest. Public attention began to focus on Anglo-Hispano relations in the region, not as an intramural, domestic concern, but as an integral part of the much larger question of finding the basis for a new accord between the Anglo part of the hemisphere and the Spanish part.

226  North from Mexico

Once events were seen in this perspective, thoughtful residents of the borderlands began to discuss, not “the Mexican Problem,” but the Anglo-American problem. In the past, the isolation of the Spanish-speaking along a broken border had obscured the obvious fact that Mexicans were a large minority throughout the region and not merely “a local problem” in a few communities. Communities in Southern California, for example, still think of the Mexicans in their midst as though they were the only Mexicans in the United States. As more and more Mexicans began to participate in “Good Neighbor” conferences and institutes, the discussion shifted from a probing of conditions long deplored to a consideration of ways and means by which the Mexican people themselves might be given a chance to improve these conditions. Although the shift in attitude is difficult to describe in a phrase, one might say that the “social work” approach has gradually been replaced by the “social action” approach. What the “Good Neighbor” conferences and institutes reflected, of course, was the rise of a new and a more general interest in the Spanish-speaking people as an ethnic group.3 The interest, moreover, began to be extended in depth as well as in general range. The studies which Dr. Michael Pijoan made of nutritional factors in “the Mexican Problem” threw a glaring light on such questions as the ability of Spanish-speaking children to learn as rapidly as AngloAmerican children.4

2. Grassroots Democracy Of the various projects sponsored by the coordinator, none was more significant than the formation of the “service clubs” in Colorado. With funds provided by the government, the Colorado Inter-American Field Service Commission was organized in the fall of 1944. By January of 1945, Mrs. Helen L. Peterson, the field director, had organized the first Latin-American service club in Rocky Ford. Before the year was over, eight additional clubs had been organized in Pueblo, Walsenburg, Trinidad, San Luis, Alamosa, Monte Vista, Greeley, and Taos. Fourteen of the clubs are now banded together in the Community Service Clubs, Inc., which publishes the Pan-American News in Denver. When the Coordinator’s Office was liquidated in December 1945, the program was taken over by the Institute of Ethnic Affairs. The Colorado service clubs are an attempt to stimulate grassroots democracy in Spanish-speaking communities. Unlike various left-wing efforts to organize the Spanish-speaking people—such as the ill-fated Spanish-Speaking Congress of 1939—the service clubs have their roots, not in international politics, but in the basic needs of the Spanish-speaking communities. Starting with some simple issue, the clubs have taken up, one at a time, the problems closest to the people. They have conducted campaigns to register voters; they have sponsored scholarships for Spanish-speaking students; financed community

After a Hundred Years   227

health surveys; brought about the elimination of discriminatory practices; fought for better recreational facilities; and secured their rights for MexicanAmerican veterans. One of the clubs has initiated a $60,000-medical-care program in the San Luis Valley. Tackling the problem of Mexican-American truants, the Rocky Ford Club succeeded in returning ninety percent of the youngsters to the schools in a week. In developing the service clubs, all that Mrs. Peterson has done has been to provide the initial leadership, direction, and guidance. Long eager for some such program, the people have quickly responded and have accomplished significant improvements in a remarkably brief space of time. In Southern California a similar grassroots, local type of organization has developed under the brilliant leadership of Ignacio López in the Pomona Valley. For all practical purposes, the Unity Leagues which he has organized are similar in purpose and function to the service clubs in Colorado. With Mexican-American veterans playing a leading role, the Unity Leagues have been consistently concerned with the most immediate and obvious needs of the Mexican communities. Indigenous and organic, they represent grassroots democracy at its best, for it is their purpose to enlist the energies of the people. When this approach has been adopted, leaders have been found in Mexican communities without much trouble. In essence what the Unity Leagues have done, as Ruth Tuck puts it, is to “sprinkle the grass roots.”5 In 1946–1947, the American Council on Race Relations sent Fred W. Ross, an extremely talented grassroots organizer, into the citrus belt communities in Southern California to expand the Unity League program. In a year’s work, Mr. Ross set in motion tides of interest and activity which have had the widest ramifications. In Chicago, the Mexican Civic Committee, organized by Frank M. Paz, functions in much the same way. What these activities foreshadow, of course, is a great awakening of the Spanish-speaking people in the Southwest which I feel certain will mature within the next two decades.

3. The Westminster Case Gonzalo Méndez, a citizen of the United States, had been a resident of the town of Westminster, in Orange County, California, for twenty-five years. Of immigrant background, he had come to be a moderately prosperous asparagus grower. There are two schools in Westminster: a handsomely equipped school with green lawns and shrubs for the Anglo-Americans; and a Mexican school whose meager equipment matches the inelegance of its surroundings. It was not the discrepancy between the two schools, however, that annoyed Gonzalo Méndez. Rather it was the fact, so he said, that he didn’t like the idea of his Sylvia, Gonzalo Jr., and Gerónimo, growing up with hatred in their hearts for the children who went to the beautiful school. In the nearby community of

228  North from Mexico

El Modeno, the two schools were side by side; but the Mexican youngsters were always served lunch at a different hour from the Anglo-American students. Concluding that this practice had gone on long enough, Méndez filed a suit in the federal courts on March 2, 1945, on behalf of some five thousand Mexican residents of the district, against the school officials of Orange County. Oddly enough, this issue had never been squarely raised before in California. The School Code permits segregation of “Indian children or children of Chinese, Japanese, or Mongolian descent,” but says nothing about Mexicans or Negroes. Without formal sanction, the practice of segregating Mexican children in the schools came about in California largely through default of any determined resistance on the part of Mexican-Americans. Once established, of course, the segregated schools were defended and rationalized. For example, the superintendent of one of the schools involved in the Méndez suit wrote a thesis in 1939 in which he defended segregation on the ground of “social differences” between the two groups; the higher percentage of “undesirable behavior patterns” among Mexican students (which one would assume it would be a function of the schools to correct); and the “lower moral standards” to be found in the Mexican group. In some cases, segregation was accomplished by a fancy gerrymandering of school districts; but the more common practice was to use the arbitrary linguistic device of assigning all children with Spanish names to a separate school. Occasionally the school authorities would examine the appearance of youngsters so as to prevent the offspring of a Mexican mother whose married name might be O’Shaughnessey, from slipping into the wrong school. While the practice varied from district to district, the general scheme was to segregate Mexicans from the first through the sixth, and in some cases through the twelfth, grade. In the trial of the Méndez case, the school authorities at first contended that Mexicans were a distinct and therefore an “inferior” race; but, confronted by the testimony of some world-famous anthropologists, they soon abandoned this position. As a matter of fact, it had been determined years ago—In re Rodríguez, 81 Fed. 337—that Mexicans of Spanish descent and of mixed Spanish-Indian descent were “white persons” within the meaning of the naturalization laws. The superintendent of schools then testified that Mexican children were “dirty”; that they had lice and impetigo; that their hands, face, neck, and ears were often unwashed (presumably nothing of this sort had ever happened with Anglo-American youngsters); and that, generally speaking, they were “inferior” to the other students in point of personal hygiene. In a memorable opinion handed down on March 21, 1945, Judge Paul J. McCormick ruled that segregation of Mexican youngsters found no sanction under the California laws and that it also violated the “equal protection” clause of the Fourteenth Amendment. Segregation, Judge McCormick suggested,

After a Hundred Years   229

might have something to do with the fact that Mexican youngsters were retarded in English speech. It also had the effect, he said, of “depriving them of a common cultural attitude . . . which is imperative for the perpetuation of American institutions and ideals” and of fostering antagonism. When the decision was appealed to the Ninth Circuit Court, amicus curiae briefs were filed on behalf of Méndez by the American Jewish Congress, the National Association for the Advancement of Colored People, the National Lawyers Guild, the American Civil Liberties Union, the Japanese-American Citizens League, and by Robert W. Kenny as attorney general of California. On April 14, 1947, the Ninth Circuit Court affirmed Judge McCormick’s ruling. In a brilliant concurring opinion, Justice William Denman exposed all the shabby rationalizations by which the school authorities had sought to justify their action. The segregation of Mexican students in the schools, he caustically noted, did not stand alone; on the contrary, it was part of a pattern of discrimination and could not be justified in pedagogic terms however facile. For, as he observed, the Rev. R. N. Nuñez, a Catholic priest, Eugenio Nogueros, a college graduate, and Ignacio López, a newspaper publisher, had been forced to file a suit in the federal courts to enjoin the officials of nearby San Bernardino from barring “Latins” from the public swimming plunges. This discrimination had nothing whatever to do with educational theories; nor had it anything to do with hygiene. For all “Latins” had been barred: clean or dirty, healthy or diseased, black or white; in fact, as Justice Denman pointed out, the prohibition was so broad as to have embraced the nationals of twenty-one South American nations, Mexico, Italy, Spain, and Portugal. In closing his opinion, Justice Denman suggested that the school authorities of Orange County should be punished for contempt of court in having failed to carry into effect the decision of the lower court. The filing of this precedent-shattering case was in no sense “inspired.” Outside organizations provided valuable assistance in handling the trial and the appeal, but the case had been filed simply because Gonzalo Méndez had “had enough.” That the action was long overdue is shown by the manner in which one Mexican community after the other immediately raised the same issue. In a dozen or more communities similar suits were filed or movements launched to eliminate segregated schools; and in El Modeno the Mexicans followed up their victory in the courts by electing one of the group a member of the local school board. Forty-five Mexican-American and thirty Negro families make up the little settlement of Bell Town near Riverside, California. The school in Bell Town is a four-room wooden structure built over twenty years ago. Two miles down the road is an attractive stucco “Spanish-style” modern school with excellent equipment and every teaching facility. Only Negroes and Mexicans attended the Bell Town school; while the other school was “lily white” with the exception of three

230  North from Mexico

Mexicans and one Negro student who lived so close to the school that they could not be excluded on any pretense. “White” children, regardless of where they lived, were invariably assigned to the better school. Hearing about the Méndez case, the residents of Bell Town decided that they, too, had “had enough.” Under the guidance of Fred Ross they founded the Bell Town Improvement League and petitioned the authorities to do away with segregation. In this instance, the school officials said nothing whatever about cleanliness or backwardness or godliness but frankly stated that the encroachment of Mexicans and Negroes depreciate property values. After a long fight, the residents of Bell Town won out. On September 16, 1946, the supervisor of schools told his staff: “If there is as much as one segregated MexicanAmerican pupil see to it that he gets unsegregated immediately.” Throughout Southern California, Mexican-Americans have been moving toward a new awareness, a new consciousness of their rights as citizens of the United States. Chavez Ravine, in the City of Los Angeles, has a large Mexican population. In June 1946, the 4,500 residents announced that they had been “walking and walking for years and years and now we’re very, very tired” and proceeded to form a civic organization and demand that the area be provided with the bus service which it had lacked for twenty-five years. Today buses are running in Chavez Ravine. When the Mexican-American veterans returned to Clearwater, where most of the Mexicans lived on Illinois Street, they decided that it was about time, after all these years, that the street was paved. Today Illinois Street is paved. This same process is at work all over Southern California: streets are being paved; lights are being turned on; buses are running; and Mexican children are beginning to attend general schools along with other children.

4. “Utilizable Cultural Residues” Part of the change that is taking place in Anglo-Hispano relationships in the Southwest can be traced to the new interest and leadership that has developed at the colleges and universities. As early as 1912, Dr. E. D. Gray wanted to found a Spanish-American university in New Mexico, for he regarded the bilingual population as a national and international asset. Today, thanks to the leadership of Dr. Joaquin Ortega, the University of New Mexico has a School of Inter-American Affairs which is training students who want to work in LatinAmerican countries as teachers, businessmen, and technicians. “New Mexico,” writes Dr. Ortega, “still possesses utilizable cultural residues.” It has the most homogeneous Spanish-speaking and the most cohesive Indian communities in the United States as well as a typical cross-section of AngloAmericans and other immigrants. Here is the place, he has insisted, to study the process of acculturation in the Americas for nowhere else can one find the

After a Hundred Years   231

three major cultural groups—Indians, Spanish, and Anglo-Americans—living together in large numbers with a common national allegiance yet maintaining their traditional cultures. “New Mexico,” Dr. Ortega has also said, “is the shortest route to Mexican goodwill,” and to the goodwill of all Latin-America. During the war, we had no more effective spokesman on our radio programs broadcast to South America than Senator Dennis Chavez of New Mexico. While we have not yet learned to utilize the cultural resources of the Southwest, some steps have been made in the right direction. We discovered during the war that Pueblo Indians from New Mexico made excellent spokesmen for the United States among the Indians of South America. And we have begun to use a few Spanish-speaking people in the diplomatic and consular services (Daniel Valdez, who did one of the first studies of the Spanish-speaking people in Colorado, was recently appointed attaché to the American embassy in Montevideo). In emphasizing the importance of understanding and developing these resources, the University of New Mexico has played a prominent part. In 1936 the university came into possession of the Harwood Foundation through a gift from Mrs. Lucy C. Harwood and immediately set about using the foundation as the spearhead for an educational program designed to serve the needs of Spanish-American villagers. By way of preparing this program, Dr. George Sanchez made a study of Taos County which was published in 1940 under the title of Forgotten People. The publication of this extraordinarily fine, sensitive, and perceptive study of Spanish-American culture might be said to mark a new chapter in the history of the Spanish-speaking people of the Southwest. Based on this study, the university then launched the Taos Project of community and adult education which has attracted nationwide attention. Essentially a self-help, cooperative, community organization project, the Taos Project has some impressive, if limited, accomplishments to its credit.6 At about the same time, Mr. and Mrs. Cyrus McCormick made a grant to enable the University to revise the work of a small rural school in the village of Nambe. This school has since become an important “pilot” school in devising better teaching methods for Spanish-speaking children.7 All of these projects, as well as the founding of the San Jose Training School in Albuquerque in 1930—again to improve teaching methods—and the fine work being done at New Mexico Highlands University at Las Vegas by Quincy Guy Burris— indicate the kind of leadership the universities have shown in the last decade. Similar developments have taken place at the University of Texas, particularly since the arrival there of Dr. George Sanchez. The Committee on Inter-American Relations in Texas has sponsored such excellent studies as Dr. Wilson Little’s report on Spanish-Speaking Children in Texas (1944), which has been the basis for many improvements in the educational system. Under the guidance of the university, the State Department of Education is now

232  North from Mexico

making a serious effort to improve the teaching of Spanish in the public schools and free textbooks are furnished for teaching Spanish at every grade level. A summer school is now conducted by the university in Mexico City and an Institute of Latin-American Studies has been founded at Austin. Recently the First Regional Conference on the Education of Spanish-Speaking People in the Southwest was held at Austin (December 13–15, 1946)—one of the first attempts to consider the education of Spanish-speaking people as a regionwide problem in inter-American education. From this conference has come the Southwestern Council on the Education of the Spanish-Speaking Peoples whose recommendations have begun to have a wide influence in the Southwest. Some fifteen years ago, J. Frank Dobie said that there were only about twenty or thirty Mexican-American students at the university; today one hundred and fourteen are enrolled.* All this activity of the last decade—the new role being played by the universities, the calling of conferences and institutes, the appointment of Good Neighbor commissions in Texas, and so forth—indicates that leaders of opinion in the Southwest have come to recognize that in discussing the education of Spanish-speaking people and related issues, they are dealing with aspects of a unitary region-wide problem which cannot be precisely correlated, say, with the assimilation of Italian-Americans in New Haven. For the Spanish-speaking minority in the Southwest, rather like the Negro issue in the Deep South, presents a problem in masses. Spanish-speaking people in the borderlands are the fringe of great masses of Spanish-speaking people in Mexico, Central America, and South America. To regard them as “merely another minority” is to gravely minimize the significance of the borderlands as a bridge to inter-American understanding.

5. From De Anza to Juan López The movement traced in the foregoing section should be regarded as the latest chapter in the Anglo-Americans’ belated discovery of the Southwest. The oldest settled portion of the United States, the Southwest is the newest in point of Anglo-American interest. In this sense, it was discovered by Hubert H. Bancroft and Bandelier in the 1880s. Bancroft’s History of the North Mexico States and Texas appeared between 1884 and 1889; his History of California from 1885 to 1891; and his History of Arizona and New Mexico in 1889. Bancroft worked with documents; the man who really “discovered” the Southwest, as a cultural province, was Adolph F. A. Bandelier. Trained in geology at the University of

* “See Texas-Born Spanish-Name Students in Texas Colleges and Universities: 1945–1946, by Ruth Ann Fogartie, University of Texas Press (March 1948).

After a Hundred Years   233

Bern, Bandelier came to New Mexico in 1880 and in the course of a decade visited every nook and cranny of the Southwest. His Final Report on the Southwestern United States, Part II, appeared in 1892. These works by Bancroft and Bandelier, along with Frank W. Blackmar’s Spanish Institutions of the Southwest (1891) and Charles Fletcher Lummis’s The Land of Poco Tiempo (1893), represent the initial Anglo-American awareness of the cultural importance of the Southwest. Those who first discovered the cultural riches of the Southwest emphasized, quite naturally, its archaeological and antiquarian interests. Largely as a result of Bandelier’s work, the Archaeological Institute of America selected Santa Fe as the headquarters for its School of American Research in 1908. At about the same time, the influx of writers, artists, and intellectuals began, when Ernest Blumenschein and Oscar Berninghaus arrived in Taos in a covered wagon. Through the work of the Santa Fe and Taos artists, the nation became increasingly aware of the cultural importance of the Southwest. First discovering the Indians, the artists and writers gradually began to discover the Spanishspeaking. From 1920 to 1930 the native New Mexicans were “discovered” with a vengeance. Just as the somewhat earlier discovery of the Indians found expression in a movement primarily aimed at the reconstruction of Indian arts and crafts, so the discovery of the Hispanos coincided with a movement aimed at reconstituting the Spanish-Colonial handicraft arts. What both movements lacked was a social program by which the basic economy of the Indians and the Spanish-speaking might be reconstructed, for the arts and crafts could only flourish as the culture was vigorous and life sustaining. This missing element—the social program—was supplied by the New Deal agencies in New Mexico in the thirties. Faced with heavy Spanish-speaking relief loads, these agencies were compelled to undertake, in conjunction with the Bureau of Indian Affairs, a thoroughgoing survey of natural and human resources in the Rio Grande Valley. The monumental Tewa Basin studies initiated by the Soil Conservation Service represent perhaps the first serious attempt to view the whole human scene in New Mexico and to ferret out the correlations between Anglo, Hispano, and Indian influences in the region. Much of the present-day interest in Spanish-speaking people in the Southwest can be traced to the work of federal agencies and federal officials in New Mexico in the period from 1933 to 1940. If the Indians received more attention than the Spanish-Americans, it was because they were somewhat better organized to demand attention for their problems. But both efforts—to rehabilitate Indian life and to reconstruct Spanish-American communities—were and still are closely interrelated. What has happened in this process of discovery is that the focus of the Anglo-Americans has finally come to center in the contemporary scene. An interest in mission ruins and Indian relics has been known to lead to an

234  North from Mexico

interest in Mexicans and Indians. Hence the focus of interest has gradually shifted from the entrada of Juan de Oñate in 1598 to life in the village of Truchas in 1947; from Junipero Serra’s first celebration of the mass in San Diego to a concern with the present-day Mexican-American colonias. Similarly an interest in Spanish-Colonial arts has gradually ripened into an interest in the handling of the chili crop and in tenant-herding in the Cuba Valley. The nearer the focus has shifted to the contemporary scene, the more the Anglo-Americans have been surprised by their discoveries. Nowadays, a hundred years after the signing of the Treaty of Guadalupe Hidalgo, they have finally begun to study the actual social structure of the widely varying Mexican-American communities to be found between Brownsville and Los Angeles. From studies of this sort has come the realization, as Lee Casey puts it, that perhaps the Anglo-Americans should “do a little adjusting themselves.” It is significant that the growing maturity of Mexican-American communities throughout the Southwest is closely related, in point of time, to the appearance of this new interest, not in Juan Bautista de Anza, but in Juan López.

16

“One and Together”

“No one who has grown up in California,” wrote Josiah Royce, “can be under an illusion as to the small extent to which the American character, as here exemplified, has been really altered by foreign intercourse, large as the foreign population has always remained. The foreign influence has never been for the American community at large, in California, more than skin deep. One has assumed a very few and unimportant California ways, one has freely used or abused the few (Spanish) words and phrases, one has grown well accustomed to the sight of foreigners and to business relations with them, and one’s natural innocence about foreign matters has in California given place, even more frequently than elsewhere in our country, to a superficial familiarity with the appearance and the manners of numerous foreign communities. But all this in no wise renders the American life in California less distinctly native in tone. . . . You cannot call a community of Americans foreign in disposition merely because its amusements have a foreign look.”1 To Royce—the most perceptive and sensitive of California historians—this summation seemed quite clear and obvious in 1886. But it was not quite accurate, even then, to say that “the California ways” which had survived were “few and unimportant.” More deeply than Royce imagined, the customs, the laws, and the economic practices and institutions of the native Californians had exerted a definite influence on the culture which began to emerge after the American conquest. However it did appear in 1886 that the native Californians had suffered an irreparable defeat and that the initial contact between the two cultures had resulted in the eclipse of the one without any substantial modification in the basic pattern of the other. But today the ineluctable facts of geography and history dictate a somewhat different conclusion. Mexico is not France or Italy or Poland: it is geographically a part of the Southwest. Residing in the Mexican states immediately south of the border are approximately 2,500,000 Spanish-speaking people; in the American border states approximately the same number of Spanishspeaking reside. Essentially these are one people, occupying a single cultural

236  North from Mexico

province, for the Spanish-speaking minority north of the border (a majority in some areas) has always drawn, and will continue to draw, support, sustenance, and reinforcements from south of the border. Our Spanish-speaking minority is not, therefore, a detached fragment but an integral part of a much larger population unit to which it is bound by close geographic and historical ties. Furthermore, Hispanic influences in the United States have a strong anchor in New Mexico where these influences are actually older, and perhaps more deeply rooted, than in the Mexican border states. The Spanish-speaking and the Indians of the Southwest have the highest birthrates of any ethnic groups in the region. Infant mortality rates are declining, for both groups, throughout the borderlands: between 1929 and 1944 the rate decreased in New Mexico from 145.5 infant deaths per 1,000 live births to 89.1. With high birthrates and rapidly declining infant mortality rates, the Spanish-speaking element will retain its position relative to Anglo-Americans for many years to come, barring unforeseeable contingencies. These facts alone would indicate that the Hispanic minority cannot be regarded as merely another immigrant group in the United States destined for ultimate absorption. In this instance, however, demographical considerations are fortified by the facts of geography and the implications of history. While Spanish cultural influences have retreated in portions of the Southwest, they have never been eclipsed. “Whether they will or not,” wrote J. P. Widney in the 1880s, “their future [that is, the future of Anglos and Hispanos] is one and together, and I think neither type of race will destroy the other. They will merge.” With the Spanish-speaking element having been reinforced by a million or more immigrants in the last forty years, virtually all of whom have remained in the Southwest, some type of cultural fusion or merger must result. In fact, a surprising degree of fusion has already taken place.

1. By Any Other Name The development of speech and language patterns not only mirrors the relationships between Anglos and Hispanos in the Southwest but is the best gauge of the degree of cultural fusion that has occurred. Needless to say, I discuss this highly complex subject not as a linguist; nor in terms of its linguistic interest; but rather to indicate what has actually happened to the two cultures in the region and to trace a relationship. For the attitude of a minority toward language and speech has an important bearing on the direction that the process of acculturation is likely to take. The language pattern in the Southwest has, of course, a number of variable factors. It varies in relation to the numerical proportion between the two groups in anyone place; the age of the community; whether it is rural or urban; the degree of isolation; the history of social relations in the community and

“One and Together”   237

many other factors. Quite apart from these variations, however, there is a larger aspect to the language pattern which can be considered from three points of view: Spanish borrowings from American-English speech; Anglo-American borrowings from the Spanish; and the development in both groups of a kind of jargon which is more “Southwestern” than Spanish or English. In 1917 Dr. Aurelion M. Espiñosa listed some three hundred words of Anglo-American origin which have been incorporated into the Spanish language as spoken in New Mexico after first being Hispanicized. Most of these words had been borrowed from necessity rather than choice, for they related, in the main, to commodities, practices, things, and concepts for which there was no Spanish equivalent (at least not in the Spanish spoken in New Mexico). Many of them had to do with commercial, industrial, and political practices unknown to the Spanish population prior to the American conquest as shown by the fact that more than fifty percent of the terms had been incorporated after 1880. In large measure the adopted words had to do with “work terms” related to the new jobs which New Mexicans had acquired; others related to slang expressions used in American sports. Obviously most of this borrowing was based on strictly utilitarian considerations.2 In another study of word-borrowing, Dr. Manuel Gamio listed the following among many terms that had been hispanicized: picnic, laundry, ties (railroad ties), matches, stockyards (estoque yardas), groceries, lunch, tickets, depot, time-check, truck, truck-driver, biscuit, omelette, bootlegger, taxes, ice cream, board and boarder, boss, automobile, sweater, jumper, sheriff, etc. In still another list, Dr. Harold W. Bentley added: home run (jonronero); scraper (cscrepa); plug (ploga); puncture (ponchar); jack (llaqui); and such expressions as “vamos jlat”—to have a flat tire (literally, “we go flat”). Generally, the Spanish-speaking people have borrowed from necessity rather than choice and have shown either resistance or indifference to other types of borrowings. Still the number of such borrowings, from necessity or otherwise, has been substantial and would probably be much greater today than when these studies were made. According to Alfred Bruce Gaarder, there are four types of Spanish spoken in the Southwest: the Spanish spoken by the “old folks,” particularly in New Mexico, which contains many archaic forms known only to the sixteenth century; the language of the “middle generation” which keeps some of the archaic and obsolete forms but adds a large vocabulary of Anglicisms developed to meet the needs of trade or business; the speech of the “youngest group” which increases the confusion by the use of slang expressions current among their schoolmates; and, lastly, the jargon of the city gangs, identical with the third grouping above, except that expressions of a shady, sinister, or double meaning have been added; often this jargon is used as a secret language.3 It is in the speech of the city gangs, “the pachuco patois,” that the attempt to fuse the two languages is most clearly apparent. For these youngsters play

238  North from Mexico

wonderful variations on both languages, Anglicizing Spanish and Hispanicizing English as it suits their purpose and often coining an expression of their own. Here are some of their inventions or fusions as reported by Dr. Gaarder: Pachuco

Spanish

English

bolar

dólar

dollar

borlo

baile

dance

calco

zapato

shoe

caldiarre

enojarse

to be angry

canton

casa

house

carlo

caló

cant

carnal

hermano

brother

carnala

hermana

sister

carrucha

automóvil

automobile

chero

policía

police

duro

dólar

dollar

frajo

cigarro

cigar-cigarette

grena

peló

hair

greta

marijuana

marijuana

huisa

muchacha

girl

jando

dinero

money

jetiar

dormir

sleep

jura

policía

police

mostacho

bigote

mustache

pildora

policía

police

rolante

automcívil

automobile

simon



yes

tonda

sombrero

hat

vesca

marijuana

marijuana

Other pachuismos are: anteojos, front window; bote, jail; chante, house (probably from shanty); choque, chalk; chillar, to cry; escamado, frightened; jefa, mother; jefe, father; lira, guitar; tambo, jail; tambique, jail; tramo, suit. The second generation uses many slang expressions also found in the talk of the pachucos: aleluyas, converts to Protestantism; birria, beer; bollío, an American; bolucha, or bolita, picking oranges; brecas, brakes; chapos, Japanese; cho, a movie; chutear, to shoot; cuivo, hello; datil, a date; diez y penny, a

“One and Together”   239

five-and-ten store; en gascado, in love; esa, hello (to a girl); ficha, money; fila, or filero, a razor; gavacho, an American; ginar, to commit robbery; gua he, a guitar; guayn, wine; jalar, to work; lorcha, a match; lucas, crazy; manil, money; mono, a movie; nagualones, imported worker from Mexico; pistiar, to drink liquor; rolar, to sleep; ser maleta, to see a movie; sut, suit; tintos, Negroes; and trola, a match.4

2. Words That Fit Anglo-American borrowings from Spanish have also been dictated by necessity, in many cases, but from other motives as well. One important grouping of borrowed words has to do with things and practices for which there was no English equivalent, as in the cattle industry, the mining industry, and in the pack-train business (as I have already shown). But, in addition to these borrowings-by-necessity, there is a long list of Spanish words which have apparently been taken over for local color, humorous effect, and, above all, for their appropriateness in an arid environment. In his Dictionary of Spanish Terms in English (1932), Dr. Bentley lists some four hundred words which have been incorporated in the English spoken in the Southwest. Actually the list is much longer than linguists such as H. L. Mencken and George Philip Krapp have indicated, for they have not been looking in the right quarter, namely, the Southwest. Considering that we had just fought a war against Mexico, it is indeed remarkable that so few Spanish place-names were changed after 1846. In addition to the names of rivers and mountains, Dr. Bentley states that there are two thousand or more cities and towns in the United States with Spanish names: four hundred or more in California; two hundred and fifty in Texas and New Mexico; and a hundred or more in both Colorado and Arizona. In Colorado the name of the state and the names of nineteen counties are Spanish. Spanish placenames also appear, with less frequency, in such states as Nevada, Wyoming, Utah, Oregon, Montana, and Idaho; in fact, they appear in every state in the union. Often the original Spanish has been Anglicized, as in Waco, California (originally Hueco); and, in many cases, Spanish and English terms have combined, as in Buena Park, Altaville, and Minaview. There are eight “Mesas,” four “Bonanzas,” and thirteen “El Dorados” in the United States. Many of the Spanish place-names outside the Southwest refer to the names of battles or of events related to the Mexican-American War. In the Southwest most of the Spanish place-names were preserved—in my opinion—because of their extraordinary appropriateness and beauty. The Spanish named places with the uncannily descriptive accuracy of poets. For example, who could improve on “Sangre de Cristo” for the name of the great range of mountains in northern New Mexico? The very persistence with which

240  North from Mexico

resident Spanish-speaking people kept calling mountains, rivers, and towns by their Spanish names must, also, have been a factor. In Southern California, virtually all the Spanish place-names were retained; but many of the street names, in places such as Santa Barbara, San Diego, and Los Angeles, were changed or Anglicized after 1846. Long familiar with an arid environment, the Spanish gave vivid and accurate names to the novel features of the Southwestern landscape. “The acequia madre (“mother ditch”) of every village,” writes T. M. Pearce, “has almost a personality of its own. It becomes the most intimate friend of every inhabitant of the place. With dancing and ceremony, the acequias are opened in the spring . . . with scrupulous care the acequias are scraped and strengthened in the villages and towns.” To call these life-giving main canals “ditches” would have been to minimize their importance in this environment. And so it is with many similar expressions, relating to the natural environment of the Southwest, which were retained and incorporated into AngloAmerican speech. The list is a long one indeed and includes such words as: malpaís, mesa, vega, cumbre, bosque, sierra, pozo, hondo, loma, bajada, ciénaga, piloncillo, potrero, arroyo, laguna, barranca, cañon, llano, brasada, chaparral, canada, and many others. “The Southwest,” writes Pearce, “with its peculiar brilliance of day and quick shadows of night-fall, with its hardbaked earth and sudden water gushes, with its thirsty sands at the very edge of soggy river bottoms, cannot be described in terms of Shakespeare’s Stratford.” For example, an arroyo is not a gully. As Pearce points out, “it is a bare rent in the side of Mother Earth where only yellow jaws yawn until a cloudburst in the mountains miles away sends the lashing torrents hurtling through it to crush and engulf everything caught in its maw.” The word malpaís means more than “badlands”: it refers to the lava ridges or serrated volcanic ash “dumps” to be found in the Southwest. It is quite impossible to convey the peculiar significance of ciénaga, as used in the Southwest, by some such expression as “marshy place,” for the latter does not carry the connotation of an encompassing aridity. Thus vega is not just “meadow”; bosque is more than “a clump or grove”; and sierra carries overtones of meaning not suggested by “saw-toothed range.” It is not by chance, furthermore, that so many Spanish names for trees, plants, and shrubs have been borrowed in the Southwest. Many of these items have never had a name other than that given them by the Spaniards: grama, sacaton, aparejo, alfilaria (grasses); mesquite, chaparral, chamiso, sahuaro, palo verde, huisache, mogote, maguey, manzanito, bellotas, álamo, tule, amole, capulin, plumajillo, and piñones, (for trees, shrubs, and plants). Similarly, the AngloAmericans borrowed many terms related to the type of architecture they found in the Southwest: portal, corbel, adobe, fogón (three-cornered fireplace), ramada (shaded arbor), azotea (the flat, platform-like roof), cabana, casa grande, hammock, presidio, hacienda, jacal, patio, placeta, plaza, viga, palacio, zaguán (open

“One and Together”   241

passage way), cañales (roof gutters), trostera (large cupboard), and many others. Many animals and insects—tecolote (owl), coyote, cucaracha (cockroach), javalina, mosquito, cougar, lobos, jaguars, conejos (rabbits), venalos (deer), are Spanish in origin or, in the Southwest, are known by their Spanish names. Names for items of dress have likewise been appropriated: sombrero, tilma, rebozo, manta, and sarape. From the Spanish-American War came rurale, machete, ley juga, mañana, hoosegow (juzgado), and a number of other words and expressions. From the contact between the British and Spanish navies came such words as armada, cask, cork, and cargo. And then there is, of course, a long list of words, Spanish in origin, which have become fully “naturalized”: vigilante, filibuster, avocado, barbecue, cockroach, corral, creole, tobacco, cannibal, vanilla, hammock, tornado, alfalfa, canary, cigar, maroon, Negro, palaver, paragon, parasol, sherry, soda, canoe, banana, alligator, cocoa, sassafras; as well as many words that came by way of South America: alpaca, armadillo, chinchilla, cocaine, condor, cougar, jaguar, llama, and tapioca. Spanish borrowings from American speech have naturally been most numerous in the speech area along the Rio Grande and immediately south of the border; while American borrowings have been most common throughout the old Spanish borderlands area. Many Southwestern words and idioms are Spanish in origin: jerky, hackamore, buckaroo, mustang, stampede, lariat, fandango, hoosegow, wrangler, desperado, vamoose, hombre, adios, agua, bandido. In this area, the corruption of Spanish has been paralleled by the cultivation of what Frank Dobie calls “sagebrush” or “bull-pen” Spanish. Most of the borrowings, on both sides, have been by ear for neither group has been a serious student of the language of the other. In the isolation of the region, each group borrowed from the other so that today part of the vocabulary of the Southwest is bilingual in origin. A kind of Spanish is still spoken in the range country along the border where, according to Doris K. Seibold, fully half the cowmen are bilingual.5 Most of the Mexicans born in the region since 1900 are, of course, bilingual. Cowboy talk is so thoroughly bilingual that, in a single issue of Lariat, a popular “western” or “cowboy” magazine, Dr. Bentley found 376 Spanish words or words of Spanish origin.6 Most authors who have written about the Southwest have felt compelled to include a glossary of Spanish terms in common use.7

3. Neighbors in Isolation Considering the degree of hostility which has prevailed between Anglos and Hispanos in the Southwest, the extent of cultural fusion which has already occurred is most surprising. The isolation of both groups has been a prime reason for this mutual borrowing and adaptation. In the absence of deeply rooted

242  North from Mexico

educational institutions, the borrowing has been unconscious, careless, and natural—a product of intimacy in isolation. However antagonistically each group may have regarded the other, the plain fact is that they have been in continuous, direct, and often intimate contact in the Southwest for over a century. There was a period, in all the borderland states, when the two groups existed, side by side, in the friendliest intimacy with mixed marriages being quite common. Dr. Espiñosa, and many other observers, have commented upon the fact that this intimacy was “much more frequent in the first years of the American occupation.” Since most of the mixed marriages of this earlier period involved Anglo-American husbands and Spanish-speaking wives, it is impossible to estimate the degree of intermixture which has taken place but it is much greater than most people imagine. This initial rapprochement came to an end when the railroads penetrated the Southwest. With the appearance of the railroads, wrote Dr. Espiñosa, “there has come a check in the race fusion and the mutual contact and good feeling between the two peoples.” Obviously it was the sequence of economic changes which the railroads initiated, not the railroads per se, which produced this effect. Previously the motive for dominance was largely lacking, for, in the absence of markets, a barter rather than a profit economy prevailed. In the isolation of the frontier both groups felt compelled to seek a degree of cooperation to mitigate the rigors of a harsh and unfriendly environment; on the frontier, as someone has said, “all churches look alike.” While this earlier intimacy came to an end, the relationships which came out of it could never be effaced. Thus there exists in the Southwest an antecedent pattern of fusion and merger which continues to exert an influence, however imperceptible, upon present-day relationships. It must also be remembered that the process of acculturation is somewhat different in the Southwest than elsewhere in the United States. Here we adopted the Spanish-speaking minority; they did not adopt us. It is this difference which accounts for the tenacity with which the Spanish-speaking have clung to certain aspects of their native culture. As late as 1917, Dr. Espiñosa estimated that there was not one Spanish-American family out of a hundred in New Mexico that had entirely abandoned Spanish as the language of the home. Nowadays nine out of ten of the native-born New Mexicans speak English but Spanish is still the mother tongue for most of them. The persistence of Spanish speech, however, is due to many factors: the prevailing isolation; constant discrimination; the lack of educational facilities; the existence of segregated schools; the migratory pattern of employment, and so forth. Whatever the reasons may be, the point is that this persistence in Spanish speech has been most influential in forcing a degree of cultural fusion. To appreciate the importance of this factor all one has to do is to compare the rapidity with which the Nisei or native-born Japanese have abandoned or lost all

“One and Together”   243

familiarity with the Japanese language. Regardless of length of residence, only a small proportion of Mexicans in the United States have lost all knowledge of Spanish or have abandoned its use in the home, no matter how meager their training in the language may have been or how imperfectly they may speak it. While Mencken and others have suggested that Southwestern Spanish is doomed to vanish, sooner or later, the facts would seem to cast grave doubts on this conclusion. This becomes more apparent when one considers the thorny issue of language instruction in the schools.

4. Who Is Being Stubborn? In the bundle of issues that is called “the Mexican Problem” none has occasioned more discussion and controversy than the language issue in the schools. Both the history and latter-day ramifications of this issue are most complex. Prior to 1846 the borderlands were without schools, public or private; illiteracy was the rule, literacy the exception. The first school systems were dominated, in administration and personnel, by Anglo-Americans who knew little or no Spanish. While official concessions were made to Spanish speech in New Mexico, school officials in the Southwest have always insisted upon English as the language of instruction. They still invest their position, on this issue, with an emotional halo of moral and patriotic self-righteousness. To a generation of American teachers trained in the normal schools of the period from 1890 to 1910, it seemed both heretical and disloyal, despite the guarantees of the Treaty of Guadalupe Hidalgo, to tolerate any form of bilingualism. In some areas, the issue has even been colored by religious prejudices of one kind or another. According to Dr. Ortega, Anglo teachers have actually changed the names of Spanish students, on the first day of school, to some English equivalent by way of emphasizing the “terrible handicap” that Spanish speech is supposed to be. In other cases, Hispano teachers in rural schools made up of Spanish-speaking children have used Spanish surreptitiously for fear of being called on the carpet by some irate Anglo administrator. The natural consequence of this official attitude has been to foster a generation illiterate in both languages, for the teaching of Spanish has been as systematically neglected as instruction in English has been systematically stressed. Spanish-speaking children often come to the schools without a word of English and without the environmental experience upon which school life is based. In many cases, they are not even familiar with the concepts for which they are supposed to learn English names. The use of standard curricula, books, and instruction materials in such schools has been ludicrously inept. Once Anglo-American teachers had “retarded” Spanish-speaking students, they sought to rationalize their incompetence as teachers by insisting on segregated schools which only aggravated the problem. Notoriously bad linguists,

244  North from Mexico

Anglo-American teachers have been known to show an unreasoning irritation over the mere sound of a Spanish word or phrase spoken in their presence. This irritation is often reflected in a hostile attitude toward Spanish-speaking students. Over a period of many years, I have heard Anglo-American teachers in the Southwest complain bitterly about the “stubbornness” of MexicanAmerican youngsters who just will persist in speaking Spanish on the playgrounds, etc. Actually the language issue in the Southwest is part and parcel of a much larger set of socioeconomic issues from which it cannot be separated. It is most absurd, therefore, to attempt to isolate this issue and to regard it as a special problem which might be solved, apart from the larger issues, by the development of special teaching techniques and especially trained personnel (important as these items would be). Obviously the issue is related to bad housing, lack of nutrition, migratoriness, social disorganization, segregation, dominant group hostility, and a dozen other factors. The language problem, in short, is a community problem; a problem involving the relationship of the school to the community and of the community to the school. Today the issue is widely recognized as the major educational problem in the Southwest. Furthermore, the conclusions now being drawn are indirectly premised on the assumption that a type of cultural fusion actually exists. “The intermingling of different home languages in the Southwest,” to quote from one report, “is a relatively permanent condition, for here the waters from two great reservoirs of language flow together, constantly renewed from sources back from the border.”8 (Emphasis added.) In other words, the borderlands have consistently remained the borderland of the two cultures; neither has prevailed in toto and neither is likely to win a complete victory over the other. Each group has gained recruits; the number of bilinguals is steadily increasing; and the area of fusion is expanding south and, to some extent, north of the border. Dynamic factors are involved in the extension of the borderlands for experience has shown that Mexican immigrants cannot be kept out of this area. We stopped Mexican immigration but imported 180,000 Mexican workers in wartime; workers are still crossing the border illegally, as they have for years, and the number of “wetbacks” is currently estimated at around 80,000. Regardless of how it changes, the Southwest is “an ever-normal granary” so far as Mexicans are concerned. Emphasis on the language issue as “relatively permanent” merely reveals the true cultural background of the region. There are more persons of Italian than of Mexican descent in the United States but no one has suggested that bilingual instruction is a major problem in the education of Italian-Americans. By insisting on regarding “the Mexican Problem” as part of a familiar Americanization process, identical with European immigration, we have consistently missed the point. Failing to recognize the degree of cultural fusion

“One and Together”   245

which has actually occurred, we have steadily belabored the Mexican for his “stubbornness” in adhering to a culture which prevails in the very areas of the United States in which he resides. It is as though we were to accuse the Eskimos of Alaska for their “stubborn” adherence to the only culture they have ever known. In this instance, the “stubbornness” is ours in not recognizing the real character of the culture which prevails in the borderlands. “Once I had a dream,” writes Dr. Ortega, “that there was in Latin America a republic in the midst of which lived an English-speaking population just as a Spanish-speaking population lives in the midst of New Mexico. If that could be made a reality, we would have the right Pan-American setup for two complementary links, and then, perhaps, the delicate problems of adjustment might be solved with a measure of equity and mutual understanding. As it is today, the Spanish-speaking population of the Southwest represents a living example of disorientation, of American political and social failure as a colonizing metropolis.  .  .  . How are we going to bid friendship to the Latin Americans, with what face are we to talk of democracy and equality to them, what are our titles, besides those of purely material power, to aspire to be the big brothers in the Pan-American empire, if we have made within our borders a mess of the relatively simple problem of dealing with an Hispanic group? How dare we, in all fairness, to call backward the Latin American republics and blame their lot on misgovernment, when we here have not managed in nearly a century to do a better job with the same human material?”9

5. The Indelible Imprint Throughout the Southwest the imprint of Spain and Mexico is indelible; not as Spanish or Mexican influence per se but as modified by contact with Indian and Anglo-American culture. The three influences are woven into nearly every aspect of the economy, the speech, the architecture, the institutions, and the customs of the people. For the people of the Southwest share a mixed cultural heritage in which the mixtures, rather than the pure strains, have survived. In a Navajo rug, an adobe house, or an irrigated farm, one may find elements of the three cultures inextricably interwoven and fused. The rug may be of Indian design, woven by Indian hands, and colored by native dyes; but the loom may be Spanish or Mexican and the wool probably came from some New Mexican’s herd or it may have been purchased from an American mail-order catalog. The rug, however, is most likely to be owned by an Anglo-American. The irrigated farm may lie within a district irrigated by water from some huge dam or reservoir built by American engineers, but the fields will be tilled by Mexicans, using a knowledge of irrigation which, in part, was acquired from Indians. “Three types of domestic architecture,” writes Ruth Laughlin, “have come down to us in their chronological order—the Pueblo, the Mexican, and the

246  North from Mexico

American-Spanish. They are seldom found absolutely true to type for the needs of men have overlapped since the days of the first Americans. In each we find resemblances to the others, like the faces of mothers and daughters.” Where these elements have been mixed, as in the domestic architecture of New Mexico, they have attained the most enduring expression. Where the Indian element has been lacking, as in the so-called “Spanish-Colonial” architecture seen in Florida and Southern California, the fusion has been least successful. Even the public buildings of the Southwest tend more and more to derive from Indian and Spanish-Mexican sources with the Anglo-Americans showing great ingenuity in adapting these forms to modern uses. In short, this mixed heritage belongs to all the people of the Southwest; not to any one group or to the combination of any two. Of paramount importance to the future of this culture is the role that the coming generation of Mexican-Americans will play. The region has yet to experience the impact of the first articulate generation of persons of Mexican descent. In another generation, Mexican-Americans will be found in all walks of life—in the arts, the professions, in the colleges and universities—and in significant numbers. In the past, Mexicans have been a more or less anonymous, voiceless, expressionless minority. There has yet to be written, for example, a novel of Southwestern experience by an American-born person of Mexican descent or a significant autobiography by a native-born Mexican. The moment the group begins to achieve this type of expression, a new chapter will be written in the history of the Southwest. For as the Spanish-speaking attain cultural maturity, as they achieve real self-expression, they will exert a profound influence on the culture of the region and Spanish-Mexican influences that have remained dormant these many years will be revived and infused with new meaning and vigor. It is the borderlands, not the border, that is important in the Southwest. For the borderlands unite the Anglo-American and the Hispano-American worlds and the area in which this mixed culture exists is expanding north and south. As the possibilities of the Good Neighbor Policy are realized, the border will have even less meaning than it has had in the past. By simply exploring the neighboring state of Sonora, Arizona businessmen have been able to increase the total traffic through the port of Nogales from $50,000,000 in 1945–1946 to $76,000,000 in 1946–1947. Incredible as it may seem, these same Arizona businessmen had for years assumed that merchandise destined for Sonora had to be shipped to Mexico City and then rerouted north to Sonora. By investigating the situation at first hand, they discovered that trucks could transport merchandise from Arizona to any point in the province, just as the pack-trains had done two hundred years ago. Today machinery, wire, pipe, cement, steel, farm implements, glass, crockery, paint, and plumbing fixtures are moving south across the border and

“One and Together”   247

Mexican minerals, shoes, fish, flax, bamboo, guano, tomatoes, chickpeas. and other products are moving north. Furthermore the imbalance between imports and exports changed from $7,000,000 in favor of the United States to somewhat less than $1,000,000 in the space of one year and even this margin may soon disappear.10 Just as Arizona is discovering that Nogales is the logical gateway to the west coast of Mexico, so Texas is discovering that its bordertowns are the logical gateways to eastern Mexico. Over how large an area, therefore, is the cultural fusion of the borderlands likely to expand in the next quarter-century? As the borderlands expand, both the Anglo and the Hispano elements will receive numerical reinforcements, so that the process of cultural fusion will be repeated; in fact, it is already apparent that this process is a constant factor in the life of the borderlands. Hence it is extremely difficult to imagine any workingout of this process that would involve the complete absorption of one culture by the other. In a sense the settlement of the United States has always moved against the grain of geography, for the east-to-west movement of the American people has been against the flow, the natural movement, of the landscape. In moving westward, the American people have crossed mountain ranges, crossed the plains, crossed the rivers, crossed the deserts. Yet the geographical flow of the continent is not from east to west but from north to south; our major mountain ranges run along north-south lines as do most of our great river systems. Unwittingly we have been bucking geography, not cooperating with it. With the lodestar being ever in the West, we have simply failed to change our vision and to note the natural contours of the country. Prior to the settlement of the eastern seaboard by European colonists, the continent was orientated on a north-south, rather than an east-west, axis and it may yet be orientated in this fashion. Since the westward movement of the American people “leapfrogged” over the intermountain states to the West Coast, we have failed to let our eyes follow the natural lines and contours which run in the opposite direction. Hence it involved an abrupt turnabout when the New Mexican frontier, with its face turned anxiously east, became a part of the last American frontier, with its face turned eagerly west. Not only is the movement “North from Mexico” older in point of time than the westward movement, but it has remained constant through the years; it is continuing now and is likely to continue indefinitely.

6. “The Sun Has Exploded” On July 16, 1945, a rancher went to visit his sheep camp in the San Andres Mountains in New Mexico. “As usual his sheepherders,” writes Ruth Laughlin, “had started out before dawn that morning in spite of a mountain rain storm.

248  North from Mexico

They had not gone far when they saw a terrific flash at the other side of the eighty-mile sweep of prairie. They ran back to the sheep camp, shaken and terrified, and cried to their patrón, The sun has exploded, señor. We saw it. It was so bright that we fell on our knees and our sheep stampeded. Take us back to our families and let us go into the church. It is the end of the world.’” When the great mushroom-like cloud of smoke and dust cleared away from the testing ground beyond Alamogordo—in this first release of atomic power in world history—the isolation of New Mexico—the isolation of all men everywhere—ended once and forever. Today New Mexico is the center of American research and experimentation in the use of atomic power and the corner where the states of New Mexico, Arizona, Colorado, and Utah meet is reported to be one of the richest centers of fissionable materials in the United States. Over the radio, as I write these lines, comes word of still another mysterious plant in New Mexico which is now employing sixty-five thousand people. The explosion at Alamogordo unlocked the latent richness of the mineral resources of the Southwest. What Emerson said many years ago has now come true: “To science there is no poison; to botany no weed; to chemistry no dirt.” The science that released atomic power in the Southwest can now find new uses for resources long regarded as worthless and can reclaim large portions of its arid wastes. Here, in the heart of the old Spanish borderlands, the oldest settled portion of the United States, a new world has been born and the isolation of the region has been forever destroyed. Like the peoples of the world, the peoples of the borderlands will either face the future “one and together” or they are likely to find themselves siftings on siftings in oblivion.

17

Chicano Leadership and Organization Matt S. Meier

World War II brought dramatic changes to America. Even more than World War I, it had a significant impact on people of Mexican descent in the United States, economically, educationally, and socially. For the most part these changes were positive. The massive war effort created earlier-undreamed-of opportunities for education, jobs, training, and economic advancement for both Mexican-American men and women. During the war years thousands of Chicanos1 moved out of the traditional homeland, the rural and small-town Southwest, to urban centers throughout the country. Higher-paying and more regular employment in war production factories and related industries became more accessible for the first time to large numbers of them. Numerous workers, especially from New Mexico, moved to Los Angeles and other West Coast cities to work in shipyards and other industries, while others, mostly from Texas, went to Chicago and Detroit to take jobs in war industries and to Kansas City and Denver to work in packing plants and beet-sugar refineries. Thousands of others, approximately one-third of a million, volunteered or were inducted into the United States Armed Services where they served with devotion and valor. Many of them fought in the more dangerous branches of the services, for example the paratroopers and the marines. From overseas they sent home to families their monthly allotments, which reduced the amount of desperate prewar poverty that existed in some areas. On the home front wives suddenly found themselves in new gender and culture roles that later helped to provide a basis for the Chicana awakening of the 1970s. The movement to the cities at times led to racism and conflict as in the Los Angeles Sleepy Lagoon affair and Zoot Suit race riots. Both military service and urbanization tended to push Mexican-Americans toward greater acculturation by distancing them from older, traditional life styles and values. The World War II

250  North from Mexico

experience strengthened their sense of citizenship, which in turn resulted in organizational development focusing on civil rights.2

1. Effects of World War II Both in the armed services and in the cities, Chicanos often found themselves in a milieu that was more tolerant and accepting than some of the societies in which they had grown up. During the war years discriminatory practices against Mexican-Americans became less overt and somewhat diminished. In the urban Midwest and West Coast and in the ranks of the armed services, many of them encountered conditions that were often less repressive, less circumscribing, allowing them not only to expand their current horizons but also to increase their hopes and expectations for a better future. The war against the fascist powers and the tremendous demand for labor that it created, plus efforts of the War Manpower Commission and the Fair Employment Practices Committee helped to reduce open discrimination against them in employment, and they found themselves no longer completely excluded from unions. At the end of the fighting some of the gains made during the war years were lost. Many wartime jobs disappeared, social advances failed to hold in some areas, and many Chicano veterans returned to find little increase, if any, in societal acceptance. On the other hand, many veterans, having had their aspirations raised and their ethnic awareness enhanced, refused to submit to indignities of the past. Determined to have the freedoms and opportunities for which they had fought and for which many had shed their blood, some were able to use the benefits of the G.I. Bill to obtain college educations or on-thejob training, to start businesses, and to buy homes; some gains were maintained. One school of historians has seen the two decades after the war (1945–1965) as a period in which Mexican-Americans devoted their energies largely to goals of personal, social, and economic improvement. This may quite well be a correct interpretation, but they were also laying the groundwork for the next period, the politics of protest.

2. Demographic Changes After the second World War occupational patterns of Chicanos responded primarily to technological changes, urbanization, and developments in internal migration. Agriculture became more mechanized with the introduction of cotton-picking machines, mechanical tomato pickers, and other “inventions.” More recently canneries and other food-processing plants increasingly automated, eliminating thousands of jobs that once were the first rung on the ladder to better-paying employment. Meanwhile, work opportunities for

Chicano Leadership and Organization   251

urban Chicanos developed or increased in construction, trucking, building maintenance, the garment industry, hotels and restaurants, retail clerking, assembly-line work, gardening, and domestic service. From earlier workers in these fields a small merchant sector emerged. Typically these barrio businessmen might be contractors, restaurant operators, small grocery or bakery owners, mechanics, metalworkers, or local publisher-printers. In some towns and cities they organized Mexican-American chambers of commerce and other service organizations. By 1988 there were more than a quarter of a million Latino-owned businesses in the United States, a majority of them in California, generating nearly $100 billion in annual revenue. Some see this development as an indication of increasing upward mobility. The heavy flow of rural peoples to the city in recent decades has been a national phenomenon, not limited to Mexican-Americans. But for them especially, the second World War was a major factor in the movement to urban centers, as workers moved to the West Coast and to inland industrial centers for employment. In the postwar years continuing heavy immigration of agricultural workers north from Mexico added to the pressure on MexicanAmericans to move to cities and to the Midwest as jobs available to them in southwestern agriculture decreased.3 More were pulled into urban barrios by a changing economy in the Southwest—from heavy dependence on agriculture, ranching, lumbering, and mining to manufacturing, government installations, and service industries in which long term, better-paying employment was available. In 1930 nearly half of employed Chicanos worked in agriculture; by 1982 less than seven percent were so employed. Already in 1950 two-thirds of Mexican-Americans lived in urban centers, and by 1980 the number had risen to more than ninety percent. However, a few rural areas remained predominantly Chicano in population, for example southeastern Texas and northwestern New Mexico. In barrios and colonias the expanded urban Mexican-American population often was led by members of the developing business and professional classes and became the basis for more aggressive community groups and politically-conscious veterans organizations. In the long run, perhaps the most important results of wartime experiences for la raza were that many Mexican-Americans were able to improve their standard of living, while some were afforded the opportunity to develop their leadership capabilities. After the war the organizing vitality among Chicanos in the Southwest was supplied primarily by veterans. They returned to barrios and colonias unwilling to accept the old strictures and aware that they could use their newfound skills to organize their communities in order to achieve their rightful place in a democratic American society. Certainly, some roots of the movimiento, the movement of the late sixties and early seventies, found fertile soil in this new consciousness of returning Chicano veterans and in their demands for full citizenship.4

252  North from Mexico

3. Organizational Development Many Mexican-American veterans of the second World War were children of migrants who had come to the United States during the years of World War I and in the 1920s. During the World War II years these second-generation Americans had the opportunity to learn how the American system functioned on a day-to-day basis, how American society could be influenced for change. Many also obtained some practical training in leadership skills and techniques. Early leaders like Héctor P. García, Edward Roybal, and Bert Corona performed the important initial function of dramatically calling attention of the majority society to the difficulties being experienced by la raza. Hard on the heels of these first postwar pioneers, the most noted leaders of the Chicano generation, César Chávez, José Angel Gutiérrez, “Corky” Gonzales, and Reies López Tijerina, brought the plight of Mexican-Americans into the mainstream of American consciousness by tying it in with the concept of civil rights and human dignity for all men. Each of these four leaders, with his own particular background, stressed certain aspects and areas of the multifaceted problems facing Mexican-Americans, particularly the widespread denial of civil and economic rights. Each formed a distinctive organization to achieve his particular objectives.

4. Community Service Organization Until World War II Mexican-American organizations had been either mutualist self-help or assimilationist-citizenship in their orientation. However, the leadership lessons that Chicanos learned during the conflict and their own experiences then and later taught them that the most powerful tool for social change was the pressure that organized groups could exert on government and society. So the postwar years witnessed the creation of various organizations that proposed to use political power and societal perceptions to achieve their goals. In addition, these groups stressed American cultural diversity and demanded greater Anglo acceptance of Mexican-Americans. One of the first of these new pressure groups was the Community Service Organization (CSO), which followed the community organizing ideas of the Chicago-based Industrial Areas Foundation, developed and headed by Saul Alinsky. Led by Fred Ross, an Alinsky disciple who had earlier helped found the Unity Leagues in southern California to elect Mexican-Americans to city councils, in 1947 the CSO was organized in Los Angeles to support the political candidacy of Edward Roybal. Roybal had been defeated in an earlier bid for a seat on the city council; now with the backing and active political support of the CSO he was elected by an overwhelming majority of 20,000 to 12,000. After the election the CSO became involved in a wide variety of civic action.

Chicano Leadership and Organization   253

While continuing its voter registration drives, it also turned to the numerous problems of the Mexican-American community, especially civil rights violations. Directing its focus to police brutality and toward the twin goals of political and social integration through naturalization and assimilation, it quickly spread from Los Angeles to other areas, especially to urban centers in California and Arizona. One of the first mass membership Chicano organizations, the CSO reached a membership of 50,000 to 100,000 in the second half of the sixties but then declined as the result of various factors: termination of financial support from the Industrial Areas Foundation, increasing competition for membership from other newly-developed Mexican-American organizations, and the more activist mood of the times. When Chicano activism declined in the late seventies, the CSO regained some of its lost membership. However, it has remained essentially a California-based group.

5. American G.I. Forum In Texas at the end of World War II, the new Mexican-American leadership was clearly demonstrated in the 1948 Longoria case which developed in the town of Three Rivers. When the mortuary there refused to handle the reburial of Félix Longoria who had been killed on the island of Luzon in the Philippines and initially interred there, Chicano veterans protested loudly, and a bitter dispute ensued. Ultimately, at the request of the fledgling U.S. Senator from Texas, Lyndon B. Johnson, Longoria was buried in Arlington National Cemetery. The following March, Mexican-American veterans met at Corpus Christi to complain about discrimination and poor service in the Veterans Administration hospital of that city. This protest led to the establishment of the American G.I. Forum (AGIF) under the leadership of Dr. Héctor Pérez García, a World War II combat surgeon who had taken an active role in the Longoria affair. Although membership in the Forum was limited to Mexican-descent veterans, the AGIF quickly developed a broad and balanced list of goals that stressed development of Chicano leaders and achievement of raza goals through wide participation in political, civic, and community affairs. Active at all levels of the political system but rejecting confrontation tactics, the Forum led voter registration drives during the fifties and filed lawsuits against civil and social discrimination, especially against segregation in Texas schools. After some membership decline in the 1960s, the seventies saw a revival and expansion of the G.I. Forum as college-educated businessmen and professionals infused the organization with new blood. The Forum was, and is, nominally nonpartisan; but its members showed themselves predominantly sympathetic to the Democratic Party, and in the 1960 presidential election they were prominent in developing Viva Kennedy clubs during the campaign. Although initially a Texas organization, the AGIF has expanded all over the Southwest.5

254  North from Mexico

6. California MAPA Voter registration drives and other activities of the American G.I. Forum, the Community Service Organization, and other Mexican-American groups increased the strength of the Democratic Party, but to many politically-minded Chicanos the party seemed to take their support for granted and appeared unwilling to include them in party structure, decisions, and benefits. As a result of this feeling a group of Chicano activists under the leadership of Bert Corona and Eduardo Quevedo met in 1959 at Fresno, California, to consider what action might be taken to advance Mexican-American political interests more rapidly. After prolonged discussion the assembled California leaders formed the Mexican-American Political Association (MAPA), an organization with an overwhelming emphasis on political action, thereby initiating a new stage in Chicano political activism. Edward Roybal was elected its first president. MAPA had little direct interest in social assimilation; its main emphasis was political, ethnic, and nonpartisan. It sought Mexican-American candidates in both major parties and supported their candidacies, it took public stands on issues affecting la raza, it lobbied with public officials for Mexican-American interests, and it carried out political education programs in the community. Initially the Mexican-American Political Association aroused enthusiastic and widespread interest among Chicanos throughout the Southwest. Early in 1960 a group of Mexican-American political leaders in Texas formed an organization they called Mexican-Americans for Political Action (sometimes referred to as Texas MAPA from its initials). However, in the presidential election year it was overshadowed by the Viva Kennedy clubs, which captured the imagination and support of most Mexican-Americans. Although California MAPA fell short of its early goal of forming a chapter in each of the eighty state assembly districts, it remained a powerful force in California politics during the early 1960s and had some success at influencing both major parties and in electing raza officials. With MAPA support, in 1962 Edward Roybal won a seat in the U.S. House of Representatives, a position to which he has regularly been reelected since. After some decline in MAPA’s power in the early seventies, it underwent a revival during the latter years of that decade, only to languish again in the 1980s due to the greater conservatism of the times, internal factionalism among the leaders, and financial problems. Despite efforts to extend MAPA nationally it has remained essentially a California organization.6

7. A National Organization? Because of the Mexican-American Political Association’s early successes, there was widespread interest in the possibility of developing a national

Chicano Leadership and Organization   255

organization, possibly along the lines of MAPA. To this end representatives of California MAPA, Texas MAPA, the CSO, LULAC, Viva Kennedy clubs, and other groups met in Phoenix, Arizona, after the presidential election of 1960 to try to put together an organization with national membership and concerns. The elections had shown the importance of the Mexican-American vote and indicated that political control of south Texas by conservative Anglo Democrats could be broken. Under the leadership of Dr. Héctor P. García and others the Political Association of Spanish-Speaking Organizations (PASSO/PASO) was formed after some, at times acrimonious, discussion. Dr. García was elected its first president. However, California MAPA and the CSO withdrew their support from the group, with the result that PASSO’s membership was limited largely to Chicanos active in the Texas Viva Kennedy clubs. It never became the national organization that had been envisioned. PASSO, taking the place of Texas MAPA, has been active mainly in that state and concentrates its efforts on civic education and direct political action rather than social or economic integration. Its most notable success was at Crystal City, Texas, where, in conjunction with the Bishops Committee for the Spanish Speaking, the Teamsters Union, and other groups, it succeeded in securing the election of an all-MexicanAmerican city council in 1963. However, the Crystal City campaign was also divisive and led to a split in PASSO with most of Hector García’s more moderate followers later withdrawing from the organization, which then turned somewhat more militant. Unsuccessful efforts by PASSO during the early 1960s to establish chapters in Arizona led instead to the founding at Phoenix of the American Coordinating Council of Political Education (ACCPE), which was modeled on PASSO. Tailored more specifically to the political needs of Arizona Chicanos, ACCPE had some successes at local levels, electing Mexican-American candidates to city councils and school boards. Although it spread to most of the state it was unable to achieve success at county or state levels and remained a rather pallid reflection of PASSO.

8. Leadership in the Protest Movement By the middle sixties the political impact of these organizations, while still limited, was clearly evident, and Mexican-American voters in the Southwest began to find themselves wooed regularly by candidates seeking public office. In addition to being made aware of their newly-acquired political potential, they were exposed at the same time to the heady influence of the black civil rights movement and to the ideology of the Black Power movement of that decade, as well as to the militancy of anti-Vietnam activists. All of this led to a further step in the Mexican-American awakening. Illustrative of and central to

256  North from Mexico

this change were four prominent new Chicano leaders who explicated postwar goals for la raza in the second half of the sixties: César Chávez, Rodolfo “Corky” Gonzales, Reies López Tijerina, and José Angel Gutiérrez. Each created an organization to achieve the goals he had defined for the Mexican-American community. In the long run the most successful of these leaders was César Chávez.7

9. Man of the Migrants: César Chávez After a childhood spent on the migrant harvest circuit, a two-year hitch in the United States Navy during the latter part of World War II, and a decade as a CSO organizer, César Chávez Estrada in 1962 began developing the National Farm Workers Association (NFWA), which was ultimately to become the United Farm Workers (UFW). In 1965, faced with low harvest wages in California vineyards and unsanitary living and working conditions, the NFWA joined with about 600 Filipino members of the Agricultural Workers Organizing Committee of the American Federation of Labor led by Larry Itliong in a strike against table grape growers of California’s San Joaquin Valley. The charismatic Chávez, who quickly came to dominate the strike and molded its nature, envisioned the Delano Grape Strike, or “la Huelga” as it became widely known, as no simple labor dispute but as a crusade for social justice, for human dignity—la causa–based on nonviolence and civil rights. With help from Dolores Huerta, Fred Ross Sr., Jim Drake, and many others, César Chávez was able by his charisma and hard work to arouse and organize the aggrieved grape workers. After a 1966 march from Delano to Sacramento to dramatize the strike, a twenty-five-day fast by Chávez in the following year to renew the strikers’ dedication to nonviolence, and a nationwide 1968–1970 table grape boycott, the union finally won “la Huelga” in July 1970 and signed a series of favorable three-year contracts with a majority of the table grape growers. The successful conclusion of the long strike was a testimonial to César Chávez’s ability to organize and lead, as he combined a doctrine of nonviolence with the support of students, labor unions, civil rights groups, the Catholic Bishops’ Committee for the Spanish Speaking, the California Migrant Ministry, and the symbol of the Virgin of Guadalupe. It was also a tribute to the support of farmworkers who, as a group, tended to be the most traditional, the most “Mexican” culturally, and certainly the most exploited and ill-treated of Mexican-Americans.8 When contract renewal time came around in three years, the UFW lost most of its contracts, in many cases to its archrival and favorite of the growers, the Teamsters Union. This reverse was offset only temporarily by the 1975 passage of a state Agricultural Labor Relations Act that Chávez and the union strongly supported and that enabled the UFW to win back many of its

Chicano Leadership and Organization   257

contracts in field elections. However, the benefits of the act for the UFW were soon ended as a result of its being gutted by purse-string control in the growerdominated California legislature. César Chávez’s initial success in the grape strike led to the next target for the UFW, the lettuce industry in California’s Salinas Valley. Here the union found the Teamsters deeply entrenched and working closely with the large lettuce growers. In spite of lawsuits, picketing, boycotts, and repeated efforts to come to an agreement with the Teamsters union about jurisdictional divisions, Chávez was never able to repeat in Salinas his earlier success in the central valley. There he had won a remarkable victory in the vineyards, but its value was largely symbolic. During the 1970s, despite setbacks in California, UFW ideas and influence spread to the rest of the Southwest and even to Florida agriculture, although its success in the fields was limited. Some internal dissent in UFW ranks, especially over Chávez’s use of the Synanon Game, a harsh personal-encounter technique, and resultant defections from the union in the latter seventies further weakened the organization. However, during the 1980’s César Chávez, using direct mailing and computers, hoped to turn around the national trend against unionism and especially the decline in UFW contracts. In 1984 he initiated a third grape boycott to protest the excessive use of dangerous pesticides by growers9 and to call attention to the emasculation of California’s Agricultural Labor Relations Act by agricultural interests. In August 1988 Chávez completed a thirty-six-day fast to gain support for the four-year-old UFW grape boycott.10 Although the UFW does not now have any table grape contracts in California, it does have a membership of about 40,000 and over 100 contracts in vegetables, citrus, and other areas of agriculture. An important factor in the union’s decline has been the continuing almost unlimited supply of undocumented workers. In spite of serious setbacks, over the years Chávez and the UFW have had a considerable impact on agricultural wages and working conditions, and it seems unlikely that workers will ever be returned to their earlier, inhumane conditions of work and pay. As a result of his role in this revolution in harvest agriculture César Chávez became the first Chicano leader to achieve nationwide recognition and also became a national symbol of social justice for both Chicanos and Anglos. As this symbol he has been a galvanizing force in awakening the consciousness of the American people to the many injustices suffered by MexicanAmericans. In addition, Chávez and the UFW have had indirect impact in other areas of the Chicano experience, for example, politics.

10. Champion and Crusader: Rodolfo “Corky” Gonzales The year after Chávez initiated the Delano Grape Strike, Rodolfo “Corky” Gonzales, ex-Golden Gloves boxing champion and former Democratic

258  North from Mexico

political leader in Denver, Colorado, began to establish his Crusade for Justice (Cruzada Para la Justicia). He based the Crusade on a group called Los Voluntarios, which he had founded earlier. He envisioned it as a civil rights organization to promote Chicano self-determination and nationalism and to fight against police brutality and judicial bigotry toward Chicano youths. In October 1967 he was one of the Mexican-American leaders who boycotted the El Paso White House Conference on the Mexican-American. During the following summer he participated in the Poor People’s March on Washington, D.C., playing an important but secondary role to Reies López Tijerina and the black leader, Ralph Abernathy. Meanwhile he also enlisted widespread support among young people by giving talks at various colleges and universities. His immensely popular poem “I Am Joaquin” and his “Plan of the Barrio” (1968) added to his following. Corky’s Crusade for Justice provided much of the leadership and direction to the politics of protest during these years. In March 1969 his first Chicano Youth Liberation Conference, held in Denver, produced the “Plan Espiritual de Aztlán” which set goals of cultural nationalism and defined a concept of the mythical Aztec land of Aztlán, identified with the Southwest, as the Chicano homeland. At the second annual conference the following March, Gonzales called for the organizing of a Chicano political party as well as the establishment of Aztlán. A few weeks later he announced the forming of the Colorado La Raza Unida party (LRU), which was able to secure a place on the ballot and ran several candidates for local and state offices later that year. At the September 1972 El Paso convention of La Raza Unida, delegates engaged in a vigorous debate over the party platform, political tactics, and national leadership. Supporters of Gonzales and followers of José Angel Gutiérrez, a founder and the leader of the Texas LRU, vied for control of the convention. Paradoxically the latter won the struggle and emerged as the party leader, but the convention supported much of Gonzales’s somewhat socialistic program of ethnic nationalism. The year 1972 marked the beginning of the decline of the Crusade for Justice, as a majority of la raza in the Southwest, sharing in a national conservative trend, began to move away from Corky’s strongly activist and nationalistic stance. However, he continued to maintain the Crusade’s Denver complex with its languishing learning center, Escuela Tlatelolco, and its medical, legal, and financial services. As its head he continued to defend the civil rights of Chicanos. Increasingly he became more and more isolated from the Mexican-American mainstream and therefore lost power at the national level. During the late seventies he made a return to boxing, training amateurs at first and then professionals in the early eighties. In October 1987 he was seriously injured in a car accident from which he made a very slow recovery because of complications.11

Chicano Leadership and Organization   259

11. King Tiger: Reies López Tijerina During the late 1960s the activities of Corky Gonzales and César Chávez were often overshadowed in the United States press by the more flamboyantly militant and confrontational tactics of Reies López Tijerina and his followers in the Alianza Federal de Mercedes (Federal Alliance of Land Grants) in New Mexico. Like Chávez and Gonzales, Tijerina had grown up on the migrant harvest circuit and as a result had limited formal schooling. At seventeen he became a convert to a fundamentalist Protestant sect and, having undergone a period of training in a small Texas Bible institute, engaged in migrant missionary work along the Mexican border. After a decade spent in this activity, Tijerina broke with the Pentecostals over matters of church discipline, and by the mid-1950s social justice, rather than any specific creed, had become his primary concern. Against persistent and vicious local Anglo opposition he spent two years in Arizona trying to establish a utopian cooperative community, Valle de la Paz. In 1960 he moved permanently to New Mexico where he had already begun to study the land grant question and had gradually become convinced that all the problems of Nuevo Mexicanos derived from the loss of their lands—particularly the pueblo (town) grants. He set the regaining of these grants as his ultimate goal. In 1963 he began to recruit followers in an organization he called Alianza Federal de Mercedes, and after May 1967, the Alianza Federal de Pueblos Libres (Federal Alliance of Free Towns). At first Tijerina used the legal system and the courts in his attempts to persuade government officials of the validity and justice of Nuevo Mexicanos’ land claims. By the time of the Delano Grape Strike, the Alianza claimed a membership of 20,000 and in the following year initiated direct action designed to draw public attention to its demands. In July 1966 the failure of a march from Albuquerque to the state capital to secure official response to demands was followed by more confrontational tactics such as an October attempt to take over Echo Amphitheater in Kit Carson National Forest. In June 1967 Alianza members “raided” the Rio Arriba county courthouse at Tierra Amarilla in order to make a citizen’s arrest on District Attorney Alfonso Sanchez. By these actions Tijerina hoped to further his objective of calling attention to injustices suffered by Nuevo Mexicanos. Arrested and charged with numerous offenses in connection with both events, Tijerina argued his own defense, won some “not guilty” verdicts, but ultimately was sentenced to prison in 1970. While fighting in the courts, he continued his aggressive tactics, playing a leading role in the 1968 Poor People’s March to Washington, D.C., where he demanded a meeting with Secretary of State Dean Rusk to discuss the 1848 Treaty of Guadalupe Hidalgo. He also

260  North from Mexico

announced his intention to run for governor of New Mexico bur was disqualified because of his felony conviction. In the following year he attempted a citizen’s arrest of Supreme Court nominee Warren Burger, New Mexico governor David Cargo, and other prominent figures. On July 27, 1971, after serving more than two years in federal prison, Reies Tijerina was paroled with the condition that he hold no office in the Alianza during the next five years. After his release he turned to the theme of “Brotherhood Awareness.” At the El Paso Raza Unida convention in the following year he rook the role of elder statesman while Corky Gonzales and José Angel Gutiérrez struggled for leadership. A month later, October 1972, he lost many of his youthful followers when the Tierra y Cultura congress, which he had called, voted to support the LRU in the November elections over his objections. He walked out of the conference in high dudgeon and returned to his northern New Mexico interests. When his five-year parole period ended in 1976 he returned to the presidency of the Alianza, which had, in the meantime, devoted its main activities to issuing public statements of policy, publishing pamphlets, and filing law suits to regain land grants. During the second half of the seventies and the early eighties, Tijerina, now leader of a greatly weakened organization, directed much of his energies toward attempts to interest the successive presidents of Mexico in supporting the Alianza in the land grant issue on the basis of the Treaty of Guadalupe Hidalgo—without success. Logically Mexican policy has been to take no public position on problems facing Mexican-Americans except to be broadly critical of denial of civil rights. Partly as a result of Tijerina’s more restrained style he attracted less national attention, and his support was ultimately reduced to the older land grant people of the Rio Arriba region of northwestern New Mexico. A twentieth anniversary celebration of the Tierra Amarilla incident, held in June 1987, drew fewer than 200, mostly older Nuevo Mexicanos, who came to listen to Reies López Tijerina. In his time on the historical stage this complex man played an important role by dramatizing the land grant question not only for New Mexico but for the entire Southwest, by focusing attention on rural poverty in northern New Mexico, and by reawakening Nuevo Mexicano pride in language, culture, and ethnic identity.12

12. A Student Leader: José Angel Gutiérrez While Tijerina was involved in the dramatic and violent confrontation at Tierra Amarilla, in southern Texas young José Angel Gutiérrez, a graduate student in political science at St. Mary’s University in San Antonio, was organizing his fellow students in a group called the Mexican-American Youth Organization (MAYO). MAYO’s principal objective was to work for social

Chicano Leadership and Organization   261

change in a state dominated by conservative rural Anglo interests. Gutiérrez was elected its first president. Son of an immigrant medical doctor who had settled in Crystal City, Texas, he returned to Cristal, as many Chicanos referred to it, where he began implementing some of his ideas about organizing to achieve political power within the system. Following raza difficulties with the Anglo school board and a resulting high school walkout late in 1969, he organized local Mexican-Americans into the Raza Unida party. In the April 1970 elections Gutiérrez, along with two other Chicanos, won seats on the school board and two more Mexican-Americans were elected to the city council. Raza Unida candidates also won city council and school board elections in several nearby Texas towns where, for lack of organization, their majority position had previously been politically unavailing. This local success, a first for Mexican-Americans in Texas, led to pressure on La Raza Unida to go national in the presidential election year 1972, and a national convention was held in El Paso. At the convention Gutiérrez’s pragmatic political position, which emphasized working within the two-party system, won out over Corky Gonzales’s insistence on the creation of a nationwide third party and offering voters a national LRU ticket. In ensuing elections local LRU candidates had some success, but clearly the party’s power had already peaked. However, La Raza Unida did show enough power that the two major parties were induced subsequently to nominate Mexican-American candidates in areas of large Chicano population, but the party was never able to develop even a regional permanent presence. During the national trend toward greater conservatism in the late seventies and eighties the LRU became fragmented, moribund, and finally expired because of internal conflict and lack of voter support. Attacks by conservative state leaders in Texas also aimed at destroying the LRU and contributed to its demise. José Angel Gutiérrez was elected judge of Zavala County, Texas. in 1974, but ongoing difficulties with the Anglo-dominated legal establishment caused him to resign in 1981, and he turned to college teaching, first at Colegio César Chávez in Oregon and later at Western Oregon State College.13

13. Summary The activities of these four leaders were, and are, an important part of the Mexican-American’s long struggle for full citizenship rights. They appealed to the varied segments of Mexican-American society; each, with his own particular background, emphasized certain aspects of the plight of la raza; all stressed the denial of civil rights, all worked for social justice for Mexican-Americans. César Chávez, leading a civil-rights–labor organization, developed the widest appeal, both within the community and among Anglos. Even though he concerned himself with a minority within a minority, Mexican-American

262  North from Mexico

agricultural workers, he broadened his appeal by formulating his concern in terms of civil rights and human dignity. José Angel Gutiérrez, as leader of a civil-rights–political group, held a special appeal for many second- and thirdlevel Chicano leaders working within the United States political structure; on the other hand, his often strident rhetoric attracted many youthful activists from various student groups. Reies López Tijerina, heading a nationalist, irredentist alliance, attracted a mixed following of conservative older Nuevo Mexicanos, most of them land grant descendants, and of activist Chicano youths, mostly students. Corky Gonzales’s emphasis on cultural nationalism and political activism brought him a large following of high school and college-age Chicanos, attracted by his idealism and rhetorical vigor. All four leaders made important contributions to bringing the many socioeconomic injustices facing la raza to the attention of United States society. All made significant contributions to the advancement of the movimiento, as the sometimes aggressive outburst of the struggle for civil and political rights as well as for cultural self-assertion and liberation became known. It goes without saying that their achievements were aided significantly by the commitment and hard work of many devoted second- and third-level leaders and support from their many loyal followers.

18

Politics, Education, and Culture Matt S. Meier

By the late seventies and early eighties the four principal Chicano leaders: Chávez, Gonzales, Tijerina, and Gutiérrez, had all declined in appeal and power, some more than others. Their weaker positions resulted in part from changes taking place in the Mexican-American community. Already by the mid-1970s the civil rights movement, the black movement, and the student movement had all become less militant or had all but disappeared from the American scene. Radical stances and confrontational styles diminished in effectiveness, and those who continued in that mode became less and less influential. A majority of Mexican-Americans rejected extreme militancy but still felt their civil rights should provide them a greater say in their own governance. In 1974 two politically moderate Mexican-Americans were elected state governors: Raúl Castro in Arizona and Jerry Apodaca in New Mexico. Subsequently the Chicano leadership mantle seemed to be falling upon the postwar generation in the persons of people like Toney Anaya, governor of New Mexico from 1982 to 1986, Mayor Henry Cisneros of San Antonio, and Mayor Federico Peña of Denver, the latter two following in the footsteps of Mayor Raymond Telles of El Paso (1957–1961) and working within the political system. A kind of milestone was marked in 1984 when Raúl González became the first Tejano elected by popular vote to a statewide office: Associate Justice of the Texas Supreme Court.

1. Chicano Powerlessness Although the Mexican-American population has increased from about two million in 1930 to nearly six million in 1970 and to almost nine million in 1980, its members have obtained little decision-making power. Between 1976 and 1980 nearly 665,000 Chicanos were registered in the Southwest, and by

264  North from Mexico

1984 another 443,689 Latino1 voters, mostly Mexican-Americans, were added to the rolls. However, in the nonpresidential election of 1982 only 52 percent of eligible Latinos were registered and only 37 percent voted. In California the number of voting-age Latinos increased a dramatic 117 percent in the decade 1970 to 1980, and by mid-decade they formed twenty-one percent of the state’s population, yet they wield very limited political power. Why do Chicanos have a lower level of participation in the political process than the general population? The answer lies in the many factors that sap their political potential. A few are cultural, but most arise out of their socioeconomic position in American life and out of discrimination practiced against them. To begin with, by far the most important factors in their political emasculation are gerrymandering their areas of population concentration, the device of at-large rather than single-district elections, other discriminatory election practices such as threats, as in Orange County, California, in the 1988 election, and, until fairly recently, poll taxes. Residency requirements have also militated against those who remain in the migrant stream. A relatively high level of non-citizenship further lessens the number of political participants; about thirty percent of Mexican-Americans are not citizens, compared to three percent for the total population. The fact that they form the most recent immigrant wave explains part of that difference in citizenship. Additional factors encouraging naturalization seem to be property ownership, marriage, and children, especially when born in the United States.2 In addition, there is a cluster of factors that further reduce MexicanAmerican political participation. It is a matter of statistical record that low levels of age, education, and income negatively affect voter registration and turnout. In all three categories Chicanos are at, or close to, the bottom in American society. In 1980 the median age of Chicanos was 22.1 years; for the whole population it was thirty-one years. Only about fifty percent of Chicanos complete high school, and median family income is only two-thirds of the national average. Another factor is the denial of the vote to convicted felons. For a number of reasons Chicanos are in that category in higher percentages than the general population. A final factor might be called cultural; it is a cynicism about politicians and the political process. This may come from their historical experience both in Mexico and in the United States.3

2. Government Response At the national level government reaction to increased Chicano political power has led to some improvement. In response to Mexican-American demands, President Lyndon Johnson in June 1967 announced the formation of the Inter-Agency Committee on Mexican-American Affairs and appointed Vicente Ximenes to head the new office. Two years later it became a Cabinet

Politics, Education, and Culture   265

Committee on Opportunities for Spanish Speaking People with a five year life. In its early days the Nixon administration announced a sixteen-point program to hire the Spanish-speaking in government. However, during his five-andone-half years of tenure the President appointed fewer than ten Latinos to high-level positions, while his successor, Gerald Ford, made over twenty such appointments in two and one-half years. Between 1977 and 1981 President Jimmy Carter appointed about 200 Latino officials. Under his successor, Ronald Reagan, the total was down, but he did appoint the first MexicanAmerican to a Cabinet post. Lauro Cavazos, president of Texas Technological University, became Secretary of Education in September 1988 and in the next year was reappointed to that position by the new President, George Bush. President Bush also selected Nuevo Mexicano Manuel Luján as his Secretary of the Interior; he was approved by the Senate early in February 1989. By 1986 there were 183 Latino mayors, 117 state legislators, and over 1,000 members of city councils and other elected city officials, the majority of these MexicanAmericans, according to the National Association of Latino Elected and Appointed Officials (NALEO).4 At the end of the following year there were an estimated 3,000 elected Hispanic officials. While Mexican-Americans often seemed less than firmly committed to Latino unity, which might give them greater political power, they clearly had moved beyond brown-power rhetoric and sloganeering to a position of working for substantial political and social benefits.

3. The Chicano Student Movement Over the years Mexican-American educational advancement, one of the routes to greater political power, has been held back by various factors: continuing heavy immigration from Mexico, demographic concentration in the Southwest and in urban barrios, low income, occupational and geographic transiency, language and cultural differences, and inferior schools and teachers. At the end of World War II rekindled demands for equality in America led in 1945 to the filing by LULAC of Mendez v. Westminster School District against segregation in Southern California schools. Two years later the Ninth Circuit Court upheld a lower court decision outlawing segregation, and the school district was forced to comply. In Texas a similar LULAC suit, Delgado v. Bastrop Independent School District led to a court decision in 1948 that segregation was prohibited by the equal protection clause of the Fourteenth Amendment. Some Texas schools responded to this outlawing of segregation by integrating black and Chicano schools, rather than Chicano and Anglo schools. Despite elimination of many exclusionary practices since World War II, Mexican-Americans have lacked sufficient political strength to win definitely their war for educational equality.

266  North from Mexico

Reaching maturity in the late sixties and early seventies, post–World War II Chicano youths were in the forefront of the student movements of that period. As more Chicano students began to enter the upper levels of the educational mainstream, a heightened stress on equality in education led increasingly to concern about secondary school and college curricula. On university, college, and high school campuses they developed student organizations that formed such a vital part of the movimiento. Among the most important of the groups they formed to address their concerns were the Mexican-American Youth Organization (MAYO) in Texas, United Mexican American Students (UMAS) and the Mexican-American Student Association (MASA) in southern California, the Mexican-American Student Confederation (MASC) in north central California, and the Chicano Associated Students Organization and Chicano Youth Association (CYA) in New Mexico. All of these groups came into being during the mid-sixties, sparked in part by the civil rights movement as well as concern for quality education. Thousands of Chicano students, by challenging the system and sometimes their more conservative and hesitant barrio elders, acted as a moral goad to school administrators and community leaders and thereby rendered a valuable service to the movement. The student groups and their demands for better educational treatment and opportunity were generally supported by Mexican-American communities. With the example set by the four leaders in the movimiento, young urban Chicanos, who generally represented the more acculturated segment of la raza and many of whom had previously been identified with the ideology of LULAC and the G.I. Forum, began organizing their fellow students in high schools and colleges during the second half of the 1960s. The MexicanAmerican Student Association was founded in 1966 at East Los Angeles Community College, and United Mexican American Students, while having earlier antecedents, seems to have been organized directly as a result of a meeting held the following May at Loyola University in Los Angeles. UMAS was service-oriented, maintaining closer ties to the Mexican-American community than most student groups. In the San Francisco Bay area the MexicanAmerican Student Confederation, founded at this same time, stressed college educational programs and the need to bring more Chicanos into the college and university system, both as students and teachers. It tended to be more politically oriented and also more radical in its leadership than UMAS in the south. In New Mexico the Chicano Associated Students Organization, at first named the Spanish American Students Organization, succeeded in obtaining innovative multicultural courses at Highlands University in Las Vegas and capped student efforts by getting the first Mexican-American university president in the United States: Dr. Frank Angel. The Chicano Youth Association in the state was primarily a high school organization and was mostly concerned

Politics, Education, and Culture   267

with problems of discrimination in the schools, the limited number of Chicano teachers and counselors, and the fact that schools ignored the pressing educational needs of Chicanos at that level. In Texas MAYO was officially organized in 1967 at St. Mary’s College (University) by José Angel Gutiérrez and other student leaders, although it had an informal existence since 1964 at Texas A & M College (University) in Kingsville. Widely active in community concerns, its principal goals were aggressively political. It quickly spread to other college and high school campuses in Texas. Early in 1969 it protested injustice and discrimination, issued a Del Rio Manifesto condemning racism and asserting Mexican-American cultural identity, and demanded reinstatement of a canceled community Vista program. In California the rapid spread of student organizations led to a conference in 1969 at the University of California at Santa Barbara to unite the various university and college groups under the name Movimiento Estudiantil Chicano de Aztlán (MEChA). The conference, made up of students, faculty, and community activists, also drew up the Plan de Santa Barbara, a comprehensive approach to the development of Chicano college programs. The plan included sample degree programs from several California universities and colleges, descriptions of courses, and a select bibliography, as well as an exposition of the ideology of the movement.5 In Mexican-American high schools, which tended to be overcrowded and poorly equipped in comparison to Anglo schools, student groups were primarily concerned with separate but unequal education. Guided in some instances by adults like Corky Gonzales in Denver and Sal Castro in Los Angeles, the often militant leaders of these organizations led walkouts, sometimes called “blowouts,” at the secondary school level all over the Southwest. Their demands included a wide range of clearly articulated academic goals which included more Chicano teachers with whom students could identify and use as role models, social science courses more relevant to Chicanos, and programs to encourage Chicano students to complete their high school education. In March 1968 nearly 10,000 students walked out of five Los Angeles high schools. As a result of this example, similar demands and walkouts took place at high schools in Denver, San Antonio, Albuquerque, Phoenix, and other cities in the Southwest and even some in the Midwest. Protesting students were in many cases treated by police with excessive roughness and sometimes with brutality.6

4. Student Movement Results Directly as a result of these confrontations, the students obtained staff, policy, and curricular reforms in most schools. Their protests also caused other

268  North from Mexico

schools, not directly faced with walkouts, to accept many of the demands, including creation of ethnic studies courses, recruitment of minority teachers, and programs to educate school staffs about the educational needs and problems of Chicano students. At the college level, as the result of mass demonstrations, sit-ins, and other direct confrontation tactics, departments of Chicano or ethnic studies were created, and courses in Chicano history and sociology were added to the curriculum. In a few institutions degree programs in Chicano Studies were developed, and for the first time serious efforts were made to recruit minority faculty and students. One of the major weaknesses of the Chicano movement in education was the lack of an institutional academic base, such as blacks have had since the Civil War, for example, Howard University in Washington, D.C. During the 1970s efforts were made to develop a Chicano university base, but they were less successful than other educational efforts. In Austin, Texas, the Juárez-Lincoln University was organized, and in northwestern Oregon Mt. Angel College was converted into Colegio César Chávez. In California at the beginning of the seventies a former army base near Davis became Deganawidah-Quetzalcoatl University (DQU), shared by Chicanos and Native Americans. A decade later and fifty miles to the southwest the National Hispanic University was founded in Oakland. Among the many problems faced by these institutions were difficulties in securing accreditation and the external financial support that often depends on being accredited. Although the walkouts and sit-ins drew national attention to the educational plight of Chicanos, in the long run improvement was limited. A recent study of high school students showed that the aspirations of Anglo, Chicano, and black students were very similar, but that expectations of Chicano students, conditioned by American society, were much lower. As a result, MexicanAmerican students still have an unacceptably high dropout rate. Nevertheless, the student movement had extraordinary importance, especially among middle-class urbanized Mexican-Americans. It provided invaluable leadership and support for off-campus as well as on-campus groups.

5. Other Chicano Organizations Important to the entire movement, and especially to the high school walkouts, was a community organization called the Brown Berets. Originally known in 1967 in East Los Angeles as the Young Citizens for Community Action, made up mostly of high school youths, over the months it moved to a more active stance in confrontation with the authorities. An apolitical, paramilitary group active principally in California and Texas, the Berets took upon themselves the function of defending the community against police harassment and other

Politics, Education, and Culture   269

forms of the majority society’s repression. In 1969 in response to the Vietnam War they formed the National Chicano Moratorium Committee, and on the following August 29 a national moratorium march was held in Los Angeles. In the course of police monitoring of the event, which attracted between 20,000 to 30,000 participants, Rubén Salazar, a prominent Los Angeles news and television reporter, was killed by a direct hit from a high-velocity tear-gas projectile. The subsequent coroner’s inquest failed to indict the officer responsible for Salazar’s death and he was never tried. Most Chicanos and many Anglos felt that the inquest was seriously flawed.7 The Brown Berets exhibited varying degrees of suspicion toward Chicano professionals and university students and appealed strongly to less-educated barrio youths. By the beginning of the seventies the group had forged a program of concerns about barrio issues, especially unemployment, housing, and civil rights. However, internal factionalism and weakness in organizational discipline which led to negative publicity, plus intense police intimidation, caused the group to lose some barrio support. A part of its leadership became more radicalized and split off to form a rival group, La Junta. In 1972 David Sánchez, founder of the Brown Berets, announced the disbanding of its ninety chapters to avoid factional violence. The organization had a valuable function in arousing and politicizing many moderate members of the community. It also led to some spin-offs, imitators, and rivals in the community.8 Among other important community groups were the Community Action Patrol of San Jose, California, and Latinos Unidos por la Justicia, a reformist and somewhat paramilitary organization in the greater Los Angeles area. Also in Southern California the Congress of Mexican American Unity, founded in 1968, served as a conduit for the concerns of about 200 local community groups. In Texas the Mexican-American Unity Council and the Community Organized for Public Service (COPS) worked successfully for the economic and political interests of the raza community. Among the institutions that felt the rising pressure of Chicano militancy was the Catholic Church. Until the 1945 organizing of the Bishops Committee for the Spanish-Speaking under the leadership of San Antonio’s archbishop Robert E. Lucy, the Church hierarchy had generally kept clear of Mexican-American social problems. During the Delano Grape Strike some clerics and nuns participated in UFW activities, while others feared to alienate grower parishioners. However, in 1970 the Bishops Committee on Farm Labor played a paramount role in bringing the strike to a successful conclusion. At this same time in the Los Angeles area an activist Chicano group, Católicos por la Raza, assailed the hierarchy for ignoring Mexican-American Catholics and demanded reforms. Undoubtedly because of the movimiento, during the 1970s the Vatican appointed about ten Mexican-American bishops, and greater efforts were made to

270  North from Mexico

recruit Latino priests. A group of Chicano priests, in the meantime, had begun their own organization named PADRES (Padres Asociados Para Derechos Religiosos, Educativos, y Sociales). PADRES, with about 300 members, worked for greater Church concern for Mexican-Americans and sought to arouse in their parishioners more awareness of their own power to bring about social and economic change. By 1989 there were two dozen Latino bishops who could articulate Chicano concerns in hierarchical circles.9

6. National Organizations Despite the failure of activists to develop a nationwide political umbrella organization, there were some successes at the national level. In 1964 the National Organization of Mexican-American Services (NOMAS) was instituted to coordinate the efforts of Chicano groups nationwide. It was useful in that function for a few years until undercut by the Lyndon B. Johnson administration, which bypassed it in dealing with Chicanos. However, in 1968 a more successful and enduring agency was created. With the help of a $2.2 million grant from the Ford Foundation, the Mexican-American Legal Defense and Education Fund (MALDEF) was established to protect the legal rights of Chicanos and to help raza youth gain entrance into the legal profession. Five years after its founding, the financial basis and the areas of concern for MALDEF were broadened by the presidency of Vilma Martínez, a dynamic Chicana lawyer and civil rights leader. MALDEF has concerned itself primarily with educational segregation, job discrimination, jury exclusion, and other civil rights and consumer issues. Since the passage of the Simpson-Rodino Immigration Reform and Control Act of 1986 (IR&CA), investigating and litigating complaints of discrimination resulting from the law has become one of its priorities. It has headquarters in San Francisco and offices in Los Angeles, Denver, Albuquerque, San Antonio, and Washington, D.C. A somewhat similar organization was the Southwest Council of La Raza, a coalition of some two dozen groups, also begun with a Ford Foundation grant in 1968. When the grant money ran out in 1970 the Council moved to Washington, D.C., and three years later was metamorphosed into the National Council of La Raza as it widened the scope of its activities. During the seventies the Council developed ties with more than 100 community organizations as it sought to strengthen community-based groups by supplying them with advice, knowhow, and technical assistance in program development, fund-raising, and dayto-day operation. Similar to the Council and MALDEF in its concerns is the Southwest Voter Registration Education Project (SVREP), founded in the mid-1970s, which works with community groups to register MexicanAmerican voters. It not only has filed suits challenging denial of civil rights to Chicano voters; but with the help of Voting Rights Act amendments in 1975

Politics, Education, and Culture   271

and 1982, it has more than doubled the number of registered MexicanAmerican voters.

7. Chicanas The many efforts to organize Chicano interest and pressure groups that made up the movimiento had one aspect that was a change in degree rather than a completely new facet. This was the increasing participation and leadership of Chicanas, especially in student and community organizations. Long before the 1960s Mexican-American women were politically active in auxiliaries to male organizations and in a handful of more militant or radical groups. A few were able to achieve leadership roles. As far back as the latter part of the nineteenth century, Lucia González Parsons had been active in radical labor movements and was one of the founders of the Industrial Workers of the World (IWW) in 1905. A decade later the Liga Mexicanista Femenil and the radical Club Liberal de San Antonio were founded by women and in the 1920s María L. Hernández and her husband Pedro founded La Orden Caballeros de América. This San Antonio offshoot from La Orden Hijos de América had an important part in establishing LULAC at the end of the decade. María Hernández herself remained active in politics, seeking civil rights and justice for MexicanAmericans and even taking part in the Chicano movimiento during the late sixties despite her age. In the 1930s, when Chicanas made up a sizable segment of workers in the border garment industry, pecan shelling, and other sweatshop industries, a new generation of Chicana leadership appeared. Isabel Malagrán, educated at New Mexico A & M College (now New Mexico State University), led a harvest strike by pea-pickers in the mid-thirties; after moving to Colorado she became active in politics and in the next decade ran for the Denver city council— unsuccessfully. In San Antonio Chicana Emma Tenayuca began a lifetime of concern for the working class while still in high school. She is best known for her leadership role in the 1938 pecan-shellers strike in that city. Also prominent in that same strike was Luisa Moreno who went on to play an active role during the forties in labor organizing in Southern California until she was deported to Mexico during the McCarthy anticommunism of the 1950s. Another contemporary and friend of Tenayuca was fellow Tejana Manuela Solís who helped organize workers in agriculture and the garment industry. Although successful in organizing for the South Texas Agricultural Workers Union, she was unable to achieve any gains for union members because of strong grower opposition and aggressive antiunion tactics. In 1938 Solís joined Tenayuca in organizing San Antonio pecan-shellers in their strike against the industry. In the post–World War II years she continued to be involved in causes of special concern to Mexican-Americans, including the Chicana movement

272  North from Mexico

and the rights of undocumented immigrants. Many Chicanas from this earlier period helped to form the cadres of groups organized by the movimiento during the late 1960s and to give them the benefit of their experience.10

8. Chicana Liberation The 1960s viewed the liberation of women as an important part of the civil rights movement of that decade. Although most Chicanas tended to view Anglo women’s liberation groups with some suspicion, they were influenced by the feminist movement and did greatly increase their own public activities. During this decade María Hernández became a frequent spokeswoman for young Mexican-Americans at rallies in Texas and was especially important in La Raza Unida party, stumping widely in support of its candidates. Virginia Musquiz, one of the founders of the LRU, took an active role at Crystal City, Texas, throughout the sixties in organizing the Mexican-American community for political action. In Colorado Helen Gonzales, wife of Corky Gonzales, played an energetic role in the Crusade for Justice, which was centered on family involvement from its beginnings. In the San Joaquin Valley of California, Dolores Huerta, after a stint in the CSO, became deeply involved in the UFW, in which she served as an organizer, contract negotiator, lobbyist, and union vice president. Soon she had become César Chávez’s prominent secondin-command and was widely recognized as an articulate and effective labor leader. She has received numerous honors, including having a corrido (ballad) written about her. To the south Julia Mount took an active part in a local MAPA chapter and in 1967 tried unsuccessfully to win a seat on the Los Angeles school board. Another Southern California Chicana, Francisca Flores, played a major role in politics and Chicana affairs; for a while at the end of the sixties she published a journal that addressed Chicana feminist issues, at first titled Carta Editorial and then in 1970 Regeneración, retitled after the earlier publication of Mexican revolutionary Ricardo Flores Magón. In the field of welfare rights, another active Los Angeles feminist leader in the late sixties was Alicia Escalante who was closely involved with the Chicano National Welfare Rights Organization and the militant Católicos por la Raza.11 As the movimiento progressed, active participation by Chicanas led them to become increasingly aware of the limitations on their roles. Within the movement they found that they were often victims of sexist discrimination, just as they were victims of racist and ethnic discrimination outside the movement. In 1969 the rejection of Chicana rights by the first Chicano Youth Conference in Denver served to dramatize their situation and acted as a further catalyst to the Chicana movement. Their new awareness sometimes led to heated debate with the male leadership of various movimiento groups and

Politics, Education, and Culture   273

ultimately led them to establish a number of their own organizations with varied constituencies and agendas.

9. Chicana Organizations Although male-dominated organizations had provided Chicanas with little opportunity for leadership experience, by the beginning of the eighties they had developed over forty regional and national groups aimed at freeing them from cultural and sexist role restrictions. The most important of these were La Comisión Femenil Nacional, the first strictly Chicana organization, founded in California in 1970; Chicana Service Action Center, also in California; MexicanAmerican Women’s National Association (MANA), founded in 1976 and headquartered in Washington, D.C.; Mexican-American Women’s Political Caucus, established in Texas; Mexican-American Business and Professional Women, also in Texas; and Mujeres Activas en Letras y Cambio Social (MALCS), a predominantly academic group founded in California in 1981. Another expression of the new Chicana liberation was the publication of a number of journals and periodicals with a Chicana interest or viewpoint. In 1973 the first of these, Encuentro Femenil, appeared and was soon followed by Hijas de Cuauhtémoc, La Comadre, and Hembra, among others. Also, by means of conferences and workshops, as well as by social and political pressure, these groups strove to raise Chicana consciousness, to advance their rights, to increase educational and professional opportunities, and to provide counseling and support for selfimprovement. During the National Chicano Political Conference in San Jose, California, in 1972 the Chicana Caucus, despite some opposition, was able to push through a position paper on the concerns of Chicanas, and later that year the Raza Unida party convention at El Paso made a promise to support Chicana issues.

10. Chicanas in Business and Government It was not only in organizations and social activism that Chicanas began to assert their rightful place; they also improved their status and position in business and government. Even before the second World War, Mexican-American women had begun to enter the labor market in increasing numbers. During the 1930s about one-third of border garment industry workers were Chicanas, and in the pecan-shelling industry they formed over three-fourths of the workforce. In agriculture a large number of wives and daughters worked in the harvest fields alongside the men. During the war years many more of them entered the labor market outside the home, working in war production factories, in shipyards, and on the railroads. By 1950 twenty percent of Chicanas were officially in the workforce, and by 1980 the percentage had risen to

274  North from Mexico

forty-four percent, comparing favorably with the forty-eight percent for all women in the United States. Chicanas are most heavily represented in lower-paying jobs in sales, service, and clerical work; relatively few are in higher-paying executive and administrative positions. The 1980 census indicated that, although Chicana employment had brought about some improvement, Chicano family income averaged only about seventy-five percent of Anglo family income. About twenty-five percent of Chicano families had incomes below the poverty level and fully fifty percent of female-headed Latino families remain below the poverty level. Chicanas from rural and small-town families, especially, had totally inadequate opportunities and income. Whereas Chicanos may encounter discrimination, as a group Chicanas suffer from a triple oppression because of their gender, race, and social class. In the business world only a handful of Chicanas have been able to achieve leadership roles. One of them, Romana Acosta Bañuelos of Los Angeles, developed a food-processing business in the 1950s, and she later moved into banking as well. In 1969 she was named Outstanding Businesswoman of the Year in Los Angeles and two years later was appointed Treasurer of the United States by President Richard Nixon. A decade later Katherine Ortega, the Nuevo Mexicana founder of a successful firm of accountants and a banker, became the second Chicana to be appointed Treasurer, by Ronald Reagan. Also in New Mexico Nina Otero Warren became a prominent leader in real estate and insurance in Santa Fe, although primarily an educator for most of her life. In Arizona María Elba Molina, an immigrant from Mexico, attained a high position in Tucson banking circles and then organized her own company, the J. Elba Corporation. Tejana Cathi Villalpando achieved distinction in communications and in the oil business, and in California actress Lynda Córdoba Carter has her own production company and is also associated with the cosmetics business. Another actress in California, Carmen Zapata, organized a bilingual theater company in Los Angeles in the early 1970s, and a decade earlier Hope Mendoza Schechter established a successful stenographic business. Nearly all these Chicanas also took active roles in community or political affairs. In government the most highly placed Chicana was Marí-Luci Jaramillo of New Mexico who, after a distinguished career in education, was appointed U.S. ambassador to Honduras in 1977. Many other Chicanas have held high positions in state and national government. Texas-born Belinda Cárdenas Ramírez in 1980 became the first woman to be appointed to the Civil Rights Commission, by President Jimmy Carter. In 1978 Polly Baca-Barragan, an active Democrat, became the first Chicana to be elected to the Colorado state senate; Linda Chávez, an equally active Republican from Colorado, was appointed a U.S. Civil Rights commissioner by Ronald Reagan and in 1985 was appointed

Politics, Education, and Culture   275

deputy assistant to the President, thus becoming the highest-ranking woman in the White House. Her subsequent bid for a senate seat from Maryland in 1986 failed. In California Gloria Molina, after government service in both Sacramento and Washington, in November 1982 became the first Chicana to be elected to the state assembly. Some of the women mentioned above and other Chicanas, like Cecilia Preciado de Burciaga of Stanford University, have played prominent leadership roles in education; while others, like Vilma Martínez and Graciela Olivárez, have become important leaders in national and community organizations and have had considerable impact far beyond the Mexican-American community.12

11. Chicano Culture: Alive and Well The objectives of the movimiento, while generally aiding Chicanas in their struggle for liberation and Chicanos in their search for civil rights, were not only political, economic, and social; they were also, and perhaps most importantly, cultural. Like their leaders, a majority of Chicanos saw themselves separated from, or deprived of, their cultural roots in American society. As a result, an important current in the movement was an emphasis on Mexican-American culture, especially in literature and the arts. Novels, short stories, essays, poetry, and plays flowed in ever-increasing volume from the typewriters of Chicano writers. Suffering from a deep sense of alienation from U.S. society, Chicano intellectuals saw a need to define their identity, to construct their history, and to explore and elaborate their culture. From the beginning Chicano leaders in the movimiento viewed literature as the primary way of achieving these objectives. The concern for defining rather than stereotyping the Mexican-American was paramount and is perhaps best illustrated by Corky Gonzales’s poem “I Am Joaquín” (1967). Another important aspect of this auto-definition is seen in the proud emphasis by writers during the 1960s of their Indian heritage; they gave their journals names like Aztlán and El Azteca and frequently used Aztec and Mayan drawings and symbols to illustrate and decorate their writings. Clearly the Chicano literary renaissance both nourished the movement and was nourished by it as writers explored the themes of discrimination, prejudice, exploitation, and the pains of cultural adaptation.

12. Literature The first significant Chicano novel was Pocho, the semiautobiographical story of a Chicano growing up in California’s Santa Clara Valley, by José Antonio Villarreal, published in 1959. Unfortunately it was a decade ahead of its time, and not until the movimiento gathered impetus did it achieve widespread renown. In 1965 a more critical view of socioeconomic injustices in American

276  North from Mexico

society appeared with the publication of The Plum Plum Pickers, a portrait of contemporary migrant life by a Californian by adoption, Raymond Barrio. By the late sixties and early seventies greater interest in Chicano literature had developed, and a number of anthologies became available to Chicano readers. In 1970 Richard Vásquez came out with Chicano, a novel of the immigrant Mexican experience in the United States. The publication in 1971 of the late Tomás Rivera’s award-winning . . . y no se tragó la tierra set high standards for a new generation of Chicano novels. In that same year the manuscript novel Bless Me, Ultima by Rudolfo Anaya received the Premio Quinto Sol award and was published in the following year. It was followed by Rolando Hinojosa-Smith’s prize-winning Estampas del valle y otras obras in 1973 and three years later by his Klail City y sus alrededores, which won the prestigious Casa de las Americas award. Among other notable Mexican-American authors are novelists Nash Candelaria, best known for his Memories of the Alhambra (1977); Ron Arias with his social commentary, The Road to Tamasunchale (1975); and John Rechy, author of City of Night (1963) and six subsequent novels dealing with homosexual life in large U.S. urban centers. These and the novels of other Chicano authors arose out of the Chicano experience and helped to conceptualize that experience as well as to make Anglos more aware of its essential meaning. However, the best of Chicano writing has had a universal application as well. Along with the novel, the short story has been a favorite Chicano literary genre, with a number of outstanding practitioners like Sabine Ulibarrí, Ron Arias, Miguel Méndez, Rosaura Sánchez, Sergio Elizondo, Estela Portillo Trambly, and Francisco Jiménez. Everywhere the signs indicate that Chicano literature is alive and flourishing and is increasingly being recognized by mainstream readers, students, and critics as an intrinsic part of American literature. In the field of drama the plays of Carlos Morton, Estela Portillo Trambly, and Luis Valdez have excited widespread interest. Valdez, important as well for his creation of the Teatro Campesino during “la Huelga,” is also outstanding as a producer of plays and film. In many ways he set the pattern for other Chicano dramatists. His Teatro Campesino inspired the formation of numerous other teatro companies in colleges, universities, and barrios all over the United States; in 1970 they joined together to form TENAZ (Teatros Nacionales de Aztlán) which has held annual drama festivals. However, the recent trend has been away from popular theater toward more diverse forms of drama. This move is illustrated by Luis Valdez’s 1978 production of Zoot Suit and his 1986 I Don’t Have to Show You No Stinking Badges. In films two other independent Chicano producers of note are Moctezuma Esparza and Richard “Cheech” Marín. Excellent television films like The Ballad of Gregorio Cortez and Seguín as well as the musical video Born in East L.A. have helped make their names widely known to the American public along with that of Valdez.

Politics, Education, and Culture   277

Poetry has always been a literary form of great attraction for Chicano writers, most of whom draw their inspiration as well as their metaphors from the social conditions in which they find themselves. Corky Gonzales is the best known poet of the movement because of his “I Am Joaquín.” Unquestionably the most widely known piece of Chicano literature, “I Am Joaquín” skillfully weaves together Aztec myth, Mexican history, and Chicano anguish. Poets Alurista, Abelardo Delgado, Lorna Dee Cervantes, José Montoya, Gary Soto, Omar Salinas, and Bernice Zamora are among those who follow closely behind Gonzales in stature. These and many other Mexican-American poets frequently bring together a hard realism with mysticism and great lyricism. Their poetry is often bilingual, sometimes combining Spanish and English powerfully in a single poem.13

13. Journals and Magazines Along with a rapid expansion in the field of literature the movement brought an explosion of new Chicano journals and periodicals, both academic and popular. Among the former the most important were El Grito, Aztlán, and La Red/The Net. El Grito: A Journal of Contemporary Mexican-American Thought was first published in late 1967 at Berkeley, California, by Quinto Sol Publications headed by Professor Octavio Romano-V as a forum from which Chicanos could articulate their concerns. Aztlán: International Journal of Chicano Studies and Research, now more simply subtitled A Journal of Chicano Studies, began in 1970 as a quarterly published by the Chicano Studies Center at the University of California, Los Angeles under the guidance of Juan Gómez Quiñones. La Red/The Net: The Hispanic Journal of Education, Commentary, and Reviews is a quarterly which focuses on educational issues of importance to Chicanos. It developed out of an earlier monthly newsletter of the National Chicano Research Network. These journals and others of more irregular issue like Campo Libre, Atisbos, Caracol, Con Safos, De Colores, Tejidos, Chismearte, and La Cucaracha carried poetry, short stories, and scholarly as well as popular articles on topics in the social sciences and arts, and made it possible for many young Chicano writers to break into print. In addition to these publications there were a handful of more specialized journals like the Chicano Law Review, which discusses legal issues affecting Mexican-Americans; Bilingual Review/ Revista Bilingüe, primarily a bibliographic and research journal; and Revista Chicano-Riqueña, now called The Americas Review, specializing in creative literature and literary criticism. At the popular level the best known periodicals are Caminos, La Luz, Nuestro, and Hispanic, all national journals aiming at the broader middle-class Latino readership and featuring articles on Latino personalities, cultural events, current news, and topics of special interest to the nation’s Latinos. By

278  North from Mexico

1980 Hispanic Business, published by Chicano Jesús Chavarría, had become the bible of the growing Mexican-American business class. There were also other more regional and local publications, often published by academic Chicano studies centers; most of them appeared irregularly and had disappeared by the mid-eighties. Because most Chicano writers found it extremely difficult to get published by mainstream presses, the late sixties and early seventies saw an eruption of Chicano publishers and small publishing houses. Inevitably many of these were ephemeral, and some, like Raymond Barrio’s Ventura Press, were, and are, one-man operations. Among the most important and productive private publishing companies were Quinto Sol Publications and its successors Tonatiuh International and Justa Publications in Berkeley; La Causa Publications, Inc., of Oakland, California, which published El Plan de Santa Bárbara; Trucha Publications in Lubbock, Texas; Arte Público Press in Houston; and Pajarito Publications at Albuquerque. Among the outstanding Chicano publishing organizations connected with academic institutions are the Chicano Studies Center Publications at the University of California, Los Angeles; the Center for Mexican-American Studies at the University of Texas in Austin; the Mexican-American Studies and Research Center of the University of Arizona in Tucson; and the Bilingual Press/Editorial Bilingüe, now at Arizona State University in Tempe.

14. Chicano Music and Art In music also the movimiento challenged Chicanos to express themselves creatively. Pushing aside society’s view that Mexican-American music had no place in the American mainstream, it stimulated and redirected earlier musical forms, kindling a Chicano musical renaissance. There was a surge of MexicanAmerican songs, especially in support of la Causa and in criticism of aspects of Anglo society’s treatment of Chicanos. The Delano grape strike led to a record album titled “Viva la Causa—Songs and Sounds from the Delano Strike.” Corridos, often a form of oral history, were composed and sung on the topics of migrant labor, mistreatment of Chicanos and Mexicans, the deaths of John F. Kennedy and his brother Robert, and the Vietnam War. In the entire border region so-called Tex-Mex music from south Texas developed wide popularity among Anglos as well as Mexican-Americans. Outstanding Mexican-American singers like Joan Báez, Linda Ronstadt, Vikki Carr, Trini López, Freddy Fender (Baldemar Huerta), and, most recently, Los Lobos have helped make Spanish language music more widely known and popular among Anglos and Chicanos. The 1987 success of Luis Valdez’s film La Bamba, on the life of Chicano vocalist Richie Valens (Ricardo Valenzuela), and in 1989 of Linda Ronstadt’s exhilarating and highly successful stage presentation and album, Canciones de mi Padre, are examples of the increased popularity of Mexican music.

Politics, Education, and Culture   279

One of the most interesting results of the movimiento has been the considerable expansion in Chicano art, especially mural art. Based on the traditions of muralists of the Mexican Revolution of 1910, particularly Diego Rivera, David Siqueiros, José Clemente Orozco, and Rufino Tamayo, Chicano mural artists have painted their works on walls in nearly every city and town of size in the Southwest. Using pre-Columbian Toltec, Aztec, and Maya figures and symbols, the Virgin of Guadalupe, the Mexican Revolution, and the Chicano experience as their themes, they painted their beliefs and feelings on walls, overpasses, and buildings in a variety of styles from harsh social realism to abstract expressionism. In addition to these general expressions of social protest, some of their mural art shows specific concerns as indicated by the incorporation of the “Huelga” black eagle and phrases such as “Viva La Raza” in their works. Other Mexican-American art is less overtly focused on ethnic themes. For example, Alfredo Arreguín, the outstanding Seattle artist, paints colorful scenes that are finely executed complex drawings, often with lush backgrounds reminiscent of tropical Mexico. Porfirio Salinas, a favorite of President Lyndon Johnson, who had many of his paintings hung in the White House, is famed for his more traditional “bluebonnet” Texas landscapes. The artist Luis Jiménez, while also a muralist of note, is perhaps best known for his glossy fiberglass and epoxy “Machine man” sculptures; and Consuelo González Amezcua is unique in her intricate multicolored ballpoint pen drawings. Although Mexican-American artists have been active in all fields of fine arts, including sculpture and wood carving, unquestionably the mural was the most distinctive and important, combining, as it did, historic roots and contemporary objectives. The Chicano Park murals of Barrio Logan in San Diego are perhaps the outstanding example of this combination.14 Some Chicano art historians also consider graffiti a form of popular art; indeed, at times there has been a melding of Chicano graffiti, often carefully organized sets of symbols, with mural paintings. In addition to graffiti, lowrider automobiles, usually remodeled and meticulously redecorated both inside and out, have been seen as proud expressions of popular art among Chicano youths. There are low-rider clubs which hold regular meets and there is even a magazine, appropriately named Low Rider, a Southern California publication with a circulation of over 100,000, mostly among young Chicanos.

15. Spanish Language Use As another result of the movimiento (and of the great increase in Spanishspeaking population) there was in the 1960s and 1970s a rapid expansion of Spanish language newspapers as well as radio and television stations. By the beginning of the eighties over one hundred such newspapers were being

280  North from Mexico

published, mostly weeklies but some dailies. Local papers, for the most part weeklies, arise and wither away with discouraging regularity; but among the dailies La Opinion of Los Angeles and La Prensa of San Antonio have shown greater vitality and some claim to wider circulation. In addition, over two hundred radio stations and seventy television stations broadcast in Spanish either partially or solely. There was even a Mexican-owned network of Spanish-language television stations and cable companies—the Spanish International Network. However, while there has been a great upsurge in Spanish language television, it remains largely under Anglo management and control. Spanish broadcasting is most common in the large cities of the Southwest, and in New York, Miami, and Chicago. One of the basic causes and results of this media expansion has been the equally rapid expansion of advertising directed specifically at the Mexican-American market both through language and various ethnic preferences. As the number of Chicanos in clerical, technical, professional, managerial, and other skilled and semiskilled occupations increased, so did incomes; and more moved up into the middle classes. Manufacturers and retailers saw the increasing possibilities in the raza market and responded. One result of the expansion of Spanish-language media has been the development of an even greater cultural cohesiveness and retention of raza consciousness. The increase in Spanish language advertising also served to revivify and expand interest in the language. As with other immigrant groups and their native languages, some decline in the use of Spanish among American-born Mexican-Americans has been noted. The movement out of the barrios and out of the Southwest, higher levels of education, and an increase in acculturation have all contributed to this decline. Another factor has been the negative American attitude toward the use of Spanish; until recently in many schools of the Southwest, speaking Spanish was severely discouraged or even expressly prohibited, sometimes under penalty of physical punishment. On the other hand, segregation, continuing high levels of immigration from Mexico, and the relative ease of crossing the border for business and tourism have constantly reinforced its use. In addition, there was to some extent a sort of “Spanish is beautiful” sentiment, language being the one common denominator among Latinos, as skin color was among blacks. The movimiento with its emphasis on literature and ethnic pride also contributed to greater concern for the study of Spanish, as the bilingual heritage of Mexican-Americans was stressed.

16. Bilingualism Bilingualism became a household word in 1974 when the U.S. Supreme Court, in the case of Lau v. Nichols, unanimously held that the failure of schools to

Politics, Education, and Culture   281

provide programs tailored to meet the special language needs of minority children violated the 1964 Civil Rights Act and the Fourteenth Amendment to the Constitution as well. Special programs for Spanish-speaking (and other) children had already begun a few years before with the 1968 Bilingual Education Act, but the Lau decision greatly and rapidly expanded bilingual classes as federal funding rose from $7.5 million in 1969 to $107 million in 1980 and as bilingual education laws were passed in many states. However, the Lau case and state legislation also led to a strong reaction from conservative Anglos who felt that bilingual programs challenged basic American assumptions regarding cultural assimilation and might eventually lead to a separatist movement akin to that in Quebec, Canada. Letters to the editor and press interviews with anti-bilingual education leaders spoke of the “menace” of Spanish and even referred to it as “subversive.” The members of the MexicanAmerican community reacted to these emotional nativistic views with some bewilderment. An overwhelming majority of them recognized the need to speak English in order to function in U.S. society and to improve their economic and social position, but they rejected any implied inferiority of the Spanish language. They also saw bilingual programs as a way for their children to learn English more rapidly as well as a means of focusing public attention on the educational plight of Mexican-American children, particularly the significant number who entered school speaking little or no English. Some viewed bilingual programs as a way of fighting discrimination and increasing respect for their language and culture. Although funding for bilingual education was never more than a tiny fraction of the federal education budget, in the rising swell of xenophobia spreading over the country it became highly controversial. Under the Reagan administration in the 1980s funding for bilingual education was regularly reduced. One positive result of the programs was that many Mexican-American parents were stimulated to take a more active role in the schools in support of bilingual education. Some of the adult education programs also provided older Mexican-Americans with a way to greater economic security and mobility through an improved command of English. An important negative result was the rise of nativistic groups like U.S. English which added to a racist backlash seemingly on the rise in the United States. Angered by her inability to find an English-speaking clerk in a Dade County, Florida, government office, one Emmy Shafer founded English Only in 1978. Because of the xenophobic mood in the United States the movement quickly took off. A private national organization named U.S. English, having gotten an English-only amendment introduced into the U.S. Senate, by its honorary chairman, S. I. Hayakawa (but not passed), has since conducted regular, well-financed campaigns to get legislation passed at the state level. By the end of 1988 seventeen states had approved legislation making English their

282  North from Mexico

official language, and over forty towns, including tiny Los Altos, California, had followed suit. U.S. English urges a written English-proficiency test for naturalization. Shrill and hostile in tone, U.S. English aims overwhelmingly at Spanishspeakers and talks about “defending” English, as if it were under attack. Illustrating the movement’s bias were racist remarks in October 1988 by its cofounder John Tanton that caused the president of U.S. English, Linda Chavez, and others to resign from the organization. Surveys of Chicanos have indicated that even in Los Angeles and San Antonio eighty-nine percent are either bilingual or monolingual English-speakers. Nearly all Mexican-Americans think that America should have only one public language.15

17. The Movimiento: Radical or Reformist? Like the Mexican-American population, the movimiento was not homogeneous. It was a complex of groups and individuals with various agendas and strategies to achieve them. Its aims ran from reformist to revolutionary, from traditional forms of economic and social protest to violence, separatism, and cultural nationalism. It started as a reaction to Anglo racism and discrimination, and later, as it evolved, it began to espouse cultural nationalism. On the left it included anticapitalist tendencies and ideas of separatism. From the beginning it demanded civil rights and equality of opportunity in a more pluralistic American society. As it became more politicized, it radicalized even the usually staid Mexican-American middle class, and it developed what many referred to as the concept of Chicanismo. Chicanismo, like the movimiento, meant different things to different people. It cut across the usual boundaries of sex, class, region, and generation. Generally it rejected the idea of assimilation or acculturation. It viewed la raza as united on the basis of a common life experience and pride in heritage and culture, rather than on a basis of economic status, education, or class. It argued that, as an alternative to assimilation, Chicanos had a right to cultural selfdetermination and autonomy. In the furtherance of these ideas it agonized over identity and felt it was necessary to develop Chicanos’ self-identity, their culture, and their history, thereby constructing a new self-image. It saw Mexican-Americans cut off from their roots and dehumanized, as they became transformed into an economic commodity. As the foregoing indicates, the movimiento included within its mainstream widely differing programs and goals. Not all Mexican-Americans were enthusiastic supporters of the movimiento and of Chicanismo. Many older members of la raza found the aggressive stance of youthful activists distasteful, and some even found it threatening. Some had doubts that militant ethnic solidarity was the most effective tactic to achieve the economic and social

Politics, Education, and Culture   283

advancement of la raza. Others abjured the word Chicano itself as connoting radicalism. United States Congressman Henry B. González (Democrat, Texas) not only rejected the movimiento, but also denounced both the Mexican American Unity Council of Texas and MAYO and their programs as reverse racism. On the floor of the House of Representatives he criticized the Ford Foundation for its funding of activist groups like MALDEF and MAYO. Made up of varied economic, regional, and social groups with greatly differing life experiences, the Mexican-American population was and is extremely diverse. While a majority belongs to the working class, it includes atomic scientists, doctors, university professors, and administrators as well as barrio youths. There still is lacking an easily identifiable common goal or target around which all Chicanos can rally. In the early stages of the movement the leadership tended to stress emotional and rhetorical appeal combined with strong advocacy of cultural nationalism. The sometimes verbally belligerent posturing was especially attractive to the more alienated barrio youths of large urban centers. However, the revolutionary rhetoric of the militants was just that, and they never fused action with rhetoric. On the other hand, César Chávez with his emphasis on social justice and nonviolence appealed both to harvest agricultural workers and to urban middle classes. His selfless commitment to la causa attracted large numbers of people of good will both among Chicanos and in the Anglo community. However, not even César Chávez was acceptable to all. Mexican-Americans were not able to agree on the course that la raza should follow. Most felt that they should be accepted as Americans without reservation and without having to give up completely their distinctive cultural heritage. Most seemed to feel more comfortable working within the system, trying to bring it in conformity with its own democratic ideals and using it to achieve their objectives. With the decline of the movimiento after 1975, Mexican-Americans continued to look, in vain, for direction, guidance, and united leadership. The movimiento developed no single ideology or leader. The absence of a unifying ideology led to divided leadership, fragmented followers, and therefore a lack of political clout. Today differences of opinion concerning means and tactics for obtaining civil and political rights continue to divide Chicanos. However, conflict within the leadership usually arises over tactics rather than goals. César Chávez, the only Mexican-American who might possibly have successfully assumed the role of the leader, consistently refused to put on that mantle, limiting his leadership largely to agricultural workers.

18. The Movimiento: Success or Failure? How successful has the Chicano movimiento been? Clearly it has led to some improvement for Chicanos in many areas of American life. From the

284  North from Mexico

conflictual sixties and early seventies have come increases in civil rights and economic opportunities. It helped accelerate the dismantling of some of the more repressive aspects of the social order. It did contribute to a reduction in the political influence of conservative rural interests, the expansion of the Voting Rights Act in 1975. It promoted a considerable increase in the size of the Mexican-American middle class and a smaller but significant movement into professional ranks. It was important in the founding of a wide spectrum of raza professional organizations such as the Mexican-American Engineering Society, La Raza National Lawyers Association, National Association of Hispanic Journalists, and Mexican-American Educators. In the area of political power it has had limited but significant success. Most important it reawakened among many Chicanos a new consciousness of cultural identity and community. It aroused a dormant pride in their Mexican roots and in the Spanish language, and it brought to them an awareness of their potential as it renewed and expanded their hopes. However, some of the successes have been turned around in the eighties and many of the principal issues that faced Chicanos in the 1960s still face them today. In a few areas of the country Mexican-Americans have become a political power to be reckoned with as a result of the movimiento and the development of their own leadership. Blatant racism, strict social segregation, and crude machine politics no longer rule, and openly exploitive economic scenarios have diminished. Improvement in social and economic conditions has occurred. Chicanos have gained entry into industries and better-paying jobs that were formerly closed to them. Between 1930 and 1987 Mexican-Americans in professional, technical, managerial, and skilled categories increased from less than one percent to nearly nine percent. Despite this increase, MexicanAmericans fall far short of their fair share of better jobs and continue to experience lower levels of income and of upward mobility. Within almost all occupations they tend to hold entrance-level, low-pay positions and still experience varying degrees of economic, social, and political discrimination. In education there has been improvement, but not nearly enough. MexicanAmerican children continue to find themselves handicapped by inferior schooling and unequal educational opportunity. Census reports from 1988 indicate that Hispanics, of whom Chicanos form over sixty percent, finish high school in greater, but still disappointing, numbers. About fifty-one percent of Hispanics over twenty-four have completed high school, compared to thirtytwo percent in 1970, but compared to seventy-eight percent of the rest of the population. Of Chicanos graduating from secondary school, ten percent have gone on to complete college, compared to about five percent in 1970. A disappointing imbalance exists in the teacher population. In 1970 only four percent of primary and secondary teachers in the Southwest were Chicanos, while Mexican-Americans made up seventeen percent of the student body. There

Politics, Education, and Culture   285

seems little prospect of an extensive breakthrough in educational opportunity for Chicanos in the near future. According to the Bureau of the Census, in March 1987 there were 11.8 million Americans of Mexican descent or origin. Since the 1960s Chicanos have voted in larger numbers and more of them have run for public office. Nevertheless, they are underrepresented at all levels of elective office, on juries, and among judges and law-enforcement officers. For a variety of historical reasons Chicanos have low levels of participation in the U.S. electoral process. In the 1980 Presidential elections only thirty-six percent of eligible Hispanics registered and only thirty percent voted, compared to sixty-nine and sixty-one percent for non-Hispanics. This lack of interest in registration and election participation seems to be directly related to educational and economic levels as well as to racist discrimination, among many factors. By the mid-1970s the decline of the movimiento’s first phase was clearly evident. Militancy among Chicano organizations was manifestly languishing, as they moved from the politicization and activism of the 1960s to the conservatism of the 1980s. Reies Tijerina, subdued by his prison term and by rejection from the other principal leaders, had returned to his northern New Mexico interests. Corky Gonzales was isolated in Denver, and José Angel Gutiérrez had split the LRU by making a pilgrimage to Havana. César Chávez was losing hard-fought gains in union contracts and strengthening his ties to the California Democratic leadership. Disagreement and bitter controversy among second-level leaders in the LRU and UFW, as well as older tensions between activists of middle class and working class backgrounds, further weakened the movement. American society was also less concerned about Mexican-Americans and other minorities as well as their problems. The civil rights crusade in the Southwest had been seriously weakened by a combination of gains, accommodation, external pressures from the authorities, and internal friction. Affirmative action seemed less important, and even many Chicanos seemed more concerned about their economic improvement than about government help to improve quality of life for the poor. By the latter years of the 1980s the heralded “Decade of the Hispanic” had clearly failed to deliver its promises. At any event, the movimiento was an exciting and impressive ferment and force for change, whatever its ultimate impact may turn out to be. The change was not absolute, of course. However, the movimiento had provided useful training and experience for a new generation of young leaders. Many applied their skills and ideas to local political groups and to organizations like MALDEF and SVREP, which were more acceptable to mainstream America and which had as their concerns equality of educational opportunity, affirmative action hiring, bilingual education programs, and social justice for all— including Mexican-Americans and undocumented immigrants.

286  North from Mexico

19. Undocumented Immigrants Discussions and resolutions at the 1972 Raza Unida convention indicated the weakening of the movimiento. The delegates moved away from earlier preoccupation with self-identification and the issue of ethnic nationalism. The new issue of the late seventies and early eighties, as indicated by the convention, was immigration and particularly undocumented immigration. By the late 1970s the era of cultural nationalism and its ideology had passed, and Chicanos were moving from a doctrinaire ethnic consciousness to broader and more pragmatic viewpoints. After 1975, as emphasis on civil and human rights was more widely held, the issue of the undocumented worker began moving to a central position in the movement and the community.16

19

North from Mexico Matt S. Meier

It would be rash to try to interpret the historical experience of the MexicanAmerican without carefully studying the important role of Mexican migration to the American Southwest. When Carey McWilliams gave his history of Mexicans in the United States the title North from Mexico, not only was he paying tribute to the most salient aspect of their life experience, he was also, aptly, using a phrase that implied a long-term, ongoing process. Deeply rooted in the Mexican past, the movement north from the central Mexican plateau region began in prehistoric times and has continued over the ensuing centuries. Today people from this area form the largest national immigrant group arriving at U.S. frontiers. Crossing a border region that is in many ways a unique area of bilingual accommodation, of welcoming gateway and frightening barrier, they have had an exceptional impact in this century on Mexicans in the United States. Because of their long-term cultural, economic, and social influence on Mexican-Americans, the history of their migration demands a detailed look.

1. North from Mexico: Three Waves When Mexicans migrated to southwestern United States after the Treaty of Guadalupe Hidalgo in the second half of the nineteenth century, psychologically they did not immigrate, but rather migrated from one area of Mexican culture to another. This cultural continuity and contiguity, plus the relative ease of travel back and forth, resulted in a commitment to the move that tended to be less final and permanent than that of contemporary European immigrants. Today many Mexican-Americans have relatives among the recent immigrants and intermarry with them. In addition, Chicanos and the new immigrants share language, culture, occupations, and parts of their histories, and often live together in the barrio. Immediately after World War II Chicanos, for

288  North from Mexico

the most part first-generation Mexican-Americans, began to become more deeply concerned and involved with issues resulting from the immigrants’ and undocumenteds’ problems in American society. Between 1900 and 1989 approximately 2.4 million Mexicans with formal papers immigrated to the United States in one of the great mass population movements in the history of the hemisphere. In addition, there has been a large undocumented migration, the numbers unknown but estimated to be as many as twenty million.1 Although this migration has been a continuous process, for convenience in study it may be divided into three waves. The U.S. census of 1900 indicated that there were 103,000 people of Mexican birth in the country, ninety percent of them in the states of Texas and California and in the New Mexico and Arizona territories. It would appear then that the movement from 1848 to the end of the century was of fair size in comparison to the relatively sparse border area population. This almost continuous unregulated nineteenth century flow conditioned later migration patterns. The first wave of immigration in this century, pushed by the Great Revolution, brought as many as 100,000 Mexican nationals by 1914. A majority of these border-crossers went to Texas and many settled in San Antonio. Heavily made up of refugees, this wave was fairly heterogeneous in its composition. The second wave, which arrived from the time of the first World War to about 1930, numbered perhaps one million. These immigrants were mostly small farm operators, tradesmen, craftsmen, and peones who came largely for economic reasons. Because of their numbers they culturally overwhelmed the earlier Mexican-American population except in New Mexico, which they generally avoided because economic opportunities were better in California, Arizona, Texas, and the Midwest. Simply because of their numbers they established a majority of the present-day Mexican-American communities in the United States and significantly shaped Chicano culture and determined a large part of its content and perimeters. The third wave began with World War II and has continued to the present. Unlike the first two waves, which were partly set in motion by political upheaval in Mexico, this wave came almost solely because of perceived or hopedfor economic benefits. Although all three waves had a large majority of adult males, this wave had a greater percentage of women, children, and elderly. It was made up of visaed permanent immigrants, braceros, and undocumented workers from more varied regions of Mexico than the first two waves. All three waves might also be described in terms of the immigrants’ status. All included in various proportions: 1. permanent immigrants, visaed and undocumented. 2. commuters: daily, weekly, and seasonal, most without permanent documents; 3. braceros, on contract and freelance; and 4. businessmen, students, shoppers, and tourists. Obviously, not all remained in the categories in which they entered the United States; nor did all remain in the United States.

North from Mexico   289

2. Undocumented Entrance In the years after World War II, the largest immigrant category was that of undocumented workers. They were also unquestionably subject to much petty racketeering on both sides of the border and were among the most ill-treated workers. They took over much of the harvest farm work in the West and Southwest and later also moved into urban employment in construction, service work, and light manufacturing.2 In the years of the second World War the greatly increased need for farm, and later railroad, labor in the Southwest led to a formal bracero agreement between Mexico and the United States. The program brought about a quarter of a million Mexican workers to the United States between 1942 and 1947. These workers “advertised” to their relatives and friends in Mexico the work and pay available in the Colossus of the North. This publicity increased the number of workers who wanted to go to the United States but could not do so legally. Moreover, during the war years the relatively large number of contracted braceros probably acted as a check on undocumented migration to some parts of the Southwest. Texas was denied braceros by the Mexican government because of extremely poor treatment in that state of Mexican nationals as well as MexicanAmericans. The result was a heavy movement into Texas of undocumented workers to fill labor demands, with the Immigration & Naturalization Service (I&NS) making little real effort to hinder their entrance. Late in 1946 the Mexican government estimated that there were about 120 thousand undocumented workers in the U.S. border region, most of them in Texas. During the postwar years various factors contributed to the larger number of illegal entrants. In the decade of the forties there was a rapid expansion of irrigated market agriculture in northern Mexico; this attracted more workers from central Mexico than there were jobs available. As a result a massive buildup of unemployed agricultural workers took place in Mexican border towns. Between 1940 and 1950 the population of Matamoros rose from 54,000 to 128,000, Nuevo Laredo went from 31,000 to 57,000, Juarez from 55,000 to 122,000, Nogales (Sonora) from 14,000 to 25,000, Mexicali from 44,000 to 124,000, and Tijuana from 22,000 to 65,000. The pressure of these workers seeking employment was complemented on the American side by an increased need for agricultural labor, as more than seven million irrigated acres were added to the farmlands of the West and Southwest between 1945 and 1955. In the five years between 1945 and 1949, cotton lands in the lower Rio Grande Valley jumped from 250,000 to 600,000 acres. At the same time seasonal demand for workers increased dramatically as growers in the Texas Winter Garden district tried to get their fruits and vegetables to market as early as possible in order to benefit maximally from higher prices at the beginning of the season.

290  North from Mexico

3. Interim Bracero Hiring The World War II government-to-government bracero program ended in December 1947 when the special legislation implementing the program was terminated because wartime labor needs had declined. In the four years from 1947 through 1950, importation of workers from Mexico continued on a semiindividual basis with farm groups like the Cotton Growers Association making most of the contractual agreements and U.S. government agencies merely expediting the arrangements. With the end of the wartime program, illegal entries soared, and the smuggling of undocumenteds became a lucrative border business of “coyotes” and “polleros” or “pateros.”3 Concerned about the widespread abuse of undocumented workers, the Mexican and U.S. governments, in a series of meetings during January and February 1947, agreed to take steps to reduce illegal border crossing and to legitimize the undocumenteds already in the United States. Beginning in March 1947 the United States initiated a program of legalizing or “drying out” mojados in the U.S. border area by converting them to contracted braceros. Of the nearly 300,000 Mexican braceros contracted between 1947 and 1950 more than 200,000 were legalized undocumenteds already in the country. To discourage further illegal entry the Border Patrol stepped up its activities along the frontier, and between 1947 and 1950 the number of illegal entrants apprehended and returned to Mexico jumped from 196,000 to 469,000 per year. While the Mexico-United States agreement provided minimal guarantees for undocumenteds already in the country, clearly it failed to control illegal entrance. In September 1947 Mexico ended its support of the agreement to legalize undocumenteds, arguing that the wages paid them in Texas, where the majority of them worked, were too low and that other guarantees of the international agreement were not being respected. The following January the Department of Labor took over supervision of the braceros from the Agriculture Department, and during the next month a new Mexico-United States agreement on imported contract labor was hammered out. It outlawed the use of undocumenteds by growers and pledged the two countries to strict border regulation. Pursuant to this agreement, later that year the U.S. Employment Service set the prevailing wage in the lower Rio Grande Valley in Texas at 40 cents per hour despite anguished and vehement objections from growers.

4. The El Paso and Tivoli Incidents However, the large agricultural interests did not give up in their demands for cheap labor. In the fall of 1948 they insisted on a $2.00 to $2.50 per hundred pounds wage in the cotton harvest. Mexico contended with equal vigor that

North from Mexico   291

the going wage was $3.00, and refused to make braceros available at a lower rate. An impasse was reached. Early in October growers were complaining that their cotton would rot in the field unless they got Mexican nationals to pick it. In spite of the February agreement with Mexico, an interagency meeting was held in Washington to discuss opening the border to Mexican workers (“open border,” long a favorite idea of Texas growers). President Harry Truman expressed his opposition to this step, but on the October 16–17 weekend the border at El Paso was opened to Mexican nationals by the I&NS. An estimated 4,000 workers streamed across the border and were loaded into the trucks of growers’ agents. Within a week some 6,000 to 7,000 harvest workers crossed into Texas, enough to pick the bulk of the cotton crop. On October 18 the Mexican government formally protested this U.S. violation of the February agreement, declaring that the United States had abrogated that agreement by its illegal unilateral action. Two days later the I&NS denied having opened the border, but on October 22 the White House sent Mexico a note of “profound regret” for the U.S. action—an apology the Mexican government accepted. Soon after the El Paso incident the wage per 100 pounds of cotton was lowered to $1.50. This El Paso incident was followed in June 1949 by what is sometimes referred to as the Tivoli incident. It illustrates the inhumane treatment that braceros were often subjected to in the border region. A group of Mexican braceros working on a farm near Tivoli (fifty miles northeast of Corpus Christi) selected three representatives to present a complaint to their employer. The latter had the three arrested on disorderly conduct charges and they were sentenced to sixty days in jail in lieu of a $75 fine. Somehow Governor Beauford H. Jester was informed of the incident, investigated, and was convinced the workers had been wrongfully jailed. He sent Texas state troopers to Tivoli in cars flying the governor’s standard to secure the release of the three representatives. In the aftermath their employer was forced to satisfy the complaints of the Mexican workers. United States and Mexican concern about undocumenteds continued, and in the late summer of 1949 a new intergovernment agreement was reached. It stressed interdiction by both governments of illegal border-crossing, provided for further “drying out” of undocumenteds, and denied legal workers to growers who were hiring Mexicans without documents. Despite this understanding between the two governments, the number of undocumented workers, as measured by the number of Mexican nationals without papers who were apprehended, rose rapidly—193,852 in 1948; 289.400 in 1949; and 469,551 in 1950. In spite of the continued and rapid increase in their numbers, few undocumenteds in the United States were legalized after 1950 because of new agreements between Mexico and the United States. The government of Mexico was not pleased with the situation under the 1949 understanding and continued to

292  North from Mexico

complain that its nationals were ill-treated, especially in Texas and Arkansas, and that many growers were ignoring the minimum guarantees of the agreement.4

5. The Second Bracero Program When the Korean conflict broke out in 1950, the Mexican government informed the United States that it would be necessary to return to a World War II type of government-to-government arrangement if the United States wanted large numbers of workers. As a result Public Law 78 was passed by the U.S. Congress in July 1951 and in August became the basis of the Migratory Labor Agreement of 1951 with Mexico. Under PL78 and other legislation “drying out” was ended, braceros were denied to growers who hired illegals, and a list of ineligible employers was drawn up. The new arrangements made the U.S. Department of Labor the contractor, and the Secretary of Labor was given the task of making certain that employers fulfilled their obligations under the terms of the agreement. PL78 was originally given a life span of two years, but was repeatedly renewed, with modifications, for the next dozen years. Although the supporters of PL78 optimistically assured the nation that it would end the problem of undocumented entrance, in fact illegal entries continued to soar, reaching half a million in 1951, 839,000 in 1953, and over a million in 1954, as measured by undocumenteds returned to Mexico.5 Obviously the 1952 McCarran-Walter Immigration Act, which made concealing and harboring illegal entrants a punishable offense (but excluded employment from concealing and harboring—the so-called Texas Proviso) had no impact on the problem of undocumented entry. It also provided, under Section H-2, for the importation of workers under contract for up to three years. Almost immediately the Associated Farmers of California, fearing that Mexico might not renew the 1951 agreement (while putting pressure on the Mexican government to do so), opened negotiations in South Korea for agricultural workers. Under pressure from high levels of unemployment at home, Mexico repeatedly renewed the bracero agreement. Many American growers resented Mexico’s continuing demands for higher wages, its blacklisting, and especially its refusal to establish recruiting stations at the border. As earlier, they wanted essentially an “open border.” In partial support of their complaints the U.S. Secretary of Labor argued that the agreement with Mexico needed changes in order to end the problem of undocumented workers. In January 1954 the United States decided unilaterally to contract at the border and announced interim border hiring. At Tijuana and Mexicali Mexican police, later reinforced by army troops, tried in vain to prevent Mexican nationals from crossing the border into California. About 3,500 Mexican workers crossed over and were certified by the U.S. Labor Department;

North from Mexico   293

at the same time undocumenteds were “dried out” by the technique of having them momentarily recross the border and come back as contracted workers. Because of these U.S. pressure tactics, on March 10 Mexico signed a new bracero agreement which provided for recruitment at six Mexican border towns and included the possibility of unilateral hiring by the United States. Agricultural employer clout was also evidenced by the activities of the Border Patrol. Many public officials in Texas used undocumented workers in their businesses, and border congressmen took the lead in reducing Border Patrol appropriations in 1952 and 1954. At times border patrolmen were threatened by growers, I&NS vehicles were disabled, and patrolmen were given annual leave at the height of the Texas cotton harvest season. Both undocumented Mexican workers and braceros increased dramatically in numbers and peaked at perhaps 1.5 million in the mid-1950s. This massive influx of Mexican nationals in turn resulted in economic pressure on Chicanos in the Southwest. Many employers deliberately discouraged local applicants for agricultural jobs and used braceros and undocumenteds maximally. Some crops were completely dominated by Mexican nationals. Their widespread use resulted in underemployment and unemployment for local Chicano farmworkers and tended to hold down agricultural wages to bracero levels. As labor surpluses developed along the border, especially in low-skill and low-pay job categories, a considerable out-migration of Mexican-Americans took place from such labor reservoir areas as El Paso, San Antonio, and Los Angeles. Texas became both the largest importer of Mexican labor and the largest exporter (to other states) of Chicano workers as thousands of MexicanAmericans were forced to take to the migrant circuits because of depressed wages, poor working conditions, and limited job availability in the areas where they lived.

6. Operation “Wetback” Early in 1954 Attorney General Herbert Brownell announced an imminent mass roundup of undocumenteds, citing possible illegal entry of subversives as his principal reason for doing so. In May of that year the Justice Department informed the nation that on June 17 a Special Mobile Force under General Joseph Swing, Commissioner of the I&NS, would begin operations in Southern California. In July the roundup moved from California to Texas, again highly publicized in advance. There was some infringement of civil liberties in the operation, and the treatment of those arrested by the Special Mobile Force was often harsh and sometimes abusive. By the end of 1954 more than a million undocumenteds had been apprehended and returned to Mexico. Between 1950 and 1955 a total of about 3.7 million undocumented nationals were sent back to Mexico, more than all Mexicans reported as immigrants to the United

294  North from Mexico

States since 1848. In its 1955 Annual Report the I&NS confidently announced that “the wetback problem no longer exists.” The Mexican-American community has been ambivalent in its feelings about undocumented workers. On the one hand they were members of la raza; on the other, they clearly took jobs from la raza in the United States. Neither undocumenteds nor Mexican-Americans felt that the former’s illegal status was criminal, or even that any moral or social stigma was attached to it. Many undocumented Mexicans had been in the United States as long as ten years and were fully integrated into the Mexican-American community.6 The postwar bracero movement reached its peak in 1956 when 445,000 contracted Mexican nationals were brought into the United States to work, primarily in agriculture, seventy-five percent of them in Texas and California. Largely because of increasing mechanization, after the mid-1950s the demand for agricultural workers began to taper off and then decline, especially after 1960. Between 1949 and 1965 farm employment dropped from nearly ten million to a little over 5.5 million. At the same time support in Congress for the bracero program began to dissipate as the power of agricultural interests in government declined. The enabling legislation to extend the bracero program was passed regularly, but only after long and often heated discussion; and difficulties in arriving at agreement with Mexico increased. Already in 1959 Congressman George McGovern introduced a bill to phase out the program, as labor, church, other anti-bracero groups, and such notable MexicanAmericans as César Chávez and Ernesto Galarza pressured more strongly for an end to the importation of Mexican contract workers. In 1962 the extension of PL78 limited bracero contracts to six months, prohibited the use of braceros on power equipment, increased occupational insurance benefits, raised minimum wages, and set criteria for the withdrawal of braceros in the event of strikes. In general the legislation guaranteed braceros better working conditions and greater benefits than American farm labor was entitled to by law. In fact, there were instances reported of Chicanos crossing over into Mexico in order to be contracted at the border as braceros. Finally, amid alarmist predictions of dire adverse effects on U.S. agriculture by supporters of the bracero program, it expired December 31, 1964. Agribusiness continued on its successful way.7 Between 1951 and 1965 a total of 4.5 million braceros worked in the United States, plus more than five million undocumented workers. A majority of both groups were repeaters, so it is impossible to know exactly how many individuals were involved. As time went on the program attracted Mexican workers of more diverse backgrounds. On the whole, more of the postwar braceros were married and older than the World War II braceros. Only about eight percent of them remained in the United States. For this and other reasons their influence on Mexican-American culture was much less than that of the nearly one

North from Mexico   295

million in the second wave between 1917 and 1930. However, economically Mexican-Americans were deeply affected by the bracero program since they were the workers with whom braceros usually competed for jobs. Further, bracero employment unquestionably helped to hold down wages of MexicanAmericans in the Southwest and to impede improvement in their working conditions. The various bracero programs from 1942 to 1965 brought approximately five million Mexican workers to the United States and made many more Mexicans aware of the financial potentialities in temporary U.S. employment at what were to them high wages. As long as American employers were able to obtain virtually unlimited numbers of braceros, there was limited pressure for undocumented workers. However, termination of the bracero program in 1964 left thousands of American employers addicted to cheap foreign labor and millions of Mexicans equally addicted to the financial benefits of U.S. employment. During the years immediately following the termination of the second bracero program there were a number of unsuccessful attempts in Congress to revive it. However, the braceros’ place in the U.S. agricultural economy was quickly taken by undocumented workers, commuters, permanent visa immigrants, and H-2 guest-workers. H-2 workers, so called because their importation was authorized by section H-2 of the McCarran-Walter Immigration and Nationality Act of 1952, averaged 35,000 to 40,000 annually within a decade of the ending of the bracero program and peaked at 69,228 in 1969. Permanent visa Mexican immigrants, often referred to as “green carders” after the color of their documentation, rose to more than 92,000 by 1978 and peaked in 1981 at 101,268. By 1979 California counted 513,000 permanent visa Mexican immigrants in the state and Texas had nearly 300,000. In addition there was a backlog of 173,000 Mexicans waiting for visas to enter the country. The number of commuters, those who cross the border on a daily, weekly, or other regular basis to work in the United States, also rose rapidly. Most of them were “white carders,” so called from their temporary entrance visas; but there were also green carders, and U.S. citizens among their numbers. Lastly, there was a large, rapidly increasing, and seemingly inexhaustible supply of undocumented workers. As demand for farm workers declined in the late sixties, undocumenteds in ever-increasing numbers found work in low-wage border businesses like furniture, clothing, plastics, food service, and hotels, as well as in agriculture.8 In an effort to reduce the number of undocumented workers crossing into the United States, in 1966 the Mexican government created bonded manufacturing districts along its northern border. The basic idea was to bring the work to the Mexican workers instead of the workers going to the United States to the work. This Border Industrialization Program led to the establishment

296  North from Mexico

of numerous maquiladoras, factories in which labor-intensive work was performed. By 1970 over 200 such factories had been established and were employing about 19,000 workers. By 1986 there were some 900 maquiladoras with 255,000 employees. The expectations that these factories would reduce undocumented entrance into the United States were frustrated, however. Instead of hiring Mexican males, the maquiladoras hired mostly young women who were attracted to the border cities in large numbers. In some factories the employees were ninety percent young females. One of the recent byproducts of this development has been a considerable increase in the number of Mexican women who have joined the stream of undocumenteds entering the United States. Recently observers along the border have reported that young women, almost two-thirds of them under thirty, now make up between twenty and fifty percent of the undocumented workers who enter the country.

7. U.S. Immigration Policy Of all the issues affecting Mexican-Americans, none is less amenable to sweet reason than immigration. In this issue, even more than in bilingualism, vagueness, fear, and emotion have long ago driven out logic, reason, and any sense of fairness. The immigration debate, framed by the national problems of slow economic growth and high levels of unemployment, has resulted in the Mexican being seen as an economic and cultural threat. Before we look at the picture of contemporary Mexican immigration to the United States, it would be well to review the history of U.S. immigration and immigration policy. National legislation in this field and its enforcement may reflect our convictions about what kind of future we want for the United States. Until 1875 there was no legislation limiting immigration to the United States. Between the early 1880s and World War I, U.S. immigration policy and various laws implementing that policy had the effect of excluding immigrants who were believed to be inferior to, and less suitable for assimilation than, northern Europeans. In 1907 a congressional commission was appointed to do intensive research and to report on the immigration question. Three years later this (William P.) Dillingham Commission issued a forty-twovolume report; its underlying thesis was the then widely accepted racist viewpoint that all “races” had certain innate, ineradicable traits and therefore that certain ethnic groups provided better citizenship material than others. It also included the idea of country quotas based on the number of immigrants already in the United States. However, Congress took no action on the report.

North from Mexico   297

8. The Quota System When the first World War ended, rising apprehensions about foreigners and xenophobic fears of large numbers of immigrants from war-ravaged Europe caused Congress in December 1920 to pass a new immigration bill introduced by Senator William P. Dillingham, chairman of the Senate Committee on Immigration. Pocket-vetoed by President Woodrow Wilson, the bill was reintroduced in the new Congress during the following April. It was passed quickly by both houses and was signed by the new Republican President, Warren G. Harding. This provisional 1921 Immigration Act established annual quotas for all countries of the Eastern Hemisphere. Three years later Congress enacted a more comprehensive law, the National Origins Act of 1924. It reduced the overall quota and greatly limited the number of admissible southern and eastern European immigrants. In 1927 total immigration was reduced to 150,000, and four years later, in the midst of the Great Depression, a new maximum of 45,000 was set. This 1924 immigration policy, with modifications, governed immigration to the United States until 1968. Since the 1924 immigration act excluded the Western Hemisphere from the quota system it had no direct impact on the movement north from Mexico. Along the southwestern border of the United States there had been no customs stations until 1907 and therefore no restrictions whatsoever on Mexican immigration. The immigration law of 1917, with its literacy qualification and head tax, was the first legislation to seriously affect Mexicans crossing the border into the United States. However, its effect was greatly diluted by World War I policies which exempted thousands of Mexican workers from its provisions. The Border Patrol was established in 1924, but its seventy-five members could hardly patrol the 2,000-mile border effectively. Only in 1929 was undocumented entrance made a criminal act, and even then no legal penalty was attached. During the 1920s thousands of Mexican nationals supplied labor for the postwar boom. In the 1940s World War II labor needs led to the first government-to-government bracero program as well as to increasing undocumented immigration.9

9. Refugee Legislation At the end of the war a high level of general prosperity, the plight of hundreds of thousands of refugees, and a lessening of prejudice against minority peoples created a climate in which American attitudes toward immigrants became less restrictive. There was some turning away the earlier overriding concern about the Nordic makeup of immigration. Presaging the concept of family reunification as the basis for immigration policy, in 1946 Congress passed the War

298  North from Mexico

Brides Act, under which more than 100,000 wives, children, and husbands of members of the Armed Services entered the United States. Then Congress and the country turned toward the larger and more serious problem of refugees. In 1948 the national legislature passed the Displaced Persons Act under which 220,000 were admitted during the next two years. However, these refugees were to be subtracted from their countries’ quotas. President Harry Truman denounced the bill as “flagrantly discriminatory,” because of its exclusions but signed it as the best law he could get. In 1950 a Revised Displaced Persons Act liberalized the earlier law and provided for the admission of 415,000 over a two-year period. Meanwhile, to study immigration issues the U.S. Congress established a committee which came under the control of Senator Patrick McCarran, who was opposed to liberalizing U.S. immigration policy. The result was the 1952 McCarran-Walter Act, the first general reorganization of immigration legislation since 1924. President Truman vetoed the bill, but Congress overrode his veto. The McCarran-Walter legislation continued the ethnically biased quota system and allotted eighty-five percent of the annual quota to northern Europe. Also European colonies were given separate quotas in order to reduce the number of Caribbean blacks coming into the United States as permanent immigrants. However, Section H-2 allowed them (and others) to enter as temporary “guest-workers.” The new law also ended the absolute exclusion of Asians, but gave Japan a quota of only 185 and China, 105; it also provided for the exclusion of foreigners who might subscribe to radical ideas. In 1953 Congress passed a Refugee Relief Act to replace the earlier Displaced Persons acts. It included the first weakening of the quota principle by admitting 205,000 refugees as nonquota immigrants. It was followed by legislation that opened the doors to Hungarian and Indonesian refugees. In 1960 the United Nations passed a resolution declaring World Refugee Year, and in support of this resolution the United States enacted the Fair Share Law which admitted displaced conservative supporters of the overthrown Cuban dictator Fulgencio Batista and some right-wing Chinese. Shortly before his assassination, President John F. Kennedy urged the complete revision of U.S. immigration policy with a view to ending ethnic discrimination and the national quota system. After extensive hearings and careful study, Congress passed, by a large majority, the 1965 Hart-Cellars Immigration and Nationality Act, which went into effect three years later. It was signed by President Lyndon Johnson seated at the base of the Statue of Liberty.

10. Hart-Cellars Act The Hart-Cellars Act ended the country quota system which had given northern Europeans overwhelming preference since 1924, but kept country limits

North from Mexico   299

and a total ceiling, which was now raised to 290,000. For the first time the Western Hemisphere was included in the overall total, with 120,000 places allocated, but no limit was set on an individual country’s immigration. The Eastern Hemisphere received 170,000 places with a maximum of 20,000 per country. For the Eastern Hemisphere the individual country quota principle was replaced by a system of seven preference categories which strongly stressed family reunification. First preference went to adult American citizens’ spouses, minor children, and parents, who were admitted outside the 20,000 limit. The next preferences, limited by the 20,000 per country ceiling, included six categories of relatives of adult U.S. citizens; two further preference categories covered professionals and persons with skills deemed needed in the United States. Finally, the law set aside for Eastern Hemisphere refugees six percent of the total number admissable—10,200. The inclusion of the Western Hemisphere in the total number was seen by many as an disguised effort to limit Mexican immigration—an interpretation based in part on fears expressed by some congressmen of a massive influx of Mexican immigrants. In 1976 Congress passed the Western Hemisphere Act which extended the preference system and the 20,000 per country limit to North and South America.10 The Hart-Cellars Act caused a dramatic change in the composition of immigration to the United States. Before its passage in 1965, immigration was sixty percent European, thirty-six percent Latin American, and three percent Asian. By the second half of the 1980s Latin Americans and Asians accounted for nearly ninety percent of the immigrants, and Europeans, for only about five percent. Mexicans formed the largest single national group. During the decade of the 1970s Mexican immigration averaged 64,229 persons per year. About ninety percent of these legal immigrants qualified for visas because they had relatives in the United States, according to Senator Alan Simpson. In 1980 total legal admissions reached 808,000, the highest number since the beginning of World War I in 1914. According to projections of a Center for Immigration Studies report and a Carnegie Foundation study, immigration for the decade of the eighties will surpass the previous ten-year record of 8.7 million between 1901 and 1910. With the inclusion of approximately 2.9 million undocumented amnesty applicants and hundreds of thousands of refugees, immigration figures for the 1980s are expected to reach 9 million. If other undocumenteds are added, the total could easily surpass 10 million. The 8.7 million immigrants between 1901 and 1910 equaled 11.4 percent of the U.S. population, while the projected immigration of the 1980s would equal only 3.8 percent. On the other hand, current immigration might have greater long-term impact because today the fertility rate of the immigrants is appreciably higher than that of the native U.S. population. During the first half of the 1980s immigrants have made up about one-third of the annual U.S. population increase.11

300  North from Mexico

11. Permanent Visa Mexican Immigration In 1850 the United States had an estimated Mexican-American population of between 87,000 and 118,000; during the half century between 1851 and 1900, visaed Mexican immigration, limited to those arriving by sea, amounted to 13,315 persons. For the following decade, with southwestern border immigrants being included after 1907, it reached 49,642 persons; and during the Mexican Revolution years of 1911 to 1920, the number jumped to 219,004. Between 1921 and 1930, nearly a half million Mexican nationals (459,287) came into the country with documents; but during the following Depression decade only 22,319 entered. In the decade of the 1940s the number rose to 60,589. After 1950 there was a rapid increase, especially after 1954, perhaps because there were fewer undocumented workers available for employment partly as a result of “Operation Wetback.” For the decade of the 1950s the total number of legal Mexican immigrants was 299,811, and in the sixties 453,937 Mexican nationals with visas entered the United States. During the seventies, Mexican immigration reached 642,294. Total legal Mexican immigration from 1900 to 1986 was 2,311,000; the total Mexico-born population in 1980, including undocumenteds counted, was 2,199,221, according to the decennial census. All of these statistics ignore the number of immigrants who returned to Mexico.12 In 1981 legal Mexican immigration to the United States rose to a new one-year high, 101,268. Obviously many of these immigrants came in under the first preference of the Hart-Cellars Act, outside the individual country limit of 20,000. So far in the 1980s Mexican permanent visa immigration has averaged about 66,800 yearly; from 1981 to 1986 inclusive it has amounted to approximately 400,000.

12. Green Carders Permanent visa immigrants, “green carders,” possessors of I-551 cards (I-151 before 1978), are legally and permanently admitted immigrant aliens who may, therefore, live and work anywhere in the United States. To obtain a green card, applicants need a visa from the United States and a Department of Labor certification that they are needed workers and that their employment will not adversely affect American labor. Parents, spouses, and minor children of adult U.S. citizens and of permanent visa immigrants (also political refugees) are exempted from these requirements. However, they may be denied immigrant visas on the basis of presumed likelihood of their becoming public charges. A green card is valid indefinitely unless the possessor is absent from the United States for more than a year, is unemployed for more than six months, or is involved in strike-breaking. The enforcement of these regulations is lax. Green carders are allowed to equate employment with the residency requirement.

North from Mexico   301

For a variety of reasons many Mexican green carders are not interested in living permanently in the United States. Most of these become commuters.

13. Commuters Commuter status has no basis in law, and although the Immigration and Naturalization Service accepts and uses the concept it does not identify commuters in its statistical reports. A commuter is defined as someone who lives in Mexico (or Canada) and crosses the border on a regular basis—daily, weekly, seasonally—to work in the United States. Commuters, therefore, may be permanent visa immigrant green carders, temporary entrant white carders (form I–186), U.S. citizens living in Mexico, or undocumented border crossers. Nobody knows how many commuters there are. A 1983 estimate said approximately 300,000 people living in Juarez across from El Paso had either a white or a green card, and about 3,000 Mexicans cross on a daily basis to work as maids, housekeepers, and so on. At Laredo another 3,000 cross the Rio Grande daily to work as domestics. A 1987 estimate held that there were between 150,000 and 300,000 green card commuters alone. Commuters work in various border region occupations, probably about forty percent in agriculture; the remainder is fairly evenly divided among unskilled labor, hotel and restaurant work, manufacturing and construction, retail sales, and domestic work. In some border areas commuters dominate certain fields of employment; overall they make up about sixteen percent of the U.S. border workforce. Increasingly young Mexican women have entered the commuter population. Efforts in the U.S. Congress during the late sixties to restrict green-card commuting failed in the face of strong opposition. Since then no serious effort has been made to limit commuting. There has been a difference of opinion and debate about the economic impact of commuters. One view holds that the effect is overwhelmingly negative. Commuters, supporters of this interpretation argue, take jobs which would normally be filled by American citizens. Because they are willing to work for lower wages, commuters also tend to force border area U.S. workers, mostly Mexican-Americans, to seek employment elsewhere, even out of the border states. They depress, or at least prevent or delay improvement of, wages and working conditions in those businesses in which they are employed in large numbers, and they make it difficult to organize workers into unions. Lastly, commuters are recruited as (sometimes unwitting) strikebreakers. The opposing view admits the truth or partial truth of these assertions but responds that the overall impact is positive rather than negative. Commuters, the supporters of this view stress, create jobs as well as take them. Besides, they say, a majority of the jobs they take, if not taken by them, would not remain in the United States but would go to Haiti, Hong Kong, or South Korea. Commuters are

302  North from Mexico

described as an indirect subsidy to American business, to offset low-wage foreign competition. Obviously the net impact is debatable.13

14. Undocumenteds Although the problem of commuters is of considerable magnitude, nationally it pales to insignificance as a public issue alongside that of the undocumented immigrants, many of whom are commuters. While no one knows how many undocumented aliens there are in the United States, the best-informed estimates range from three to four million.14 Of these approximately two-thirds are Mexican nationals. They form, therefore, at least fifteen percent of the Mexican-descent population in the United States. Most of these undocumenteds are employed workers or their dependents. They work principally in service industries, light manufacturing, construction, and agriculture—mainly in the states of California, Texas, and Illinois. In recent years the trend has been toward greater dispersal throughout the country. In the postwar period, not until the economic recession of 1953–1954 and the near doubling of unemployment (1.8 million or 2.9 percent in 1953 to 3.5 million or 5.5 percent in 1954), did many Americans and their government begin to view the presence of undocumented Mexican workers as a threat to U.S. labor. The result was “Operation Wetback” in 1954 which considerably reduced levels of undocumented immigration until after the termination of the second bracero program a decade later. Measured by statistics on undocumented Mexicans apprehended by the Border Patrol, there was a rapid rise in illegal entrance during the 1970s. In that decade an average of 721,000 undocumenteds per year were returned to Mexico; in the eighties the figure rose to one million per year. Most experts see the undocumenteds as sojourners rather than immigrants. However, increasingly they find employment in service industries and manufacturing, and each year they leave a residue of permanent Mexican immigrants without documentation, perhaps as many as 150,000. The 1980 census counted 1.1 million Mexican nationals without documents. Beginning in the 1970s undocumenteds became the object of a strong antialien movement spearheaded by the Federation for American Immigration Reform (FAIR). They are perceived by many Americans as not only unfairly taking jobs from U.S. citizens, but also as a financial drain on government social services and as a threat to the linguistic and cultural unity of the country. Because of their illegal status, undocumenteds have to lead a clandestine existence, often accepting below-standard working conditions, housing, and pay. They have come to form a permanent underclass. Many have become victims of “coyotes,” crew leaders, employers, landlords, and government officials. Some have been victimized by unscrupulous legal advisers and by sellers of

North from Mexico   303

counterfeit documents. All have been made the special object of I&NS attention.15 While Mexican undocumenteds form less than sixty-six percent of all illegal entrants, they make up approximately ninety percent of all illegal aliens apprehended by the I&NS. The sixty-six-mile sector of border centering on San Diego is most heavily patrolled, and in 1987 it accounted for forty-four percent of all Mexican border arrests. In late April 1982, with U.S. unemployment nearing ten percent, undocumented workers were targeted as the objectives of selective I&NS raids in nine cities: Dallas, Denver, Houston, Los Angeles, San Francisco, Chicago, Detroit, New York, and Newark. Labeled “Operation Jobs,” the raids were based on the theory that undocumented aliens would be removed from “high-paying jobs” and would be replaced by U.S. citizens who were drawing unemployment benefits. The raids seemed to be aimed predominantly at Mexican workers. Of the 400 aliens arrested in about forty businesses in California’s San Francisco Bay area during the sweep, all but a handful were Spanish-speaking. Reports from some employers indicated that the raiding officers ignored workers who did not appear to be Latino. As a result of a civil lawsuit and amid charges of racism, the I&NS suspended further raids. Spot checks two months after the raids revealed that a large number of the aliens, some now with papers, were back on the job. In the 1970s and 1980s roadblocks, raids in factories and on farms, sweeps in barrios, and arbitrary detentions in shopping centers have increasingly been accepted by the public as necessary to protect U.S. labor and to establish American control of immigration. The Immigration & Naturalization Service has contributed to this acceptance by its publicity which uses emotional terms like invasion, flood, surge, crusade, dangerous erosion of border control, and so on. Again, the object of these searches and sweeps seems to be primarily Mexican nationals. Some have seen these stepped-up internal (rather than border) activities of the I&NS as a reflection of our lessened need for Mexican oil and of Mexico’s weakened world position because of her huge foreign debt. Part of the situation is that the focus of the Border Patrol is on numbers. The apprehension of a daily commuter from across the Rio Grande in Juárez is equivalent to that of a Nicaraguan or a Guatemalan at the end of a 1,000-mile journey, or of an Irish national, or a Filipino. The Border Patrol can apprehend virtually unlimited numbers of illegal commuters, so less effort is directed toward immigrant-smuggling activities. In recent years more than one million undocumenteds have been apprehended annually and returned to Mexico. Most of these have been picked up within forty-eight hours of crossing the border and most had not yet obtained work.16 One of the hotly debated questions about undocumenteds is the nature of their role as workers in the U.S. economy. Do they take jobs that would otherwise go to U.S. citizens? Or do they take jobs that Americans refuse to accept?

304  North from Mexico

Are there large numbers of undocumenteds in certain jobs because the work is low-paying, low-prestige, or hard and dirty? Or are jobs low-paying, lowprestige, and hard and dirty because a surplus of cheap, desperate labor is readily available? The answers are not clear-cut. One thing is certain: undocumented workers are employed in a wide variety of jobs. A 1984 sampling of employed undocumenteds apprehended indicated that fourteen percent earned the then legal minimum wage of $3.35 per hour or less; fifty-eight percent earned between $3.35 and $4.25; seventeen percent earned between $4.25 and $5.25; and eleven percent earned $5.25 or more per hour. How do undocumented workers affect the U.S. economy? Do they pay their own way or are they a tax drain? A 1975 I&NS study would seem to show that they “cost” the United States $13 billion annually. However, a 1985 Rand Corporation study in California concluded that the average undocumented paid more in taxes than he/she received in services, except for education. Focusing on short-term school costs obscures the long-term benefits that accrue to the nation when the children of undocumenteds are educated. A 1985 Urban Institute study concluded that most American concerns about Mexican immigration were generally unfounded. Overall, while undocumenteds may have a negative impact on Mexican-Americans in the local job market, their employment may result in an increase in total U.S. employment. Undocumented workers, like commuters, have been partly responsible for lower wage increases, especially among Mexican-American workers. On the other hand, lower labor costs have possibly saved some marginal industries like textiles for U.S. workers and have resulted in lower prices for consumers. All in all, the broad impact of undocumenteds on the U.S. economy seems to be positive, but the issue is still in dispute.17

15. Simpson-Mazzoli: Simpson-Rodino The sustained search during the 1970s and 1980s for new immigration legislation to deal with the issue of increased numbers of undocumented immigrants coincided with persistent high national levels of unemployment and of nationalistic rhetoric about the United States’ position as a world leader. In this milieu widespread popular perceptions have developed (largely arising out of public statements by I&NS publicists and others) that permanent illegal immigration has been skyrocketing, that the United States has lost control of its borders and of immigration, and, moreover, that the situation has reached crisis proportions. As a result, there has been great pressure on Congress to pass legislation that would provide a solution to the perceived problems. During the late 1960s, Congressman Peter Rodino introduced the undocumented issue in the House of Representatives, but his efforts to reduce illegal

North from Mexico   305

entrance were blocked by southern and southwestern agricultural interests in Congress. In 1972 Rodino introduced a major immigration reform bill which included penalties for employers who hired undocumented workers. It passed the House in mid-September; however, it failed in the Senate. During the following January, Rodino brought up a similar bill, which again was not acted on by the Senate after passing in the House by a vote of 297 to 63. In mid-1974 Senator Edward Kennedy introduced a very similar bill, which however included a provision for amnesty for undocumented workers. The Senate failed to act on the bill. During the following year three immigration reform bills were brought forward by Rodino, Kennedy, and Senator James Eastland, but none reached a vote. In December 1976 a Gerald Ford administration cabinet-level study recommended reforms which included employer sanctions, limited amnesty, and increased funding for the I&NS. Three-and-a-half years later an amendment to the Immigration and Nationality Efficiency Act that would have provided penalties for hiring illegals failed to pass. During the first years of the Carter administration, Rodino’s bill was again introduced as was a similar administration bill. Neither was acted on. By the end of the decade eleven states and Puerto Rico had passed laws prohibiting the hiring of undocumenteds, and Congress’s Farm Labor Contractor’s Registration Act provided for a $10,000 fine and three-year jail term for similar offenses. During 1978 the U.S. Congress created a sixteen-man Select Commission on Immigration and Refugee Policy and held extensive hearings in a dozen cities. In March 1981 the Commission released its final report. The recommendations of this report were embodied in the 1982 SimpsonMazzoli Immigration Reform and Control Act. Its main features included: employer penalties, amnesty and ultimate legalization for certain undocumenteds, expansion of the Border Patrol, liberalizing H-2 worker regulations, and an absolute annual limit on immigration. The bill passed the Senate in August by a vote of eighty to nineteen, but then died at the end of the year in the House with some 300 amendments pending. In February 1983 a different version of the bill was reintroduced and three months later passed the Senate, 76 to 14. The House failed to consider the bill and again it died at the end of the session in December. It was rescheduled in January 1984 and after a great deal of emotional debate and various amendments was passed on June 20, 1984, by the House of Representatives, by a narrow vote of 216 to 211. However, in 1984 Congress failed to enact the immigration reform bill into law because it was unable to reconcile the House and Senate versions. So for a third year straight immigration reform legislation failed to pass. As earlier, the principal stumbling block was a temporary-worker policy, with opponents arguing that history indicates that temporary-worker programs tend to foster illegal immigration.

306  North from Mexico

Among the other lesser, but controversial issues in the proposed legislation was a suggested national identity card, which aroused angry objections from Chicanos and civil libertarians who protested that it would lead to increased discrimination. Other Mexican-American concerns were that the proposed legislation would cause greater job competition, lead to increased exploitation of undocumented workers, exacerbate racist feelings, and possibly arouse a backlash of discrimination. Additionally, representatives of various groups opposed to the bill predicted that it would do little to reduce illegal immigration or to enable the United States to “regain control” of its borders, as its supporters maintained. During the summer of 1985 Peter Rodino introduced another version of his bill in the House. Meanwhile the Senate debated and on September 19, 1985, passed the Simpson-Mazzoli bill, which included a controversial threeyear, 350,000 annual guest-worker provision as well as amnesty for undocumented workers who had resided in the United States since January 1982. Again it appeared that no compromise could or would be reached. During the first half of 1986 House Judiciary Committee discussions looked for compromises to solve the temporary-worker impasse. In June 1986 Congressman Charles Shumer introduced the concept of a second amnesty category, undocumenteds who had worked twenty days in perishable agriculture between May 1, 1985, and May 1, 1986. His amendment also provided that those who qualified under this agricultural amnesty and later left farm work might be “replenished” beginning October 1, 1989, by bringing into the country replacements who could then apply for permanent residence. By increasing the agricultural work experience requirement from twenty to ninety days and by clever parliamentary procedure, Shumer got the bill passed in the House on October 9, 230 to 166. The inclusion of a provision for a special, relatively easy amnesty program for agricultural workers in the Shumer Amendment had finally broken the deadlock resulting from agribusiness’s insistence on a guest-worker clause and revived a bill that all had considered dead two weeks before. On October 14 a House-Senate committee was able to hammer out a compromise bill. The next day the House voted to accept the new version, 238 to 173, and two days later the Senate followed suit, 63 to 24. On November 6, 1986, President Ronald Reagan signed the Simpson-Rodino Immigration Reform and Control Act (IR&CA). During the final debate there was strong feeling that the IR&CA was a “bad” law, but the sentiment that “something had to be done” won out, and it passed with the deletion of the guest-worker provision. The Simpson-Rodino law is based on the idea that heavy undocumented Mexican immigration to the United States is almost entirely the result of economic pressures and job-seeking. This assumption appears less than absolutely certain in view of two facts. First, urban migration within Mexico would seem

North from Mexico   307

to indicate that unavailability of standard jobs and even scarcity of substandard ones is not a sufficiently strong deterrent to dissuade the determined migrant. Second, ninety percent of current and recent legal immigration from Mexico has been under the family reunification provisions of the 1965 HartCellars law and there is a backlog of approximately 170,000 persons awaiting visas. The frustrated hope of family reunification is an important cause of undocumented immigration, and the IR&CA does not consider that problem. The law has three principal provisions: two categories of amnesty, employer sanctions, and beefed-up Border Patrol. Of these three, amnesty was, for Mexican-Americans, the most important.18 The concept of amnesty in the Simpson-Rodino bill was not a completely new idea. The Registry Act of 1929 provided for the naturalization of aliens who had been living in the United States since June 3, 1921. A quarter of a century later the McCarran-Walter Act of 1952 included the provision that persons with continuous U.S. residence since June 30, 1924, could legalize their position without leaving the country, and five years later an amendment moved the residency date up to June 30, 1940. As finally passed, the IR&CA law listed two ways that aliens might qualify for amnesty: provable residence in the United States since January 1, 1982, or provable employment in farm work for ninety days between May 1, 1985, and May 1, 1986. Applicants for amnesty had to pay a $185 fee ($420 for a family) plus the costs of a medical examination and other smaller expenses. According to the best estimates, the average applicant had to spend about $300 on his or her application. The complete process consisted of three stages to be passed: application, resulting in temporary legal status; application, leading to permanent legal status; and, finally, citizenship. The Immigration and Naturalization Service began accepting applications for legalization on “Cinco de Mayo,” May 5, 1987. Although officials spoke of expecting a “flood” or a “crush” of applicants, six months later only about 850,000 had filed. Meanwhile the law and considerable confusion as to its interpretation by the I&NS created uncertainty and fear in the minds of undocumenteds in the United States as well as among those in Mexico who were thinking of entering the country. Initially a decline was noted in border area arrests. No flood of undocumenteds returning to Mexico developed, as some had predicted. In fact, apparently many sojourner undocumenteds had remained in the United States rather than risk being barred from reentry. The May 4, 1988, deadline for the first category of eligible undocumenteds in the program showed approximately 1.4 million applicants under the general provision and 480,000 under the special provision for workers in agriculture whose deadline was November 30 of the same year. The I&NS reported that between eighty-five and ninety-five percent of eligible undocumenteds had applied. In late June thousands of Mexican nationals applied at the border for

308  North from Mexico

entry in order to claim temporary working visas under the agricultural worker provision, alleging that they qualified. They were allowed to enter on the basis of their word, the Border Patrol announced. By November 30, 1988, a total of nearly one million farm workers had filed, half of them in California. The total number of aliens who had applied for amnesty by January 1989 was about 2.96 million (1.789 million and 1.168 million), of whom approximately seventy percent, or 2 million were Mexican.

16. Results of the Simpson-Rodino Law Two years after the passage of the IR&CA, what can be said about its impact, its effectiveness in controlling undocumented immigration? At first it seemed to result in a decline in illegal entry. Both the I&NS and President Reagan announced dramatic decreases in the number of undocumenteds apprehended. However, in addition to the new law, other factors contributed to this drop: the removal from the border of I&NS personnel to operate the amnesty program, the rising of the Rio Grande to its highest levels in forty years, and the continuing decline in illegals’ funds due to the peso devaluation. By June 1987 there began a noticeable upswing in undocumented entrance, which continued during the rest of the year. By May 1988 the numbers were back up to pre-IR&CA levels. There is general agreement that the new law has not stopped the undocumented “flood” as its supporters had promised. The February 1989 U.S. proposal to build a four-mile “Berlin Wall” ditch at Otay Mesa near San Diego seems a tacit admission that the Simpson-Rodino law has failed to regain control of the border. The law has made it more difficult and expensive to become an undocumented worker and therefore has led to greater selectivity. On the other hand, because of Simpson-Rodino, more illegal workers now tend to bring their families and to remain in the United States. There seems to be little reduction in employer demand for their labor, despite sanctions. A 1988 Dallas Federal Reserve Bank study estimated that sanctions would reduce the hiring of undocumenteds by only fifteen to twenty-five percent. It seems that Mexicans will continue to come without documentation, perhaps not in the earlier volume, but they will come. The Simpson-Rodino legislation has led to numerous problems, some anticipated, some not. There clearly has been a rapid increase in the use of fraudulent documents, both by illegals and employers, and much more is anticipated as undocumenteds try to prove the length of their U.S. residence. The law also seems to be affecting legal workers negatively. There has been a rapid rise in reports of employer discrimination against U.S. citizens and green carders on the basis of their presumed ethnicity. Complaints have been numerous and the General Accounting Office announced in mid-1988 that over a half million

North from Mexico   309

employers had initiated policies that resulted in discrimination. Employers, who are required by law to ascertain the legal status of new employees, were assured by the I&NS that they would not be held responsible for the authenticity of documentation proffered by job applicants. The I&NS may thereby ensure greater ostensible compliance with the law, but also may further impair its effectiveness. Another problem was the question of spouses and children who do not qualify for amnesty, although one spouse does. Initially the I&NS created great outrage and anger by announcing that they might be “required to depart,” but that deportation might be delayed for “compelling humanitarian reasons.” However, in May 1988 the I&NS changed its position, saying spouses and children would be given special visas to avoid breaking up families.19 Obviously it will be years before the full impact of the Simpson-Rodino law can be known or measured. There are a number of questions still to be answered. What will be the impact of stage two with its tests? How many undocumenteds will be able to move from temporary to permanent legal status and ultimately to citizenship? What will happen to those who are unable to complete the qualification process? Will they become a permanent underclass? What will be the final costs, financial and economic, to the United States? To the undocumenteds? Will amnesty lead to better working conditions and pay for the legalized undocumented workers? Will the legalized workers remain in low-pay, low-prestige employment or move to better jobs that their legal status may make available to them?

20

A Demographic Profile of the Mexican-Born Population in the United States Alma M. García

Crossing a 2,000-mile border with the United States, Mexican immigrants have always found themselves in a situation unique to all other immigrant groups.1 Since this border is one of the most frequently crossed in the world, Mexicans have always shaped the fabric of the United States. For Mexican immigrants, life in “El Norte” represents both cultural change and cultural continuity.2 This interaction between immigrants and their host country has not been without social and cultural upheaval, one that has long challenged the national metaphor for cultural relations—the American melting pot. Since the arrival of first major wave of Mexican immigration in the early twentieth century, this immigrant population has settled in such cities as El Paso and San Antonio, Texas, and Los Angeles, San Diego, and San José, California.3 These cities and others maintain vibrant, flourishing communities of Mexican immigrants who live in close proximity, often as next-door neighbors. These communities are the site of both cultural replenishment as well as social, political, and economic discord between Mexican immigrants and the U.S.native born.4 Still, a small but growing sector of foreign-born Mexicans and U.S.-born Mexican immigrants have surpassed major obstacles as they search for a better life for the children they brought with them during their precarious journey to El Norte. Oscar Handlin, the noted U.S. historian, provided an overview of immigrants. In his landmark book The Uprooted: The Epic Story of the Great Migrations That Made the American People, Handlin stated that “once I set out to write a history of the immigrants in American. Then I [Handlin] discovered that the immigrants were American history.”5 Borrowing

312  North from Mexico

Handlin’s assertion, Mexican immigrants were and will always be part of American history. A demographic profile of Mexican immigrants residing in the United States is essential to understanding the twenty-first century’s political, economic, social, and cultural issues currently facing future generations of Mexican immigrants and later generations of U.S.-born Mexicans in the United States. Educators, politicians, and social-service providers cannot adequately design public policies for the problems facing Mexican immigrants, both authorized and unauthorized, without understanding key demographic features of this population.6 Beginning in the early 1960s, the foreign-born population living in the United States experienced marked changes in their countries of origin, settlement patterns, population size, age, and other demographic characteristics. As the European-origin immigrant population in the United States declined, the U.S. foreign-born population, documented or not documented from Latin America increased. Under the provisions of the Immigration and Nationality Act of 1965, a preferential system replaced the existing quota system for countries in Latin America and Asia, regions that each received significantly larger allotments than European countries. Although many hoped that this policy would curb undocumented immigration, it did not due to economic stagnation in Mexico, the difficult bureaucratic process required under the Immigration and Nationality Act of 1965, and the pull from U.S. manufacturing and service industries for a pool of cheap labor.7 This “new wave” of immigrants transformed the fabric of society in the United States by contributing to racial and ethnic diversity in long-standing communities of immigrants and in new geographic areas. Latin American and Asian immigrant communities were replenished with recently arrived immigrants from their home country. New immigrant communities emerged as immigrants from new sending countries began to arrive in the United States. Both types of communities experienced cycles of acceptance and rejection, ones that have historically shaped all aspects of their lives and that of the U.S. population. The impact of a growing foreign-born population raises the following questions. First, will the government recognize the specific needs of a growing population of both authorized and unauthorized immigrants? If so, what will be the nature and scope of the political debates surrounding the allocation of funds to meet these needs? How will the government reassess the efficacy of existing government agencies to adapt to communities of immigrants who are coming from different countries of origins in comparison to historically longstanding ones such as Mexican immigrants? What will be the outcome of restrictive immigrant legislation, such as those regarding the use of public services by immigrants and, more recently, the resident status of citizenship

A Demographic Profile of the Mexican-Born Population    313

paths for young children brought to the United States by their unauthorized parents? What decisions will be made to address the educational needs of unauthorized children who accompanied their unauthorized parents or guardians? What rulings will be made on immigration cases by the different U.S. Courts of Appeals and the U.S. Supreme Court? What policies will be developed to address the specific needs of immigrant women? Lastly, what role will naturalized Mexican immigrants and second-generation and later-generation U.S.-born Mexicans play in politics at every level and every type, but perhaps most importantly, in presidential electoral politics? These critical questions, among others, and subsequent policy decisions will continue to fuel the political and public debate on Mexican immigration. The Mexican-born population in the United States has increased exponentially over the past several decades, surpassing every other immigrant group living in the United States. Many factors contributed to the growing influx of Mexicans crossing the border into the United States. Mexico’s economic recessions in the 1980s and 1990s led the Mexican federal government to adopt vastly unpopular austerity measures that greatly diminished the standard of living for the middle class, the working class, and those in the informal labor sector, such as street vendors and domestic workers, most of whom were women. As a result of significant cuts in key social services used by those Mexicans most at risk, rising costs of all consumer items and food, and diminishing wages, Mexicans decide to make the journey to El Norte in their attempt to start a new and better life. From 1970 to 2000, immigrants from Mexico surpassed the combined total of immigrants from Central America. Many factors account for such a pattern beginning with the historical and contemporary ties between Mexico and the United States that pulled Mexicans to the United States. Long-standing communities of Mexican immigrants facilitated the migration of Mexicans to unite with family members, not only those from their nuclear families but also extended family, friends, and even friends of friends. In 2010, immigrants made up 25 percent of the total population of California, 11 percent of New York, 10 percent of Texas, and 9 percent of Florida. Together these four states accounted for 59 percent of the total foreign born. One in four residents in California were immigrants, primarily Mexican immigrants. In New York and New Jersey, one in five residents were foreign born. Except for Illinois, with 14 percent of its population being foreign born, other states in the Midwest had lower than 8 percent foreign born, including North and South Dakota with only 3 percent. Immigrants from Cuba and the Dominican Republic were concentrated in Florida, Latin and Central Americans in California, with Mexican immigrants forming the majority in this state. Mexican immigrants have always been concentrated in a few regions and states in the United States. In 2011, 58 percent of all Mexican immigrants lived

314  North from Mexico

in California and Texas. Four million lived in California, representing 37 percent of all Mexican immigrants, and 2.5 million in Texas (21 percent of all Mexican immigrants). Illinois and Arizona rank the next highest. In 2010, about 25 percent of the Mexican born clustered in three cities: Los Angeles, Houston, and Chicago. In 2011, Los Angeles and its greater metropolitan area were home to 15 percent (2 million) of the Mexican-born in the United States. Chicago ranked second with 6 percent (684,000) and Dallas ranked third with 5 percent (610,000). Historically, Mexican immigrants have concentrated in several metropolitan areas. Between 2008 and 2012, the top five metropolitan areas with the largest percentage of Mexican immigrants were McAllenEdinburg-Mission, Texas (27 percent); Los Angeles-Long Beach-Santa Ana, California (14 percent); Riverside-San Bernardino-Ontario, California (13 percent); San Diego-Carlsbad-San Marcos, California (11 percent), and Houston-Sugar Land-Baytown, Texas (10 percent).8

1. The Nexus of Education, Occupation, and Poverty An overview of education, occupation, income, and poverty illustrates the structural inequalities confronted by Mexican immigrants. The United States has always prided itself on its public school system, one developed to be free and accessible. Nevertheless, equal access to quality education remains a pressing social problem, one that results in limited upward mobility. The educational attainment level of parents remains one of the most significant predictors of the educational attainment levels of their children. Education is a major avenue to upward mobility, a foundation for surpassing a parent’s level of socioeconomic standing. Too often, it represents a long-lasting roadblock in the lives of Mexican immigrants and their children. Studies have shown that children of immigrants who are first-generation college students begin their education at a disadvantage stemming from the social-structural inequalities and limited social capital that parents without a college education possess.9 Daughters of Mexican immigrants are at a greater risk due to long-standing cultural traditions that constrain a woman from breaking expected gender roles.10 Nevertheless, institutional structural inequalities, regardless of levels of a student’s social capital, continue to reinforce educational inequalities. Such inequalities include a lack of sufficient funding for schools in cities with large immigrant communities; a lack of successful mentoring for students at all levels, particularly those wanting to attend college; a lack of sustained mentoring programs; limited access to computers; and inadequate efforts to provide school counselors who can communicate effectively with immigrant parents many of whom lack the English skills needed to understand the educational system and the needs of their children as they progress through schools. Specific programs developed to facilitate the success of immigrant children

A Demographic Profile of the Mexican-Born Population    315

and, in many cases, the U.S.-born children of immigrants have produced only limited results.11 Although educational attainment levels for some immigrants, such as some Asian immigrants, are just as high or, in some cases, even higher than those of the native born, Mexican immigrants continue to lag behind at precariously low levels of educational attainment.12 For decades, a high school diploma has no longer been a springboard to the middle class as measured by income and occupation. Based on the U.S. Census of 2010, 57 percent of all Mexican immigrants age 25 years and older had not graduated from high school or received a GED diploma in contrast to the total immigrant population (28 percent) and the total native-born population (7 percent). In 2013, only 6 percent of Mexican immigrants had a bachelor’s degree or higher, significantly lower than the rates for all immigrants (29 percent) and the total native-born population (33 percent).13 Educational attainment levels have a direct impact on the labor force experiences of Mexican immigrants. Of all Mexican immigrants in the paid labor force, 25 percent were employed in natural resources, construction, and maintenance occupations in comparison with only 13 percent of all immigrants and only 8 percent of the native-born population.14 At the service section occupational level, Mexican immigrants formed the largest group (31 percent), followed by all immigrants (25 percent) and then the native born (17 percent). Mexican women were concentrated in service and personal care jobs. A comparison among Mexican immigrants, all immigrants, and the native born reveals even greater differences in the category of “management, business, science and arts. Percentage differences in this occupational category illustrate the disparity among the three groups. Thirty-eight percent of all native born and 30 percent of all immigrants cluster in this category. Only 9 percent of all Mexican immigrants cluster in this category. In addition, the socioeconomic conditions within the country of birth for foreign-born men and women are closely tied to the placement of foreigners in the U.S. labor markets. Immigrants from India, China, and Korea have arrived to the United States with better economic and educational backgrounds than Mexican immigrants who, with some exceptions, have come from lower economic backgrounds.15 Language barriers also contribute to limited occupational mobility and overall social mobility. Unlike Indian immigrants, who have high levels of English proficiency, less than one-third of all Mexican immigrants demonstrate strong English language skills. In 2013, 69 percent of Mexican immigrants (age 5 and older) had limited English proficiency, compared with 50 percent of the total foreign-born population. A difference of 12 percent exists between Mexican immigrants who spoke only English at home (4 percent) and all immigrants (16 percent). In, 2013, levels of English proficiency among Mexican immigrants increased but were still lower than that of the total

316  North from Mexico

foreign-population in the United States. As with other immigrant groups, children of immigrants demonstrate higher levels of English proficiency. About two-thirds of Mexicans born in the United States speak English proficiently. The greater a generation’s distance from their Mexican immigrant background, the higher the percentage of English proficiency. Without an understanding of the limited English skills of a significant sector of the Mexican immigrant population in the United States, particularly newly arrived ones, attempts by policy makers to develop social services to address their needs will remain stalled.16 At the political level, the history of presidential political campaigns reveals attempts by such candidates as John F. Kennedy, Lyndon B. Johnson, George H. W. Bush, and George W. Bush to use their often-limited Spanish language skills to “woo” potential voters. Obama’s campaign organizers provided television advertisements and social media outlets to reach naturalized Mexican immigrants and their U.S.-born children who were eligible to vote. In an effort to reach monolingual Spanish-speaking customers, businesses regularly run advertisements in Spanish-language newspapers and buy television time for Spanish-language commercials. In addition, some drugstore chains provide prescription guidelines in several languages, depending on their customers, and, more recently, have translators available on call. The classic example of a potentially disastrous problem with the labels on prescription bottles involves the direction to “take once a day.” The problem is that “once” is also Spanish for eleven (once). Perhaps this remains an urban myth, but it carries cautionarytale warning of the dangers of language barriers. An examination of the poverty rates for the Mexican immigrant population requires a brief historical grounding to bring this issue from out of the shadows of political discourse. In 1960, on Thanksgiving Day, CBS aired A Harvest of Shame, one of Edward R. Murrow’s finest documentaries in which he depicted the “shameful” plight of migrant workers whose poverty was far reaching, but mostly invisible to a white, middle-class society experiencing the post–World War II affluence. Two years later, in 1962, Michael Harrington published his now classic The Other America, a scathing indictment of a society mostly oblivious to the pernicious conditions of poverty experienced by the “other Americans,” like the migrant workers depicted in A Harvest of Shame. In 1964, in his State of the Union address, President Lyndon B. Johnson declared a “War on Poverty” stating, “Our aim is not only to relieve the symptoms of poverty, but to cure it and, above all, to prevent it.”17 The civil rights movement and, specifically, the NAACP and the Urban League, put mass pressure on the Johnson administration to implement policies to set a national agenda for combating poverty. Although critics and supporters of the War on Poverty continue to debate its efficacy, poverty, however defined by the government and related agencies, remains a key indicator to assess a group’s well being in any given society.

A Demographic Profile of the Mexican-Born Population    317

Beginning in the early 1970s, statistics from the U.S. Census Bureau showed some decreases in poverty levels, but the trend was not long lasting. Certain groups, such as single African-American women with children, Native American families, and the elderly from all backgrounds and immigrant statuses, continue to battle poverty and its consequences. Similarly, those coming “north from Mexico” experience poverty levels significantly higher than that of the total U.S. society. In 2011, Mexican immigrants were more likely than the native or foreign born to live in poverty. Based on the official poverty line, 30 percent of Mexican immigrants lived in poverty in comparison to the total native population (13 percent). In 2013, poverty levels for the total native-born families (10 percent) and the total foreign-born families (18 percent) were lower than that for Mexican immigrant families (28 percent). When poverty is redefined to include those “in or near poverty.” 63 percent of Mexican immigrants are considered to live in poverty.18 Additional demographic patterns of the foreign-born population in the United States provide further insights into the contemporary experiences of Mexican immigrants and their impact on U.S. society. Levels of poverty in the United States have shifted over the years for some groups, such as AfricanAmericans. In 2012, while the poverty rate among African-Americans decreased, their poverty rate of 27 percent was more than double the rates for whites (13 percent). A larger percentage of immigrants (23 percent) lived in poverty compared with the native-born population (14 percent). Different rates of poverty exist across immigrant groups. Mexicans had the highest level of poverty (35 percent), followed by Hondurans (34 percent), and Guatemalans (31 percent). Whether they fall below the poverty level or not, accessibility of health insurance coverage represents another critical indicator of an immigrant group’s standard of living. In 2010, 54 percent of Mexican immigrants did not have health insurance, in stark contrast to rates for the total native-born population (14 percent). Many uninsured immigrants and others often used a hospital emergency department for routine care, particularly for their children’s medical care. Poverty and lack of medical insurance have a direct impact on an immigrant group’s use of the welfare system. Immigrants surpassed the native born in use of welfare programs by 13 percent. Educational attainment levels represent a major determinant in a person’s rise out of poverty, increased accessibility to health care and health insurance, and decreased need to use welfare services. For many immigrants, length of stay in the United States does not lead to upward mobility. Immigrants who have lived in the United States for a decade or more may still experience limited economic improvements. Economic gaps continue to exist between Mexican immigrants and the total U.S. population; immigrants have higher poverty rates, higher utilization of welfare services, and less access to health care.19

318  North from Mexico

Mexican immigrants have specific demographic characteristics that distinguish them from other foreign-born and native-born people. For one, Mexican immigrants have lower educational attainment levels. For example, in 2013, only 6 percent had received a B.A. degree or higher, in contrast to rates for the total foreign-born population (28 percent) and the native-born population (30 percent). Mexican immigrants are also younger than the native-born population. They are less proficient in English but are more likely to speak English at home than the total foreign-born population. Mexican immigrants cluster in certain occupations, such as service, construction and maintenance, transportation, and farm work. Mexicans have lower incomes and higher poverty rates.

2. New Destination States: Emergent Mexican Immigrant Communities Immigrants have traditionally settled in communities with immigrants from their own country. In the 1970s, California, New York, Florida, Texas, New Jersey, Illinois, and Massachusetts have become the home for about 60 percent of all immigrants and, in some years, 75 percent. Beginning in the early 1990s, immigrants moved to new-destination states with historically low numbers of immigrants from any part of the world. Between 1990 and 2000, the foreignborn population increased by 11 percent. The geographic settlement patterns for this increased population show that the number of immigrants doubled in states in the South and Midwest, two regions with historically low numbers of immigrants. South Carolina, Alabama, Tennessee, Delaware, Arkansas, South Dakota, Nevada, Georgia, Kentucky, North Carolina, Wyoming, Idaho, Indiana, and Mississippi experienced a 50 percent increase in immigrant population. A dramatic change took place from traditional immigrant areas of settlement, such as California, Texas, and New York. About one in every eleven immigrants in the United States resided in a new-destination state, compared with one in twenty-five in 1990. While the growth in the foreign-born population in the traditional settlement areas grew 40 percent between 2000 and 2010, the unauthorized immigrant population in new-destination states grew by 80 percent. Between the years 2000 and 2010, the new-destination states of Alabama, Georgia, Louisiana, Mississippi, South Carolina, and Tennessee witnessed a large increase in their foreign-born population, particularly Mexican-born immigrants. Many of these immigrants migrated to these states after living in another one, such as California. Many left once they heard of new employment opportunities and as living costs increased in California. The immigrant population increased by 77 percent in South Carolina, and Alabama and Tennessee experienced a 67 percent growth. Of the total number of immigrants in newdestination states, 35 percent were from Mexico. Immigrants from other

A Demographic Profile of the Mexican-Born Population    319

countries of origin in new-destination states included India, the Philippines, and China, but none of these countries had an immigrant population larger than 4 percent. The unauthorized immigrant population in new-destination states grew by 80 percent between 2000 and 2010, while in traditional-destination states it increased by 40 percent. In both traditional and new-destination states, Mexicans make up the largest group within the unauthorized.20 Immigrants in new-destination states differ in some key demographic characteristics from those in traditional destinations. In contrast to immigrants in traditional-destination states, immigrants living in new-destination states are more likely to be men of working age who participate in the paid labor force; are less likely to have health insurance; live in poverty; and are less likely to hold management, business, and finance occupations. In traditionaldestination states, more women (6 percent) had jobs in the health care system than those in new-destination states (3 percent). Immigrants in traditional and new-destination states showed similar patterns of lower educational attainment levels compared with the total native population.21 In addition to changes among immigrants in general, demographers predict continued shifts in the geographic distribution of the Mexican born. Changing manufacturing, service, and agricultural food production industries will continue to require a cheap and steady source of low-skilled or unskilled workers. Mexican immigrants, specifically undocumented ones, fill such needs. For example, Vermont’s milk industry, the state’s major agricultural food production industry, is in constant need of workers. As increasing numbers of U.S. workers moved out of the state looking for better employment opportunities, the milk and milk by-products industry looked south to Mexico in search of a replacement labor force. Eventually, Mexican immigrants, usually undocumented, made the trip north from Mexico to Vermont in search of more lucrative jobs than those in the West and Southwest. In 2013, Mexicans represented about one-third of the dairy workers. Interestingly, unlike the growing opposition in Arizona, the National Milk Producers Federation expressed concern over the widespread deportation of unauthorized Mexican migrant workers without whom the dairy industry would face drastic economic declines. The increase in the Mexican immigrant population has not only changed the composition of Vermont’s labor force but has also affected the university curriculum. For example, at Middlebury College, students can take a new course on the U.S.-Mexico border. One year, as part of their coursework, students enrolled in this course decided to address the needs of migrant workers by creating a website containing resource materials for Vermont’s migrant workers.22 In addition, communities with a large Mexican immigrant labor force soon witnessed the establishment of new businesses, usually momand-pop business, that catered to the Mexican immigrant population. New

320  North from Mexico

issues will inevitably develop as Mexican immigrants have U.S.-born children who will enter public schools and be eligible for social services. Some communities in Vermont that have experienced an influx of Mexican immigrants have experienced incidents of ethnic violence. The town of Burlington’s Peace and Justice Center responded to the killing of Michael Brown—the unarmed black teenager shot and killed in Ferguson, Missouri, by the police— by calling for the need to fight racism and racial violence outside the South, such as in the state of Vermont where incidents of violence against Mexican workers have increased.23 Just as in Vermont, Mexican immigrants in Tennessee serve as a source of cheap labor and represent a major factor in the state’s economic growth. Workers are overwhelmingly young Mexican men who have migrated to Tennessee from other states, such as California, Florida, and Texas, instead of directly from Mexico. Using informal networks, particularly immigrant family ones, Mexicans have been moving to Tennessee, one of several new-destination states, with hopes of more stable employment. Mexican-born workers cluster in the urban centers of Nashville, Memphis, Knoxville, and Chattanooga, but trends show some movement into rural areas, where the demand for farm work is increasing. Other occupations include low-wage, low-skill, seasonal employment in construction and many year-round positions in services and manufacturing. Vermont, Tennessee, and other states without a history of Mexican immigrants, face social, political, and educational issues as immigrant workers arrive and stay in these areas of the United States. The legacy and ongoing struggles with anti-immigration prejudice, discrimination, and violence against Mexican workers in states such as California, Texas, and Arizona will no doubt face states like Vermont and Tennessee.24

3. Finding Religious Space in the New South Beginning in the 1980s, as in Tennessee, Vermont, and other new-destination states, more opportunities unfolded in the South for Mexican immigrants, and they became a driving force for social transformation. Social, cultural, economic, and political adaptation and conflicts eventually developed. On the one hand, Mexican immigrants experienced the process of adjustment to the United States, specifically within new-destination states. Unlike the experiences of Mexican immigrants in traditional-destination states, those who came to new-destination states lacked preexisting social network. On the other hand, whites and African-Americans in new-destination states did not have sustained interactions with Latino immigrants, making it necessary for them to adjust to a rapidly diversifying society.25 Mexican and Central American immigrants journeyed to the South, specifically to the states of Alabama, Arkansas, Georgia, Kentucky, Louisiana, Mississippi,

A Demographic Profile of the Mexican-Born Population    321

North Carolina, South Carolina, Tennessee, and Virginia, in hopes of improving their family’s lives. Several international factors contributed to the arrival of Mexican immigrants to the South. Beginning in the 1980s, economic transformations in the South set the stage for this new migration of Mexican immigrants. As a result of an increasingly globalized economy, the South found itself facing the ill effects of U.S. companies moving offshore, which inevitably resulted in plant closings in various industries, such as steel and textiles. At the same time, some national and foreign companies recognized the profitability of moving their production to the South, where they could take advantage of cheap, primarily nonunion labor. Mercedes, Honda, and Hyundai set up factories in Alabama, North Carolina, South Carolina, and Tennessee. Poultry processing plants in Georgia, Alabama, Arkansas, and North Carolina accounted for almost half of the entire industry in the United States. These developments created an expanding need for cheap labor that was filled by Mexican immigrants. Facing a shortage of cheap labor, food-processing, carpet, service, and construction industries adopted a variety of recruiting methods that eventually created a labor force of Mexican immigrants. Mexican and other immigrant workers built facilities for 1996 Olympic Games held in Atlanta. As a result, “the mass migration of Latin Americans [immigrants] to the U.S. South . . . triggered an unprecedented series of changes in the social, economic and cultural life of the region and inaugurated a new era in Southern history.”26 Mexican immigrants began arriving in Georgia during the 1970s when Gainesville became known as the “Poultry Capital of the World.” Mexicans also found employment in meat production and large-scale agribusiness. During the 1980s, growers preferred to hire Mexican immigrant workers, particularly newly arrived ones, many of whom were undocumented and accepted even lower wages. Mexicans made up at least 34 percent of workers in the construction industry, but there are many undocumented Mexicans in this industry, so this is a conservative estimate.27 The state also attracted Mexican immigrants to work in the carpets and rugs industry, both of which represented Georgia’s growing dominance in both the national and world economy. As early as the 1950s, textile and carpet manufacturers began to move to the South, an area with lower wages than the Northeast. Mexican workers moved into this industry, which was rapidly overtaking the poultry industry and other areas in the carpet industry. The 1990s witnessed the largest migration of Mexicans to Georgia. For example, in Dalton, Georgia, from 1990 to 2000, the Hispanic population, most of whom were Mexican immigrants, increased from 7 percent to 40 percent.28 Mexican immigrants, with their Americanborn children, became a visible part of society in northern Georgia, and Atlanta became the major hub for Mexican and Central American workers. The city of Atlanta witnessed more than mere economic growth. Social change

322  North from Mexico

and social conflict eventually developed between newly arrived Mexican immigrants and the long-established groups in Atlanta: Anglos and African Americans.29 Throughout the new-destination states, Mexican immigrants were changing the social fabric of communities. Many changes came smoothly, while others led to social conflict. The struggles of the Catholic Mexican and Central American immigrant community in Atlanta, Georgia, stands as a case study in this process. In the introduction of her study of Mexican and Central Americans in Atlanta, Mary Odem draws a vibrant picture of the community, one that struggled and won the battle to preserve its Latino Catholic roots: On New Peachtree Road in a former industrial zone located on the northern outskirts of the city of Atlanta sits a small, pre-fabricated warehouse. . . . A tall slender white cross with the words “Misíon Católica de Nuestra Señora de Las Americas” (Catholic Mission of Our Lady of the Americas) hangs from the roof of the building next to a glass-enclosed space with a 3 foot high statue of the brown-skinned Virgin of Guadalupe (La Virgen Morena) . . . an unusual site in the predominantly Protestant region. On the weekend, more than 3,000 Mexican, Central American and South American immigrants gather at this site . . . to celebrate mass; pray to patron saints; attend English job-training and computer classes and take part in family and youth programs. They come by car, subway, and on foot from surrounding neighborhoods and from nearby towns and suburbs.30

This vivid description of the Misión depicts a sacred space created out of a struggle between Mexican immigrant Catholics and the Catholic diocese of Atlanta, Georgia, as well as its nonimmigrant white and African American residents. Indeed, La Misíon stands as a testament to the tenacity and courage of Atlanta’s “newcomers” who wanted to practice their Latino Catholic religion and carve out a secular space, or “brave new world,” in Atlanta. Mexican immigrants who came to Atlanta brought with them their Catholic religion with its specific rituals and liturgy rooted in Mexican culture. The arrival of Mexican immigrant Catholics in Atlanta posed a challenge for the Catholic Church. How would Mexicans be integrated into a predominantly white, nonimmigrant congregation?31 Catholic Mexican immigrants in Atlanta hoped to “transplant” their Catholicism to their new community and churches. Mexicans wanted to preserve masses celebrated in Spanish, devotion to the Virgen de Guadalupe, religious festivals, and other events in the Catholic calendar, as well as Mexican rituals associated with their Mexican culture. Mexicans ultimately fought for the establishment of the Misión Católica de Nuestra Señora de Las Américas where they could preserve their Mexican Catholic traditions. The establishment of the Misión represented a “grassroots

A Demographic Profile of the Mexican-Born Population    323

struggle by Latin American immigrants to find a religious space where they could practice their faith in a familiar, welcoming environment and find the material, social and spiritual resources to deal with hardship of migration and adaptations to life in the United States.”32 Their efforts eventually led to conflicts with Atlanta’s Catholic archdiocese, which found itself in a quandary. The Catholic Church wanted to serve and welcome newcomers, specifically immigrants but at the same time wanted to help Mexican and other Latino immigrants adjust to the practice of Catholicism in the United States, which that would replace rituals practiced by Mexican and other Latin American immigrants.33 The archdiocese even feared the possibility of these immigrants moving away from Catholicism to evangelical Protestantism as one possible outcome of its lack of sensitivity to the specific needs of Mexican immigrants, Mexican-Americans, and other immigrant groups.34 As a result, the archdiocese designed a concerted outreach to Mexican immigrant Catholics in hopes of “integrating” them into the mainstream Catholic Church. Atlanta’s Mexican and Central American immigrant community always kept their eyes on the prize: a safe religious space where they could worship their Catholic faith as if they were still in their homeland, which they knew they would probably never see again but wanted to keep in their hearts and minds. They also sought to pass on this religious space to their U.S.-born children. As Atlanta’s immigrants moved closer and closer to building this space, albeit in a neighbor’s living room, then in a grocery store, later in a Protestant Church, and finally in the basement of a Catholic Church, a strong backlash developed from the Anglo community in general and specifically white Catholic parishioners whose immigrant roots were generations removed from the newly arrived immigrants. After a series of meetings, protests, confrontations, and ultimately negotiations, Latino immigrants succeeded in carving out a permanent religious space where they could worship using their own rituals and traditions. The Misíon Católica de Nuestra Señora de Las Americas became the heart of the Catholic immigrant community and truly ministered to the needs of the newcomers to Atlanta, both their spiritual needs and their needs as immigrants in the proverbial “strange land.” The Misíon could not eliminate the inequality and discrimination experienced by the Mexican immigrant community, but it could and did function as a safety net. Immigrants who worshiped at the Misíon eventually used this religious space as a base for civic participation. For example, immigrants learned how to write petitions, engage in petition-signing activities, and, in one case, write a letter of protest against the Immigration and Naturalization Service in Atlanta for arresting day workers without following the proper protocol. In sum, “immigrants used the Misíon as an alternative public space from which to engage in political debate over issues that affect their lives.”35

324  North from Mexico

4. Unauthorized Immigrants Living in the United States In 2013, the total number of immigrants in the United States equaled 41.3 million, representing 13 percent of the total population of the United States (316.1 million). Mexican immigrants represented 28 percent (11.6 million) of the total number of immigrants. In 2014, the total number of unauthorized stood at 11.3 million, making up about 3.5 percent of the total U.S. population. The total number of unauthorized Mexicans equaled 5.9 million, representing 52 percent of the total unauthorized population. Between 2008 and 2012, Mexico and other Central American countries accounted for 71 percent of the total unauthorized population. Half of the unauthorized came from Mexico followed by El Salvador, Guatemala, India, Honduras, China, and the Philippines. The majority of unauthorized lived in four states: California (28 percent), Texas (13 percent), New York (8 percent), and Florida (6 percent).36 Other key demographic patterns of the unauthorized include the following: (1) a decline in number from Mexico; (2) the growth in the South, particularly Georgia and North Carolina, the new-destination settlement region; and (3) the arrival after 1995 of over 50 percent. A significant number of unauthorized are younger that the general population. In addition, unauthorized immigrants are likely to have children who are U.S. citizens living with them. Taken together these demographic patterns will contribute to a change in the social fabric of the United States, a change that will challenge politicians, employers, educators, and others. Approximately 20 percent of the unauthorized population reside in five counties in the United States: Los Angeles and Orange counties, California; Harris County, Texas; and Queens County, New York. Between 2009 and 2012, the total number of unauthorized immigrants increased in Florida, Idaho, Maryland, Nebraska, New Jersey, Pennsylvania, and Virginia but decreased in Arizona, California, Colorado, Nevada, New Mexico, and Oregon.37 During the U.S. recession of 2007–2009, many Mexican immigrants reassessed the economic rationale for making the trip to El Norte. Because of the deterioration of the U.S. economy, Mexicans, at least temporarily, considered other options in their search for a better life. The recession led to other factors that contributed to the decline in the unauthorized immigration from Mexico. The collapse of the housing market in the United States led to a marked decline in the construction industry, a major employer of unskilled or semiskilled Mexican immigrants. The climate of antiterrorism inevitably led to a growing anti-immigration fervor, harsher penalties, and eventually increased deportations. The danger of border crossings also acted as a brake on the flow of Mexican immigration. Developments within Mexico also contributed to the decline. From 2010 to 2011, the Mexican economy showed signs of entering a recovery stage, one that led more Mexicans who would have decided to

A Demographic Profile of the Mexican-Born Population    325

undertake a border crossing to remain in Mexico hoping for greater economic improvement in their lives. Between 1995 and 2000, about 3 million Mexicans came to the United States, but almost 700,000 returned to Mexico, including some U.S.-born family members. The U.S.-born Mexicans who return to Mexico with their parents face major problems such as language differences, because many are monolingual English-speaking. Mexicans still form the largest part of the unauthorized immigrant population, a trend that is likely to remain the same in the future.38 In 2012, 8.1 percent of the U.S. labor force consisted of unauthorized immigrants. Unauthorized Mexican immigrants clustered in certain low-paying occupations: private household work, landscaping, service, manufacturing, maintenance, and construction. Unauthorized immigrants seem to cluster in these subsets of more general occupations more so than other immigrants and the native-born population. In 2012, 62 percent held service, construction, and production jobs, twice the share of the total U.S. population. In general, unauthorized immigrants are more likely to work in low-skilled or unskilled jobs as farm laborers, drywall installers, roofers, domestic workers, painters, carpet and floor installers, in-home child care providers, and elderly caregivers.39

5. Looking Toward the Future Although the Mexican immigrant population is not homogeneous, a review of several key demographic characteristics can function as signposts for the future of society, culture, and politics in the United States. Once again we return to Oscar Handlin’s assertion that “immigrants are American history.” Mexican immigrants coming to the United States continue to produce social changes that contribute to the fabric of diversity; however, their existence unfortunately also fuels a climate of anti-immigration hostility for others. Additional issues to examine regarding the movement of Mexicans to the United States include the following: reasons for leaving Mexico, point of entry into the United States, the socioeconomic and political conditions in the United States upon their arrival, levels of anti-Mexican attitudes, behavior settlement patterns for women and men Mexican immigrants, occupational characteristics, levels of education, upward mobility, social structural impediments for immigrants, U.S.-born children and later generations, migration to traditional-destination and new-destination states, and the unintended consequences of all of these factors. The influx of younger Mexican immigrants, such as women with higher fertility rates and who have larger families than the native population, will also reshape the nature, scope, and content of social institutions such as education, the legal system, and politics. For example, U.S.-born children of Mexican immigrants will face specific challenges as they advance in the educational system.

326  North from Mexico

Mexican immigrant parents with limited English skills will be at a disadvantage in their interaction with teachers and administrators unless school districts make efforts in predominantly Latino neighborhoods. Lastly, Mexican-born immigrants will continue to remain an immigrant group in a precarious economic standing as a result of low educational attainment levels, largely due to social structural inequities in the education system; other factors include their concentration in low-paying jobs with little or no possibility for advancement and little or no retirement and medical benefits, income stagnation, and, in general, life circumstances that impede the immigrants’ goal of making a better life for themselves and, in turn, their children. Similarly, unauthorized Mexican immigrants will continue to find themselves marginalized at all social and economic levels. The country of origin of immigrants migrating to the United States represents one of the key population characteristics. Trends in the immigration rates of sending countries have the potential to redesign the fabric of American society. The 1965 Immigration and Nationality Act resulted in immigrants from such regions as Latin America and Asia, who had previously been outnumbered by European immigrants entering the United States. Although the rates of unauthorized Mexican immigrants did not significantly decrease, Mexicans took advantage of the requirements stipulated by the 1965 legislation and stayed in the United States. The influx of Mexican immigrants, most of whom stayed in the United States, remained steady, but increased until 2013, the year of a slight decline. Two issues emerge from post-1965 increases in Mexican immigration. First, as these immigrants qualify for voting through naturalization, they have the potential, albeit often untapped, ability to contribute to the shift in political power among Democrats and Republicans. Data show, however, that Mexican immigrants have the lowest rates of naturalization among other foreign born. Nevertheless, the size and permanent stay of Mexican immigrants and their children who are born in United States represent a potential source for a voting bloc. Indeed, in 2012 Republican presidential candidate Mitt Romney underestimated the need to reach out to this group of voters. Romney received only 27 percent of the total Latino vote while Barack Obama won 71 percent. While Mexicans have a long historical tie to the Democratic Party, as evidenced in their participation in the Viva Kennedy clubs in 1960, emerging political winds can, under some circumstances, change. With the number of U.S.-born Mexican exceeding the number of Mexican-born immigrants living in the United States, the 2016 presidential race will inevitably consider this constituency.40 Due to the shared border between the United States and Mexico, a second question remains central to the discourse on unauthorized immigration: How many apprehensions of unauthorized immigrants are there per year? In 2013, the U.S. Customs and Border Protection (CBP) and U.S. Immigration and

A Demographic Profile of the Mexican-Born Population    327

Customs Enforcement (ICE), the Office of Homeland Security’s agencies in charge of deportations, recorded 662,483 apprehensions and a total of 2 million deportations, not all of whom were Mexican born. The combined numbers of apprehended individuals came from Mexico, Guatemala, Honduras, and El Salvador. The majority, 64 percent, were Mexican nationals.41 Since President Obama took office in 2008, a total of 2 million have deportations occurred. In October 2014, Obama met with the Congressional Hispanic Caucus to discuss their growing dissatisfaction with his immigration policies. The Obama administration deported about as many immigrants in its first term as the George W. Bush administration deported over a period of eight years. Contrary to some assessments, the majority of deported immigrants did not, according to DHS and ICE records, have a criminal conviction in 2013. Another trend that has developed since 2014 is the number of unauthorized immigrants and unaccompanied children entering the United States, particularly in south Texas. The case of unaccompanied children, primarily from Central America, poses a particularly thorny legal and ethical issue that is yet to be adequately resolved. Interestingly, fewer unauthorized Mexicans have been stopped and detained at the border crossing; a decline from 1.1 million in 2005 to 425,000 in 2013.42 Unauthorized children who came to the United States when they are younger than eighteen years of age with their unauthorized parents find themselves living in the shadows with their parents for fear of deportation. Meanwhile, these children often have lived in the United States continuously, never in Mexico, for an extended time and, most importantly, have progressed through the American education system. Though these children face the problem of having lived almost all of their lives in the United States, they find themselves blocked from employment, applying for financial aid for higher education and, most troublesome, being at risk for deportation to Mexico, a country that they lived in for only a few years when they were young or, in many cases, had not return to in many, many years. Unauthorized children live in fear that their parents, siblings, and other relatives will be deported. In 2012, approximately 7 percent of children enrolled in kindergarten through 12th grade had a least one parent who had entered the United States without authorization. This population of children lived in four states: Arizona (11.0 percent), California (13.2 percent), Nevada (17.7 percent) and Texas (13.1 percent). Interestingly, for public policy makers, large portions of these children were, in fact, born in the United States.43 The population of U.S.-born Mexicans is growing faster than the foreignborn Mexican population is decreasing. Such an increase brings with it major social and political consequences at various levels, including a larger pool of potential voters and the increase in the growth of both second-generation children of Mexican immigrant parents and children further removed from

328  North from Mexico

their family’s heritage of immigration. Anti-immigration sentiments, attitudes, and legislation will undoubtedly grow, but Mexicans will continue to come “north from Mexico”: El Norte. Their lives and those of their children will ultimately have a dramatic, transformative affect on themselves and, in turn, the total population in the United States.

21

Still Coming North from Mexico: Immigration Constraints and Contestations Alma M. García

Whether they crossed the Atlantic Ocean, the Pacific Ocean, or La Frontera, immigrants leave their homeland in search of a new life.1 They make the journey to a “brave new world” with a sense of fear and wonder.2 Mexican immigrants, like other immigrants, left their “old world” and entered a “new world.” Their nostalgia for Mexico was colored by their dreams of a better life for themselves, and, most importantly, for their children. Still, unlike other immigrants to the United States, such as those from Europe, Mexicans found that the “brave new world” into which they migrated bore indelible markings of the homeland they left behind. Mexicans left Méjico to travel north from Mexico but found palpable cultural evidence of Mexico in El Norte: preexisting Mexican immigrant and later-generation Mexican-American communities with all types and varying levels of Mexican cultural traditions.3 To understand immigration in general, like Oscar Handlin did in his classic, The Uprooted, scholars need to compare and contrast the experiences of one immigrant group with another.4 The border between Mexico and the United States is one in which cultures meet, collide, but also blend. The immigrant experiences pose gender-specific problems, with women often facing sexual violence along the border. Mexicans navigate through a contested terrain, one characterized by various degrees of prejudice, discrimination, and institutionalized racism. Yet, Mexican immigrants maintain a spirit of hope that they will survive in this often-hostile environment and pass on their survival skills to their descendants. Mexicans do not cross the Atlantic Ocean from Europe, nor do they come from a “different

330  North from Mexico

shore” by crossing the Pacific Ocean.5 Their journey to El Norte is one involving geographic crossings, but also one with many cultural continuities, changes, and contestations. The border is always a fluid one, which enables Mexicans— both authorized and unauthorized—to cross into El Norte and then, for many, cross back to Mexico and then back again to the United States. In his classic memoir Barrio Boy, Ernesto Galarza describes his family’s journey from Jalcococtán, Nayarit. After a long and arduous train ride, his family arrived at La Frontera and crossed into El Norte. Ernesto’s mother explained to Ernesto that they were now in El Norte, “Look the American flag.” [Galarza’s] mother said “We are in the United States, Mexico is over there.”6 The North was the same place as the United States. Like Galarza, Francisco Jiménez, in his book The Circuit, recalls his family’s dream of going to La Frontera: “La frontera” is a word I often heard when I was a child living in El Rancho Blanco, a small village nestled on barren, dry hills several miles north of Guadalajara, Mexico. I heard it for the first time in the late 1940s when Papa and Mama told me and Roberto, my older brother, that someday we would take a long trip north, cross la frontera, enter California, and leave our poverty behind. I did not know exactly what California was either, but Papa’s eyes sparkled whenever he talked about it with Mama and his friends. “Once we cross la frontera, we’ll make a good living in California,” he would say, standing up straight and sticking out his chest.7

Unlike Galarza’s family, like so many other Mexicans at the time, who fled Mexico to avoid the upsurge in violence during the Mexican Revolution of 1910, the Jiménez family fled the rampant poverty of their homeland and crossed the border during the late 1940s. The stories of Mexican immigrants, like Galarza and Jiménez, represent a major component in the narrative of U.S. history. While Galarza and Jiménez depicted immigration from a male point of view, Chicana scholars and creative writers have illuminated this process from a woman’s vantage point. Chicana feminist writers, such as Gloria Anzaldúa, view the frontier, referred to as the borderlands, as a site for shifting ideologies of nationality, gender, sexual orientation, race, and social class. Anzaldúa describes this geographical and ideological spaces as an “open wound,” one in which women are likely to experience violence. Similarly, in her novels, Chicana writer Ana Castillo portrays the borderlands as a precarious geographical space where women who cross the border confront violence, particularly if they cross as unauthorized women immigrants.8 The United States has always faced a classic dilemma. On the one hand, the country has relied on the national trope of American exceptionalism embodied in the seventeenth-century Puritan hope of building a radiant “city on a hill,” one that would shine its light on the rest of the world for the

Still Coming North from Mexico   331

philosophical and pragmatic political purposes of building a democratic new world.9 On the other hand, the country had to reconcile its belief in democracy and freedom with a long list of exclusionary practices (some more pernicious than others), including the massacre and near extinction of indigenous peoples; the establishment of the “peculiar institution” of slavery; the expansion of the country under the banner of “manifest destiny,” or the providential right to rule from “sea to shining sea”; the disenfranchisement of women until the Nineteenth Amendment passed in 1920; the institutionalization of jim crow laws developed to exclude African-Americans from participation in U.S. society; the internment of Japanese and Japanese-Americans during World War II; and, relevant to this study, a series of exclusionary de jure and de facto laws and practices implemented to deprive Mexicans and U.S.-born MexicanAmericans of their civil rights.10 The United States has witnessed periods in which various sectors of Congress initiated immigration reforms. Ironically, those measures often reflected the efforts by businesses to maintain a constant influx of a cheap labor force, specifically low and unskilled Mexican workers.. Some overarching questions raised by the passage of immigration laws include the following: (1) what combination of political, economic, and social interest groups either supported or opposed a specific immigration act? (2) Did such views change over time and, if so, why? (3) What impact did a given immigration act have on U.S.-Mexico relations? and (4) What impact did a given immigration act have on U.S. international relations, particularly in the post-9/11 period? and (5) How do relations between Mexico and the United States contribute to the debate over immigration? While the passage of the Immigration Reform and Control Act (IRCA) of 1986 addressed some key issues dealing with immigration, such as family unification, immigration reform remained a critical agenda item on the national, state, and local levels. The discourse on immigration is a long-standing issue with many components: liberalization of existing controls, increased restriction, the criminalization of the unauthorized and those assisting them, paths to citizenship for unauthorized immigrants, health care, and violence along the border, particularly against women. These play out within the sociopolitical and economic relations between the United States and Mexico. Several key binational socioeconomic and political developments contributed to the dynamics of immigration in the post-1986 period. These factors created a migratory dynamic that allowed for the conditions for the movement of people north from Mexico to El Norte.

1. Maquiladoras: A Plan for Economic Development A prelude to Mexico’s economic policies in the 1990s, the maquiladora program represented Mexico’s attempt to solve its increasingly grave economic

332  North from Mexico

problems. Adopted in the mid-1960s, this binational agreement between Mexico and the United States was intended to meet the needs of growing unemployment in Mexico and the rising costs of wages in the U.S. manufacturing sector, particularly the electronics industry. Maquiladoras are assembly plants in newly created export processing zones along the U.S.-Mexico border in such cities as Tijuana, Mexico. Companies in the United States shipped partially assembled parts to these zones, free of import taxes, where Mexican factory workers completed the production process; the finished products were then “imported” to the United States free of tariffs, thus increasing profits because of the decreased cost of wages.11 Two major factors led to the development plan’s questionable economic success. Although the unemployment rate did indeed decrease slightly for male workers, the maquiladora plants, like global assembly plants in other parts of the world, sought out the cheapest group of workers—young Mexican women, many of whom did not have experience in the paid labor force. These women worked under poor conditions, and owners preferred to fire women who developed medical problems rather than provide them with medical benefits. With an available pool of Mexican women willing to work in the plants, maquiladoras never experienced a labor shortage. An increased migration to the United States of young and mostly unauthorized Mexican women represented one of the unintended outcomes of the maquiladora program and a source of increased strain between the two countries. In addition, public perceptions of this migration fueled anti-immigrant sentiments.12

2. Demonization of Mexican Immigrants: The Context for the Immigration Act of 1990 President Ronald Reagan’s policies on Mexican immigration contributed to a demonization of the Mexican immigrant, a long-standing trend in the history of the United States. During his two terms as president from 1981 to 1989, Reagan faced a series of challenges that would ultimately cast Mexican immigrants as “threats” to the core of U.S. society: its values, culture, economic wellbeing, and national security. The Reagan administration and its supporters cast immigrants, both Mexican and Central American, specifically unauthorized immigrants, as a threat to the very fabric of “American” society. Mexican immigrants, like their late nineteenth and early twentieth century European counterparts, did not tear the national fabric: they wove it into it a different, more vibrant society. Several factors shaped the Reagan period and its overarching antipathy toward immigrants, factors that became largely responsible for anti-immigration bills, state initiatives, and negative perceptions of immigrants. At the beginning of his first term as president, Reagan and his foreign policy advisers

Still Coming North from Mexico   333

identified the rising upheavals in Central America as ones fueled by Communist insurgents. The fear, or near hysteria, of a Communist takeover of countries such as Nicaragua and El Salvador, shaped U.S. foreign relations with Latin and Central America. Moreover, political developments in Central America were believed to lead to unrest with Mexico, which the Reagan administration argued would prove catastrophic given Mexico’s shared border with the United States. Relations between the United States and Mexico have been shaped by Mexico’s internal politics, particularly its economic ones. As the Mexican economy imploded, anti-immigration advocates in the United States called for more stringent, even draconian measures, to secure the border. The series of economic crises that crippled Mexico from 1970 through the 1990s exacerbated the widespread fear in the United States that immigrants would be “pushed” out of Mexico and set the stage for a new movement of people coming north from Mexico to seek redress from such conditions who would “take” jobs from Americans. From 1950 to 1970, Mexico, like many countries in Latin America, such as Brazil, turned to a national import-substitution industrialization development plan that would increase domestic manufacturing and, as a result, decrease the country’s reliance on imports. Although some economic growth and limitation on imports did indeed develop, import substitution eventually experienced an economic bottleneck. As a result, Mexico sought foreign loans to shore up its economy, but the country paid a heavy price: slashing wages and reducing social services, which combined to increase social unrest among the poor and contempt from a growing middle class that had been experiencing limited, but gradual, levels of upward mobility and increased savings. In 1976, the devaluation of the Mexican peso to 50 percent of the U.S. dollar reflected a near demise of the national economy that resulted in a downward spiral of the middle class, cuts in wages and social services, and an increase in food prices, putting the working class in an even more precarious position. The discovery of huge oil reserves in the Gulf of Mexico in 1974 had raised the hopes of Mexicans that their country was on its way to economic recovery under the leadership of President José López Portillo. Mexico’s petrolization policy, however, would lead to further economic crises. Unanticipated costs of revamping Mexico’s infrastructure to bolster rapid and lucrative oil exports led Mexico to incur an astronomical foreign debt, one that the country could not repay. Mexico eventually—but not surprisingly—was forced to implement a draconian measure: the suspension of its debt repayment. Grave international consequences soon followed. In addition, corruption within the highest levels of the presidential administration and ill-advised national investment contributed to Mexico’s deteriorating economy. Mexico’s oil industry collapsed as the price of oil from other oil-producing countries decreased the price of oil

334  North from Mexico

overall. Mexico’s development bubble burst in 1982 when it declared bankruptcy. Later, between 1988 and 1994, Mexican president Carlos Salinas de Gortari initiated some reforms but, once again, political corruption, scandals, and growing urban violence led to Mexico’s further economic fall.13 In addition to Mexico’s severe economic problems, the Cold War intensified Reagan’s antipathy to Latin American immigration. The 1980s would become a high-water mark in the history of U.S.-Soviet relations. Reagan used every opportunity to remind the American people that the Kremlin continued to bolster Castro’s Cuba, although the Soviet Union had “winked” by removing its nuclear-armed missiles from Cuba in 1962. The political slogan that “Cuba was only ninety miles from the United States” had immigration implications. On the one hand, the United States welcomed Cuban refugees who fled as a result of the 1959 Cuban revolution and who mainly settled in Miami, later referred to as “Little Havana.” On the other hand, its fear that Cuba would “export” its Communist revolution throughout Latin America, particularly Central America and Mexico, intensified and further validated the demonization of immigrants because they were seen in conservative Reagan circles as both immigrants and Communists. As a result, the cry for increased national security measures and control mechanisms along the U.S.-Mexico border became a conservative mantra. Both the efforts to reform existing immigration laws, such as the Immigration Act of 1990, and a resurgence of anti-immigrant state initiatives such as California’s Proposition 187 of 1994, reflected the continuation of the national and state dialog on immigration.14

3. The Immigration Act of 1990 A brief review of the 1986 IRCA serves as a context for the Immigration Act of 1990. Passed by Congress in 1986 and signed on November 6 by President Ronald Reagan, the IRCA contained the following provisions: reduction of unauthorized immigration, implementation of sanctions against emp­­ loyers  that hired unauthorized immigrants, and establishment of new amnesty requirements for individuals living in the United States without authorization.15 The IRCA did not answer all the immigration issues, and as a result, the debate over further immigration reform continued. Many legislators and immigrant rights groups called for greater reforms related to the process of obtaining a pathway to citizenship for the unauthorized. Some members of Congress maintained their opposition to immigration reforms out of their strident belief that unauthorized Mexican immigration, if left unchecked, would jeopardize the fabric of American society. Their pragmatic political considerations for being reelected in conservative, anti-immigrant states deepened their opposition to immigration reform.

Still Coming North from Mexico   335

Signed on October 27, 1990, by President George H. W. Bush, the Immigration Act of 1990 represented a significant change since the Immigration Act of 1965 and the Immigration Act of 1986. Sponsored by the late Senator Edward Kennedy and the late Bruce Morrison, the former Democratic congressman from Maryland, the 1990 act increased the number of authorized immigrants that entered the United States every year from 500,000 to 700,000. It also addressed one of the major criticisms of the 1965 act: its limitations regarding family reunification. The 1990 act defined familybased immigration of immediate relatives of U.S. citizens and added four preferences by increasing the number of family-based visas intended to facilitate increases in family reunifications in immigrant families, specifically in the United States. The IRCA of 1986 allowed unauthorized persons who had lived continuously in the United States for five years after 1982 to avoid deportation, become eligible for temporary resident status, and then, if all requirements were met, apply for permanent legal residence—a green card. Although this provision represented a significant reform, it failed to take into consideration family immigration patterns from Mexico: some members of Mexican immigrant families did not meet the residency requirement and as a result many families became had mixed legal status. In May 1988, the Immigration and Naturalization Service announced that spouses and children would be given special visas in order not to break up families. In an effort to address what many in Congress and sectors of the general public identified as a major weakness in past immigration laws, the Immigration Act of 1990 provided for a temporary halt to deportations of those family members who missed the cutoff date. In addition, it provided work authorizations for spouses and unmarried children of those who did not qualify under IRCA, although they were barred from receiving aid from some public assistance programs. The 1990 act also gave admission preference to immigrants with designated skills and occupations in response to the businesses that were experiencing a shortage of highly skilled and educated professionals, particularly those in high-tech industries. This provision reshaped the demographics and nature of immigrant communities in Silicon Valley, California, and other high-tech areas. The 1990 act also included a provision that established visas that became known as Morrison visas after one of the bill’s cosponsors. This is also referred to as the “diversity” provision because it was designed to address immigration patterns since the passage of the Immigration Act of 1965. Under this provision, the attorney general was designated as the person to identify those countries and regions of the world whose admission rates were underrepresented within the total number of immigrants who were eligible for admission under the 1965 act. It allotted 40,000 visas each year for three years to countries that had been disadvantaged by the 1965 immigration legislation. To accomplish this, the Immigration Act of 1990 created the first permanent visa lottery program that

336  North from Mexico

would increase immigration from those countries whose admissions rates were low. Although these changes were called the “diversity” component, critics and supporters remained skeptical. Opposition to this section led to an increase in anti-immigration sentiments that would increase the political and social friction surrounding immigration, particularly in areas with large Mexican immigrant communities. Another provision aimed primarily at Central Americans contributed to increased anti-immigration rhetoric. The act created a “safe haven” for individuals fleeing their homelands, such as El Salvador, for fear of persecution or even death. Those who met this criterion received protected status and were allowed to stay in the United States. As a result of the increase in immigration from El Salvador, historical Mexican immigrant communities such as the Mission District in San Francisco, California, were forever changed. Ironically, the passage and implementation of the Immigration Act of 1990 came at a time of a severe economic recession that began in the early 1990s. This economic downturn accompanied surging anti-immigration fervor. Many believed immigrants, authorized or not authorized, were taking jobs away from U.S. citizens. The post-1990 era in immigration laws fueled a long-standing anti-immigration narrative that remained explicitly and implicitly the center of political discourse. At the state level, the smoldering anti-immigration climate in California resulted in the passage in 1994 of Proposition 187.

4. California’s Proposition 187 of 1994 Prior to the introduction of Proposition 187 in California, other states had already initiated various measures, such those stipulating that English be the official language for the state. By 1998, twenty-five states had passed “English Only” measures, but they were mainly “symbolic gestures with few practical consequences; they simply provided voters or legislators with a tangible means of registering their dislike of foreigners.”16 Several factors led to Proposition 187. Starting in the late 1980s and continuing into the years leading up to 1994, the state of California faced several problems that provided a context for the climate of support for the proposition. At the most general level, supporters of the proposition wanted to register their disgruntlement with President Bill Clinton and Congress not only over immigration but also other issues. During this period, California experienced a declining economy that future pro-187 supporters would attribute to the presence of unauthorized immigrants, mostly Mexicans, although the Central American immigrant population had also been increasing. Many of the supporters of the proposition resided in Orange County, one of the wealthiest and most conservative counties in California. Those supporters argued that the

Still Coming North from Mexico   337

greater the percentage of unauthorized immigrants who lived in the state, the greater likelihood of more jobs being taken from “Americans.” Proponents viewed access to social services by unauthorized individuals as one of the major engines driving the California economy into a downward spiral. Other economic factors contributing to a once-thriving economy included several years of recession, the elimination of military bases, and a decline in the aerospace industry due to cancellations of contracts. In addition, the increased number of unauthorized and authorized Mexican immigrants living in California fueled a historical “fear” that California, and the United States in general, would be “taken over” by Mexicans, considered as racialized “others” who would unravel the country’s social fabric. The term “American citizen” became a racialized code word understood to mean “white Americans.” The process through which racial and ethnic groups were defined in inferior racialized terms was not a contemporary phenomenon; it prevailed in historical and contemporary circumstances throughout the United States.17 Supporters of Proposition 187 often referred to it as the “Save Our State” (SOS) initiative. Critics of the proposition quickly pointed out that SOS is the international code signal of extreme distress, an urgent call for rescue. As such, supporters of this ballot initiative saw themselves as coming forward to “save” California” from the surging “wave” of unauthorized immigrants, primarily from Mexico. At the time, an estimated 1.3 million undocumented individuals, including 308,000 children, lived in California, a number that further alarmed Proposition 187 supporters. Proponents referred to “SOS” as a call for relief from anticipated tax increases that would be used to provide social services for the undocumented, particularly for the education of undocumented children. However, Proposition 187 flew in the face of the 1982 U.S. Supreme Court decision in Plyler v. Doe that held that the Constitution guarantees all children, regardless of immigrant status, equal access to public education.18 Proposition 187 banned undocumented immigrants from receiving such social services as health care and education. Within a short time, thengovernor Pete Wilson, a Republican, joined other supporters and integrated Proposition 187 into his bid for reelection hoping that his rapidly declining voter approval rate would experience an upward surge, building a wider base of voters for his campaign. Supporters represented a wide range of conservative groups who organized statewide functions to raise money for commercials and print advertisements. Elected officials and former officials became vocal advocates of Proposition 187. Challenging Pete Wilson in the race for governor, California state treasurer Kathleen D. Brown, the daughter of former governor Edmund G. (Pat) Brown and the sister of the current governor, Edmund G. (Jerry) Brown Jr., went against the advice of her campaign strategists by urging voters to defeat Proposition 187.19 This support, and other

338  North from Mexico

factors, including the declining state economy, contributed to her defeat by Pete Wilson for California’s governorship. A Los Angeles Times story on Proposition 187 captured the philosophy that led to its ultimate passage when it quoted William Dannemeyer, a former Republican congressman from Fullerton, California, who had served for seven years: “If this [Proposition 187] doesn’t pass, the flood of illegal immigrants will turn into a tidal wave, and a huge neon sign will be lit up above the state of California that reads, ‘Come and get it.’”20 Former president Bill Clinton opposed Proposition 187 and urged Californians to reject it. Clinton opposed the proposition based on the premise that only the federal government, not the states, could create policies to deal with immigration rather than focusing on the humanitarian aspect of the proposition. Throughout the United States, religious groups, such as Catholics and Jews, worked with other groups, such as long-standing local and state grassroots organizations within the Latino community, to mobilize against Proposition 187. Many in the Asian-American community drew a historical parallel with Executive Order 9066, signed in 1942 by President Franklin D. Roosevelt, which ordered the removal of Japanese and Japanese-Americans from parts of the West loosely identified as military areas. A Japanese-American woman who participated in the anti–Proposition 187 campaign echoed the sentiments of many within her community: “I worry that the kind of anti-ethnic hysteria driving [Proposition] 187 is akin to the anti-ethnic hysteria that swept the country after Pearl Harbor.”21 On October 16, 1994, Los Angeles witnessed an anti–Proposition 187 march, with an estimated 100,000 demonstrators from diverse ethnic groups and backgrounds, specifically Latinos, Asians, and African-Americans. Mexican immigrants and opponents of the proposition took to the street to protest the state initiative and Governor Pete Wilson, one of the initiative’s staunchest supporters. Marching from East Los Angeles to downtown, the massive demonstration represented the success of various groups within and outside of the Latino community in mobilizing a wide swath of opponents of Proposition 187. Once at City Hall, a series of speakers, including politicians, activists, clergy, and everyday people, joined the surging backlash against this state initiative. Capturing the sentiments of the marchers who led everyday lives as hard-working parents with children who would be targeted by the state initiative, one Latino who protested against Proposition 187 said: “This proposition is not against the illegal (immigrant), it’s against children.”22 Joe Hicks, executive director of the Southern Christian Leadership Conference, urged those present to rally together and make their opposition to Proposition 187 heard throughout the country: “We’ve got to send a message to the rest of the nation that California will not stand on a platform of bigotry, racism and scapegoating,”23 A well-known community activist from East Los Angeles and

Still Coming North from Mexico   339

one of the march’s key organizers, Juan José Gutiérrez, explained the larger significance of the march: “This is not a parade, this is a social movement.”24 As a sign of cultural hybridity, a mariachi band played the “Star Spangled Banner.” A few days before the critical vote, Latino students, particularly high school students, walked out of their schools in protests reminiscent of the 1968 Chicano student “blowouts” in Los Angeles to protest segregated conditions and the inferior education system in East Los Angeles.25 An estimated 10,000 students took to the streets in an attempt to mobilize other groups to join in this struggle, one similar to the Chicano movement for civil rights in the 1960s. A statewide anti-187 student coalition further mobilized students in cities throughout California.26 University and high school students throughout California also staged protests rallies and demonstrations. In March 1994, students at the University of California at Berkeley organized a rally that was cosponsored by a coalition of student groups and community organizations.27 On November 8, 1994, California voters passed Proposition 187 by a margin of 18 percent: 59 percent of voters were in favor and 41 percent were opposed. A large divide existed between supporters from each political party: 78 percent of Republicans voted for Proposition 187 while only 36 percent of Democratic voters cast their ballot in support. White voters overwhelmingly supported the proposition (63 percent), and 23 percent of Latino voters also voted in favor. Analysis of the support by Latinos came from both political parties. Republican strategists wanted to increase their party’s outreach to Latino voters while Democratic strategists wanted to bring them back into their party, a long-time magnet for Latino voters (with the exception of CubanAmericans). Based on data gathered by voting behavior analysts, perceived economic threat influences Latino opinions on immigration issues such as Proposition 187. Those Latinos who supported the proposition, were more likely to believe that unauthorized individuals represented a personal threat to their economic status by lowering wages, taking away jobs, and increasing taxes to pay for social services, particularly health care. In addition, Latinos with higher incomes and occupational levels reported significantly greater support for Proposition 187.28 Opposition to Proposition 187 continued to gain widespread support. Not surprisingly, the day after voters approved Proposition 187, organizations such as the Mexican American Legal Defense and Education Fund (MALDEF) and the American Civil Liberties Union (ACLU) filed lawsuits to overthrow the proposition. The California State Parent-Teacher Association opposed the proposition’s stipulation that parents would be required to provide the legal status of their school-age children. Denver’s mayor encouraged a boycott of California and discouraged Denver-based organizations from having their conferences in California. The leadership of the League of United Latin

340  North from Mexico

American Citizens (LULAC) announced that its 100,000 members would support the repeal of Proposition 187.29 On November 9, 1994, the ACLU and MALDEF filed lawsuits to challenge the constitutionality of Proposition 187. The plaintiffs wanted to prevent the implementation of all of the provisions listed in the proposition. A federal district court judge issued a temporary restraining order that prevented any implementation of Proposition 187. Meanwhile, tensions deepened in California as Mexican and Mexican-Americans experienced increased incidents of discrimination. Hotlines were set up to handle the growing number of complaints of discrimination. In March 1998, Judge Mariana Pfaelzer declared Proposition 187 unconstitutional citing the 1982 Supreme Court decisions in Plyler v. Doe that all children under eighteen years of age are entitled to public education regardless of their immigration status. Judge Pfaelzer also overturned Proposition 187 on the grounds that immigration law is a federal and not a state issue. Governor Wilson appealed this decision but his successor, Gray Davis, decided to send the issue into mediation. On July 29, 1999, after five years of legal battles, a court-approved mediation committee validated the court’s 1994 court injunction, confirming the federal government’s sole authority over immigration and therefore declaring Proposition 187 unconstitutional. The settlement prevented the enforcement of Proposition 187’s major provisions. The only provisions that were upheld involved the sanctions placed on making, using, distributing, or selling false citizenship documents.30 Proposition 187 became a bellwether for the immigration debate that would continue in the decades after 1994. The anti–Proposition 187 campaigns set the stage for the continuation of a political rift between Republicans and Mexican-Americans and the Latino community in general. MexicanAmericans have had a deep allegiance with the Democratic Party. The Viva Kennedy campaign in the 1960 presidential election between Richard Nixon and John F. Kennedy showed to both Democrats and Republicans that the Mexican-American vote could not be ignored by politicians at any level. In the popular vote, Kennedy’s margin of victory was among the closest ever in the history of the United States, and Mexican-Americans voted overwhelmingly for Kennedy. This is most significant because the poll tax was declared unconstitutional in 1964 with the passage of the Twenty-fourth Amendment. The tax targeted eligible voters from low-income brackets, and MexicanAmericans were the group most likely to fill this category.31 Latinos from different countries of origin, specifically Mexican-Americans, who represented the largest population within this group, transformed their political activism against Proposition 187 into increased participation in the polls. The overwhelming support for Democrat Barack Obama in the 2012 presidential election showed that the Republican Party and its candidate, Mitt Romney, failed to win even a small block of these voters. California remains

Still Coming North from Mexico   341

the “bluest” state. In 2014, twenty years after the passage of Proposition 187, Antonia Hernández, a former head of MALDEF, assessed the impact of the anti–Proposition 187 activism: “It was [Proposition] 187. . .that unified the community.”32 Despite the repeal of Proposition 187, its introduction, passage, and successful challenges in the court fueled anti-immigrant sentiments not only in California but also in other states across the county. This climate pushed many immigrants to begin the necessary steps toward U.S. citizenship, although for many their unauthorized status precluded such efforts. In December 2000, President Bill Clinton signed a new immigration law that eased the existing residency criteria for naturalization, which disqualified immigrants if they left the country during the application process. Clinton also attached an amendment to a pending budget bill to allow for changes in the immigration law. Clinton’s amendment eliminated the requirement stipulated by the 1986 IRCA that stated that undocumented immigrants who wanted to apply for citizenship had to leave the country and apply for readmission to the United States, a process that could last as long as ten years. The immigration reforms signed into law by President Clinton represented important developments for Mexicans. During the election year of 2000, both Democratic and Republican candidates competed for the votes of Mexicans who were naturalized citizens and U.S.-born Mexicans. Immigration reform represented a key to winning their votes. Anti-immigration groups, however, continued their efforts to defeat immigration legislation that, according to them, would create increased immigration. They opposed establishing more lenient naturalization requirements, fearing a potential influx of undocumented immigrants who might become citizens. The Coalition for Immigrant Reform led the opposition to Section 245(i) of the Immigration Reform and Control Act, which removed the requirement that undocumented immigrants had to leave the country before applying for citizenship. With the election of President George W. Bush in 2000, supporters of continued legislation for further leniency in the naturalization process feared that the conservative wing of the Republican Party would pressure President George W. Bush into vetoing immigration reform laws. While he was governor of Texas, Bush maintained a record that demonstrated his political goal of establishing “good neighbor” relations with Mexico by not pressing for strict anti-immigration policies. From the earliest months of his presidency, Bush continued this policy, favoring the path of binational cooperation between the United States and Mexico. Bush and his Latin American foreign policy advisers maintained that increasing Mexican economic development would be the best way to curb illegal Mexican immigration. As a result, Bush moved closer to a centrist position regarding immigration than many conservative Republicans in the House and Senate by focusing

342  North from Mexico

more on the status of the millions of undocumented Mexicans living in the United States. Bush established a task force on immigration that would give legal status to the growing population of unauthorized Mexicans living in the United States.33

5. Unauthorized Immigration: Persistent Struggles The issue of unauthorized immigration from Mexico has always represented one of the most highly charged issues in the United States. Elections have been won or lost as a result of a candidate’s platform on immigration, specifically unauthorized immigration to the United States. On the one hand, immigrants have been the hallmark of American society, one symbolized by the classic inscription on the plaque on the Statue of Liberty: “Give me your tired, your poor, your huddled masses yearning to breathe free.” On the other hand, antiimmigrant sentiments persist, often resulting in violent episodes. Unauthorized Mexican immigration has fueled a general antipathy toward immigrants in general, but given the large population of U.S.-born Mexicans of immigrant ancestry, specifically those living in the Southwest, Mexican-Americans are often seen as immigrants at best and “illegal” at worst. A key question remains: What does the United States “do” with unauthorized immigrants, many of whom have been living in the country for decades.34 Data on the total number of undocumented individuals living in the United States is difficult to obtain, making it necessary to rely on estimates. The Department of Homeland Security and the American Community Survey of the U.S. Census Bureau, among others, are sources for estimates of the total foreign-born population. By subtracting the number of legal residents, naturalized citizens, asylees, and refugees from the total number of the foreignborn living in the United States, estimates of the unauthorized are reached. This estimate, however, is most likely to be skewed in the direction of an undercount given the difficulty in counting the undocumented population because of the fear of deportation if they interact—much less answer questions—from census takers. The Department of Homeland Security defines the category “unauthorized immigrants” as referring to foreign-born persons who entered the United States without inspection or who violated the terms of a temporary admission and who have not (1) acquired the status of a legal permanent resident or (2) acquired the status of temporary protection against removal by applying for immigration benefits. The following foreign-born persons are not considered to be unauthorized residents in population estimates: refugees, asylees, and parolees who have work authorization but have not adjusted to legal permanent resident status; and aliens who are allowed to remain and work in the United States under various legislative provisions or court ruling.35

Still Coming North from Mexico   343

Although the U.S. Census Bureau, the American Community Survey, and the Department of Homeland Security all caution against longitudinal comparisons and contrasts of the number of the unauthorized immigrants, a look at the data from one year to the next does provide a look at possible demographic trends. For example, in January 2011, an estimated 11.5 million unauthorized immigrants resided in the United States, a decrease from 11.6 million in 2010. One year later, in January 2012, the count stood at 11.4 million, indicating little or no change. Several characteristics, taken as a whole, describe this population. First, Mexicans have persistently represented the highest percentage of unauthorized immigrants living in the United States. In 2010, Mexicans accounted for 58 percent of all unauthorized individuals. Such changes within Mexico and the United States explain the small, but persistent decline or no growth from 2007 as rising unemployment rates in the United States, a slight uptick in Mexico’s economic growth, and tighter border security measures experienced by both. The decline in unauthorized Mexican immigrants coming to the United States is tied directly to a general decline in the total number of unauthorized individuals living in the United States. Second, policy debates on immigration have usually included a discussion of the average length of stay of unauthorized immigrants in the United States and the legal status of their children living with them, many of whom are U.S. citizens. In 2010, an estimated 75 percent of all unauthorized immigrants had lived in the United States for ten years or more and an estimated 50 percent were parents with minor children. The population of unauthorized immigrant families is not monolithic; some have family members whose length of time in the United States varies. Unauthorized immigrants families often have some children who share their unauthorized status having entered the United States at a very young age with their parents and other children who were born in the United States. The prevalence of families of mixed legal status stands at the center of much of the political debate regarding deportations, specifically the impact of deportation policies that do not take these family types into consideration. The issue of civil rights questions can be illustrated with two major questions. How can unauthorized immigrant parents be deported if some of their children are American citizens by birth? How can U.S.-born children be deported along with unauthorized parents? Although the Fourteenth Amendment to the Constitution states that “All persons born or naturalized in the United States, and subject to the jurisdiction thereof, are citizens of the United States and of the State wherein they reside,” a political movement to challenge this birth citizenship clause resurfaced as a result of anti-immigrant sentiments. By adhering to a strict and narrow interpretation of the citizenship clause, the anti-birthright citizenship supporters assert that children born on U.S. soil to undocumented immigrants are not protected under this amendment. Central to their position

344  North from Mexico

is one phrase in the Fourteenth Amendment: “subject to the jurisdiction thereof.” Proponents of the anti-birthright citizenship movement maintain that U.S.born children of unauthorized immigrants, such as those from Mexico, are indeed subject to the “jurisdiction” of, in this case, Mexico because their parents entered the United States without authorization. Bills continue to be introduced in Congress to address the debate questioning the automatic citizenship for persons born in the United States to parents who are unauthorized immigrants.36 Other demographic characteristics establish the need for specific immigration reforms and the development of public policies to address some of the specific issues facing unauthorized immigrants. Out of the entire population of unauthorized immigrants, Mexico has long been the major sending country. During the first decade of 2000, unauthorized Mexican immigrants made up close to 60 percent of the total unauthorized immigrants. Nevertheless, their yearly influx has been declining since 1990. During this same time period, unauthorized immigrants from Asia, the Caribbean, and Central America have all seen increases, albeit small. In rank order, the following countries that have the next largest number of unauthorized immigrants after Mexico are El Salvador, Guatemala, Honduras, and China. Unauthorized immigrants are distributed throughout the United States, but some states rank highest, far surpassing other states. States with the highest percentage of all unauthorized immigrants are California, Texas, Florida, New York, and Illinois.37 Although unauthorized immigration rates to the United States have been declining, some states with historically low numbers of unauthorized immigrants have witnessed increases in this population. Between 2009 and 2012, states experiencing increases included Florida, Idaho, Maryland, Nebraska, New Jersey, Pennsylvania, and Virginia. States showing decreases included Alabama, Arizona, California, Colorado, Georgia, Illinois, Indiana, Kansas, Kentucky, Massachusetts, Nevada, New Mexico, New York, and Oregon. In addition to knowing the states in which unauthorized immigrants have increased or decreased, their participation in the paid labor force continues to be a source of anti-immigrant near hysteria. Although some politicians, television and print journalists, and news pundits claim that unless unauthorized immigrants are deported, “legal” workers will continue to face unemployment. The view that unauthorized immigrants “steal” jobs from American workers has been used historically against immigrants, not only those from Mexico. States with the largest number of unauthorized immigrants in the paid labor force are Nevada (10 percent), California (9 percent), Texas (9 percent), and New Jersey (8 percent). However, these numbers tend to be undercounted due to the informal labor sector that unauthorized immigrants inhabit. Unauthorized immigrant women are most likely to work in the informal labor market as a result of a concentration in such jobs as domestic workers, caregivers for homebound elderly, and child care workers. These jobs are largely

Still Coming North from Mexico   345

unregulated leaving unauthorized immigrant women in an even more precarious situation than unauthorized male immigrants.

6. Immigrants, Dual Nationality, and Voting in Mexican Elections The question of dual nationality represents another issue in the immigration debate. Dual nationality for Mexicans living in the United States involves the question of whether Mexicans can cast absentee ballots in Mexican elections, particularly presidential elections. Historically, Mexican politicians considered Mexican immigrants living in the United States as potential voters. Anti-immigrant advocates, growing more vocal, viewed dual nationality as problematic, a status that would lead to divided loyalties. The U.S. government saw the potential danger of dual nationality under conditions of a war between the two countries. The Cold War exacerbated this fear. Beginning with the dramatic tearing down of the Berlin Wall, the end of the Cold War lessened the fear that immigrants holding dual nationality had the potential to destabilize the United States from within its borders.38 Facing pressure from various sectors, Mexico passed a law in 1996 that allowed Mexicans to hold dual nationality. The law stipulated that Mexican nationals living in the United States who became naturalized American citizens did not have to relinquish their Mexican citizenship. The 1996 law allowed people born in Mexico, or with one parent born in Mexico, to hold dual nationality but they would not be allowed to vote. Mexican immigrants living in the United States could maintain Mexican property rights, travel to Mexico without a visa, and invest in Mexico without the restrictions placed on foreigners. Mexicans born in the United States could apply for dual nationality by registering in any Mexican consulate. The law also stipulated that Mexican nationals living in the United States who became naturalized American citizens did not have to relinquish their Mexican citizenship. For both Democratic and Republican politicians, this law marked a key development in electoral politics primarily because it would “lift barriers that keep millions of Mexicans living north of the border from seeking American citizenship and becoming a powerful voting bloc in border states in the United States.”39 Eventually the Mexican government amended the law to provide dual nationals with the right to vote in Mexican elections, usually through absentee ballots.40 Over the next few years, the law’s implementation stalled as the mechanisms and general logistics for a smooth absentee voting process proved almost insurmountable. Mexican nationals voiced their frustration as their attempts to vote in Mexican elections proved difficult. As noted in the Los Angeles Times, one Mexican immigrant expressed his frustration in the delays in setting up a reliable system of absentee voting when he said, “We are part of [Mexican] society, and we should have a vote.” Echoing these sentiments,

346  North from Mexico

another immigrant expressed his visceral feeling for Mexico long after he came to the United States north from Mexico: “When we lived there [Mexico] we worked the land. We say the land got in us and we can’t get it out.”41 Mexican politicians recognized the importance of this voting bloc estimated at about 15 percent of the Mexican electorate. Some politicians considered such a bloc as a windfall if their political party could only rein in those voters. Mexican nationals represented a powerful voting bloc because remittances sent back home enabled them to keep a foothold in each country. Mexican politicians running for office, including presidential candidates, organized community events in California in attempts to garner votes. The Mexican presidential election of 2006 marked the first time Mexicans living abroad could vote. Of the 12 million Mexicans in the United States, only approximately 55,000 registered abroad, and only 28,000 voted. Two main factors account for this low voter turnout: difficulty understanding all the eligibility requirements and lack of public interest. According to a Pew Research study, only 55 percent of Mexicans living in the United States were not aware that that a presidential election was taking place in 2006.42 In the 2012 Mexican presidential election, Mexicans living in the United States could vote if they were a Mexican citizen by birth or naturalization, 18 years of age, registered on the Federal Electoral Institute’s Federal Registry of Voters, and had a valid voter registration card.43 Dalia Moreno, general coordinator for overseas voter outreach at Mexico’s Federal Electoral Institute, the public and independent organization that oversees Mexican elections in Mexico, noted that in a change from the 2006 presidential elections, eligible Mexican voters, including dual nationals residing in the United States and other countries would receive their ballots by mail and that they would not have to pay return mailing fees. To eliminate much of the confusion in voting, Mexican immigrants apply to vote at the nearest Mexican embassy or consulate. For those with access to the Internet, a Web site, Voto Extranjero (Foreign Votes) allowed for online registration. To be eligible to vote online a person had to be a Mexican citizen, be at least 18 years of age, be registered Federal Electoral Institute’s Registry of Voters, and have a valid voting identification card.44 Approximately 40,000 Mexicans living abroad, mostly in the United States, cast their ballots in the 2012 presidential election that brought Enrique Peña Nieto, a member of Partido Revolucionario Institutional (PRI), to office. The PRI controlled the presidency from 1929 until 2000–2006, when Vicente Fox Quesada of the Partido Nacional Accion served as president, and between 2006 and 2012 with the election of Felipe Calderón. The return of the presidency to the PRI disturbed many Mexican immigrants who viewed it as a return to “corruption as usual.”45 In addition to the struggle to cast absentee ballots in Mexican elections, some Mexicans living in the United States initiated electoral campaigns to

Still Coming North from Mexico   347

seek office in Mexico. For example, some immigrants from the state of Zacatecas channeled their preexisting community activism into political campaigns. Such Mexican immigrants have had a long-standing history of organizing social clubs and mutual aid societies. These mutual aid societies functioned to maintain strong cultural ties with their countries of origin and, perhaps most importantly, provided immigrants, particularly newly arrived immigrants, with economic assistance as they attempted to make a new life in El Norte. As such, immigrants from the state of Zacatecas turned to Mexican electoral politics, in addition exerting pressure on the Mexican government, to continue to streamline the process of absentee ballots. Interestingly, the Zacatecas state legislature created two seats for Mexicans living in the United States, referred by many Mexicans as Mexicanos de afuera [Mexican from the outside].46

7. The Immigrant Spring of 2006: Mass Demonstrations On December 16, 2005, the Border Protection, Antiterrorism and Illegal Immigration Control Act of 2005 (H.R. 4437), introduced by Judiciary Chairman Representative James Sensenbrenner (R-WI) and Homeland Security Chairman Representative Peter King (R-NY), passed the House of Representatives. Three specific sections of H.R. 4437 created panic, anxiety, and ultimately a mobilization of Mexican immigrants and others throughout the country. First, the bill further criminalized the status of those who were in the United States without authorization by changing the violation from a civil offense to a felony criminal offense. For many, this would lead to the permanent separation from their families. The issue of family reunification became an impossibility and contributed to mass opposition for H.R. 4437. Second, H.R. 4437 criminalized anyone who assisted an unauthorized immigrant. The bill stipulated that individuals “who knowingly aid or conspired to allow, procure, or permit a removed alien to reenter the United States to criminal penalty, the same imprisonment term as applies to the alien so aided or both.” This applied to members of religious organizations, nonsecular humanitarian groups, lawyers, and legal aid associations. Third, the Department of Homeland Security would be able to detain undocumented persons stopped at the border even if they could be placed with family members during the time those apprehended had their cases heard in the courts. Fourth, H.R. 4437 increased the penalties and established minimum sentences for undocumented persons who failed to leave the United States when ordered by immigration authorities.47 Although the Senate never passed, H.R. 4437, at least 3 million, or as many as 5 million people, marched in 150 cities to protest H.R. 4437. From February to March 2006, in an effort to put pressure on the Senate to vote no on the bill,

348  North from Mexico

opponents of the bill, both inside and outside immigrant communities, explained that the bill’s tenor, substance, and overall intent was intrinsically antiimmigrant. Together, these specific sections of the bill gave rise to an ideology of collective activism and a spirit of solidarity that cut across social groups from different racial, ethnic, nationality religious, sexual orientation, age, gender, education, occupation, income, immigration status, and many other sources of diversity. Their common goal was clear: to prevent the Senate from passing H.R. 4437. Millions of immigrants and their supporters took to the streets throughout the United States in a passionate effort to issue a collective cry: “Ya Basta! (“Enough!”) From the second week of February to early May, demonstrations erupted in cities with long-standing immigrant communities and in many others with a new, but growing, immigrant population. Numbering between 3.7 and 5 million, people from all walks of life joined together and marched in solidarity in forty-two states including future president Barack Obama. For many participants, the marches during the spring of 2006 became their initiation into a mass political event. More seasoned demonstrators had memories of the civil rights movement, the antiwar movement, the women’s rights movement, and, for many Latinos, the August 29, 1970, National Chicano Moratorium, an anti-Vietnam protest that ended with police violence and the killing of Los Angeles Times journalist Rubén Salazar. On that day, 30,000 Chicanos and their supporters marched through East Los Angeles, a day commemorated every year.48 In February 2006, waves of protests took place across the United States. Participants came from diverse backgrounds, including documented and undocumented Mexican and Central American immigrants, students, workers, clergy, journalists, broadcast news reporters, radio and television personalities, labor union members and activists, community and grassroots groups, and faith-based groups, including the Catholic Church. As demonstrations spread throughout the United States, they grew exponentially with expanded media coverage, including Spanish-language radio, television and newspaper. On February 11, over 500 Latino community leaders met in Riverside, California, to organize the March 6 demonstration scheduled to take place in Los Angeles. The National Alliance for Human Rights, an organization of Latino activists and college and university professors, became one of the main driving forces in bringing together representatives from a variety of organizations. Community organizations, university student organizations, labor unions, and members of the Catholic Church and other religious organizations came together to develop mobilization strategies for nationwide demonstrations.49 With most of the participants wearing white shirts or T-shirts, the development of mass demonstrations crystallized the discontent among many immigrants, specifically unauthorized Mexican immigrants who had maintained a

Still Coming North from Mexico   349

level of community invisibility for fear of deportation. Thus, they had been largely precluded from joining marches and other public forms of protests. Interestingly, one well-known scholar and journalist whose research on Latino immigration spans decades, points out that the English word “protest” does not accurately capture the nature of these protests. Roberto Suro suggests that the mass action is best described by using the Spanish word manifestaciones [manifestations]: People manifested themselves [in the spring of 2006]. They rendered themselves visible. They did not have to say anything. They only had to be there. By simply appearing, and especially because they appeared with their children, they made an existential statement, powerful for its simplicity. “We are here. We are human. Flesh and blood, parents with children.  .  .We are many.”50

The Catholic Church contributed to the organization of these manifestaciones. Although tensions between the Catholic Church and the Latino community in Los Angeles and other parts of the United States existed, these tensions were tempered during the spring of 2006. Indeed, the Catholic Church became an active participant in the struggle for immigrants rights, a role based on religious doctrines and the acceptance, particularly among Jesuits, of the need to embrace and participate in social justice movements in the United States, Central American, and other parts of the world where human suffering resulted from repressive governments. The Catholic Church’s role in supporting its parishioners, mostly Latinos, parallels its role in the 1980s sanctuary movement designed to protect unauthorized immigrants by offering them refuge in churches, a historical tradition found in most places of worship for centuries.51 On June 1, 2005, a year prior to the demonstrations of 2006, Cardinal Roger Mahony of Los Angeles published an editorial in the Los Angeles Times, “A Nation That Should Know Better,” which foreshadowed the Ash Wednesday Lenten sermon he delivered on March 1, 2006: Providing a clear route to legal status for longtime residents and providing legal entry to migrants would not only help cure the excesses of a flawed system but ensures that our nation benefits from the contributions of immigrants participating as full members of their communities. Although some in the public square consider any such rule changes a reward for lawbreakers, we should look at the issue holistically and realistically, and understand that the current law is unjust and must be changed.52

The Catholic Church in Los Angeles, and other parts of the United States, became one of the staunchest supporter of immigration reform, particularly

350  North from Mexico

with the development of the “Justice for Immigrants: A Journey of Hope” campaign started by the U.S. Conference of Catholic Bishops (USCCB).” Through this strategic measure, the USCCB identified and, most importantly, embraced its role as a source of mobilization in the struggle for immigrants’ rights, opposition to H.R. 447, and reform of immigration laws. In addition to attending Sunday Mass, parishioners listened and, in most cases heeded, the call for participation in collective activism preached from the pulpit. In addition, the Church assembled informational booklets for its parishioners: a primer for immigrants detailing the process of civic participation such as writing letters to their government officials. In sum, the Catholic Church, under the leadership of Cardinal Mahony, made a direct connection between dogma and social activism for social justice. Perhaps the most significant example of the activist role in the fight for immigrant rights involved the timing of the demonstrations and the Catholic calendar. The Lenten season, the forty days of fasting and prayer leading to the celebration of Easter on April 16, began on March 1, Ash Wednesday, the first day of Lent in Catholicism and five days before the march scheduled for March 6. Cardinal Mahony used his Lenten sermon to establish the religious tone for the upcoming demonstrations. Although some Catholic clergy and parishioners questioned the activist role of the Church, Cardinal Mahony used this sermon to explain how the immigration march represented an extension of the religious belief in good works and sacrifice. Similarly, the majority of the members of the Society of Jesus (Jesuits) throughout the United States underscored the need for Catholics to move themselves from places of worship and participate, at whatever level possible in the movement for immigrant rights. Waves of demonstrations took place across the United States. Demon­ strations erupted in Arizona, California, New Mexico, Nebraska, New York, Illinois, Texas, Wisconsin, and many other states. A demonstration largely organized by the National Capitol Immigration Committee brought together somewhere between an estimated 20,000 and 40,000 protesters to Washington, D.C. On March 10, an estimated 100,000 to 300,000 demonstrated in Chicago. In other parts of the country, students, who had already jointed demonstrations in their hometowns and others who had yet to participate, engaged in both organized and spontaneous school walkouts. These “blowouts” represented a continuation of those that took place in East Los Angeles on March 6, 1968, a coincidence not overlooked by veteran activists from the Chicano Civil Rights Movement.53 On March 25, Los Angeles witnessed its largest demonstration in the city’s history with protesters numbering anywhere between 500,000 and 1 million. In a strategic move to maximize the number of protesters, organizers turned to the Spanish-language media, particularly Spanish-language radio.

Still Coming North from Mexico   351

Well-known radio personalities such as Renán Almendárez Coello and Eduardo Sotelo, who used the monikers “El Cucuy” and “El Piolín,” respectively, hit the airwaves providing their listeners, mostly Latinos, with information regarding H.R. 4437, immigration reform, and, most importantly, details about the march.54 Radio and television stations in Mexico, whose broadcasts reached Los Angeles and other cities in California, broadcast their support.55 Hometown associations also contributed to the mobilization of the Mexican immigrant community. Fashioned after mutual aid societies, members joined together to send money to assist their hometowns in Mexico, often rural cities and villages, in building schools, roads, churches, and other types of hometown infrastructure. These associations, interestingly often offshoots of soccer clubs, brought immigrants together, increased their solidarity, and, when needed, such as during the spring of 2006, contributed to the mobilization efforts in cities such as Los Angeles and Chicago. Mass demonstrations also developed in cities without long-standing immigrant communities such as Atlanta, Milwaukee, and Omaha. Such new destination cities experienced mass demonstrations and rallies similar to those held in primary destination cities such Los Angeles, San Jose, and New York City, cities with well-established Mexican immigrant communities. These new destination cities stunned the media and, indeed, protesters across the country, as they came together from different community organizations, unions, and churches to protest anti-immigration legislation. In Nebraska, groups such as the Chicano Awareness Center, the Lincoln Hispanic Community Center, the Union of Food & Commercial Workers, and various Catholic Church organizations used their established infrastructure to facilitate the demonstrations. Although immigrant communities were composed of a diversity of backgrounds, based on legal status, nationality, gender, age, occupation, and education, Latinos made up the vast majority of the immigrants. Many new destination cities had only a burgeoning network of immigrant community support networks. Nevertheless, a combination of anti-immigrant sentiments and behavior and, in addition, the passage of H.R. 4437 produced a rising collective solidarity among immigrants with roots in the United States and those recently arrived. As in other immigrant communities, the traditional Spanish-language and print media; Spanishlanguage television news broadcasts, such as those on Univision; and social media, such as e-mails and text messages facilitated the mass participation in demonstrations during the spring of 2006. Demonstrators in new destination cities, like their counterparts in other cities, created a narrative of protest. Protesters carried both American and Mexican flags after television and radio station journalists and pundits criticized marchers for being un-American because they only carried the flag of Mexico. Protest organizers quickly insisted that people carry both flags and display them prominently when the media approached the line of protesters.

352  North from Mexico

In addition, posters carried by the demonstrators contained various messages such as “Hoy marchamos. Mañana votamos” [Today we march, Tomorrow we vote], “We love America,” and “We are hard-working immigrants not criminals.”56 Between March 27 and March 31, Mexican and Mexican-American students staged protest or “blowouts” by walking out of their classrooms and schools. Using blowouts as a protest strategy dates back to the famous 1968 East Los Angeles student blowouts.57 Other blowouts followed in San Diego, Oxnard, and Santa Ana, the latter in conservative Orange County, which had a large Mexican community. Police and other law enforcement officers moved in quickly and arrested, and in some cases deported, student protesters. On April 10, 2006, after the student protests had, for the most part, ended, a mass march and demonstration took place in Los Angeles while other cities, including Chicago, New York City, and Phoenix, participated in their own marches. Together, this national protest action became known as the National Day of Action.58 On May 1, known as International Workers’ Day, a remembrance of the 1886 Chicago Haymarket protests when 35,000 workers walked off their jobs, various groups mobilized and organized a “Great American Boycott” in which some 400,000 marchers joined together and marched to the Los Angeles City Hall. Protesters took to the streets across the country: Atlanta (2,500), Las Vegas (10,000), Los Angeles (1 to 2 million), Milwaukee (70,000), and New York (200,000). Workers, including students, truckers and dock, garment, restaurant, and hotel workers, staged a labor slowdown or complete boycott. Across the country, supporters of the boycott pledged that on this day— International Workers’ Day—they would refrain from making any purchases, attending schools, and working. Despite its success in Los Angeles, some unions and the Catholic Church voiced their disapproval with the boycott arguing that it would only serve to alienate individuals and groups that had significantly contributed to the past marches.59 It seemed to many as if cities across the country came to an almost virtual shutdown, May 1 had become a “day without Mexicans.” A transnational dimension developed when the Catholic Church, guided by Monsignor Renato Ascencio of Ciudad Juárez, Mexico, came to the support of the boycott by Catholics living in along the U.S.-Mexico Border.60 His show of solidarity with the boycott and the solidarity expressed by Mexicans living in Mexico dramatically illustrated the fluid nature of the border, a place where Mexicans lived in communities sin fronteras, communities without borders. The mass demonstrations in the spring of 2006 signaled the crystallization of collective activism triggered by growing xenophobic and racist anti-immigrant sentiments that were being translated into proposed anti-immigration legislation, such as H.R. 4437. Both the English- and Spanish-language media

Still Coming North from Mexico   353

described the marches as a combination of organized and spontaneous political behavior but often overlooked other examples of well-established forms of civic engagement within Mexican immigrant communities. Mexican immigrants, like other immigrants, had been building and participating in various forms of formal and informal political associations. Hometown associations; church groups; immigrant newspapers, such as El Observador in San Jose, California, which publishes stories in Spanish and English; nonprofit organizations; and grassroots organizations attest to a growing civic engagement among Mexican immigrants. Together these organizations became sites of political mobilization that became instrumental in increasing the participation of Mexican immigrants in the 2006 demonstrations. In such types of organizations, specifically hometown associations, Mexican immigrants maintain ties to their communities of origin and, in addition, increase their participation in community activism aimed at improving conditions within their communities in the United States for both immigrants and U.S.-born MexicanAmericans: a “civic binationality.”61

8. Gendered Journeys: Mexican Immigrant Women and Transnationalism The narrative of Mexican immigration to the United States is not complete unless it integrates the gender-specific experiences of Mexican immigrant women whose collective histories have, until relatively recently through the efforts of Mexican-American/Chicana and other historians and social scientists, been subsumed under the umbrella category of “Mexican immigrants and their families.”62 Historical and contemporary studies of immigrant women focus on specific questions to address the fact that “gender matters.” How do women’s lives change if only their partners migrate and they remain in Mexico with or without their children if they have any? Under what general conditions do women take on the journey—with or without children—to come north from Mexico? What type of family structures and power relations between men and women in families are more likely to lead women to migrate? What categories of immigration, such as temporary, permanent, undocumented, are most common among women? How do power relations with families change after arriving in the United States? In sum, “[a]nswering these questions . . .requires showing how a seemingly gender-neutral process of movement is, in fact, highly gender-specific and may result in differential outcomes for men and women.”63 Mexican immigrant women face similar obstacles as their male counterparts but experience the immigration process in different ways. For the most part, Mexican women, particularly those in lower socioeconomic positions, find themselves in traditionally patriarchal families of origin and those of

354  North from Mexico

marriage or partnership. The decision to immigrate is, in fact, a process through which family gender patterns come into play. Most studies focusing on the decision-making process have kept “women in the shadows” largely because of the assumption that males are the prime movers in making the decision to migrate to the United States. Although most studies of Mexican immigrants tend to focus solely on heterosexual unions, a growing research specialization within immigrations studies now examines the experiences LGBT immigrants, specifically their often problematic relations with border law enforcement agencies.64 The experiences of Mexican women who remain in Mexico while their husbands or partners immigrate to the United States and later either return to Mexico or relocate their families to the United States vary but some trends have developed. The likelihood of a Mexican woman immigrating to the United States is based on a combination of individual and societal factors. Historically, Mexican men have been more likely to leave Mexico for the United States with dreams of a better life for their families, who might not accompany them but may eventually join them. Some immigration agreements between the United States and Mexico, like the bracero program (1942–1964), gave rise to a labor force population of men who were contracted as agricultural laborers and unable to work in any other industries. As stipulated in the program, braceros could not bring their families to the United States, although some eventually did with or without documentation.65 Individual factors related to the sole migration of Mexican women include age, birth order, rural/urban origins, marital status, educational status, occupation or skills level, and experiences in the paid labor force. Family factors include family size, age/sex composition, life-cycle stage, family network, and family structure. Societal-level factors include community values and norms as well as gender role expectations. In addition, Mexican women, like Mexican men, make a decision to immigrate as a result of perceived opportunities in the United States with the recognition of the precarious world within which documented and undocumented immigrants live. For example, the demand in the United States, particularly in the Southwest, for Mexican women to work as domestic workers and child care providers has generally remained constant and with it the likelihood that women’s migration will remain an option. An existing network of Mexican immigrant women who are already living in the United States serves as a critical factor in the decision of Mexican women to leave Mexico.66 Thus, “the culture of the sending society [Mexico] determines the likelihood that women in various positions will migrate. . . . [A] woman’s position in the sending community not only influences her ability to autonomously decide to migrate and to access the resources necessary to do so, but also the opportunity she has to migrate at the point when the decision is being made.”67

Still Coming North from Mexico   355

Although many issues confront the daily lives of Mexican immigrant women, one of the most recurrent developments in the study of Mexican immigrant women is that of their experiences as transnational mothers.68 Unlike traditional motherhood, where mothers usually live with their children, transnational motherhood is one in which immigrant women come to the United States but leave one or all of their children in Mexico, usually with their grandparents, relatives, or, in some cases, friends. Thus, transnational motherhood is described as a situation in which “immigrant women who work and reside in the United States while their children remain in their countries of origin constitute one variation in the organizational arrangements, meanings, and priorities of motherhood. . . .[The] meanings of motherhood are rearranged to accommodate these special and temporal separations.”69 In 2012, Mexican women (26 percent) made up the largest group of immigrant women. Ironically, they had the made the least income of all female immigrant groups. Between 2008 and 2012, women were 46 percent of the 11 million unauthorized immigrants living in the United States. California ranks third in having the largest number of unauthorized women. In general, Mexican women and other immigrant women who come to the United States alone seek employment so that they can send remittances back to Mexico to support their families, specifically their children. As a result, they redefine “motherhood” by excluding the traditional view that “mothering” is done in close proximity, such as in a household. Instead, transnational mothers stress that they are not “there” to raise and nurture their children on a day-to-day basis, but they are indeed good mothers because they are “here” working in the United States in order to provide financial support for them in the best way they can. Such immigrant women migrate to the United States in hopes of earning more income than they could ever hope to obtain had they stayed in Mexico. Such a redefinition of motherhood does not outweigh an immigrant woman’s feelings of guilt, despair, and longing for their children. The choice to leave their children in Mexico to prevent them from facing the dangers of crossing the border with them, or at a later time with others, is a heart wrenching one for transnational mothers. It is common for transnational mothers to believe they will only be staying in the United States temporarily until they can save enough money to enable them to begin a better life than they had before they migrated. By seeing themselves as good mothers, because they are indeed providing for their children, transnational mothers attempt to reconcile their predicament of “I’m here, but I’m there.”70

9. The Border: Mexican Women Confront Violence The U.S.-Mexico border will always be a symbol for Mexican immigrants of the possibility of beginning of a new life. Still, for many, the border and its

356  North from Mexico

surrounding area is also a place of violence, specifically violence against Mexican women. Examples of such violence include the largely unsolved murders of hundreds of women along the Juárez-El Paso border, a threaten­­ ing  climate for lesbian Mexican immigrants and the violence endemic in “drop houses” along the border, specifically in Arizona, where undocumen­­ ted  women are held for ransom until a sum of money is paid by family or friends. These same women are often sexually assaulted after crossing the border. Ciudad Juárez, in the northern state of Chihuahua, shares the international border with El Paso, Texas. While the city’s crime rates were already among the highest in Mexico in 1993, events would unfold that caused ripples across both countries and internationally. During the decade between 1993 and 2003, the Juárez murders of women shocked Mexico and the United States. The General Prosecutor’s Office of the state of Chihuahua recorded the deaths of 268 during these years, while the Chihuahua Institute for Women recorded 320 murders during this time period.71 The murders that began in the early 1990s became more alarming when law enforcement, city government, the press, and the city’s residents learned that the murdered women shared common characteristics. Victims were young Mexican women who, for the most part, had been employed in the city’s maquiladoras (assembly plants) and had little or no family in the area. Since the introduction of global assembly plants in Mexico and other parts of the world, workers have been mostly young, unaccompanied women, most of whom had migrated from rural areas. They were vulnerable to the vagaries of a workplace in which the work culture reflected a carryover from the private world: a system of explicit patriarchal dynamics. Although Mexico had long been an industrialized country with women forming part of the formal and informal paid labor force, an ongoing tension within families and society at large simmered below the surface. A traditional cultural belief persisted that questioned the morals of single, independent women, particularly if they were from rural areas, who set out to earn a living on their own. Sensationalized feature stories in the city’s tabloids established a not-so-subtle connection between the murdered women who worked at assembly plants and what was considered “their dangerously non-traditional behavior.”72 The murders are examples of gender-based violence and bear some resemblance to the murders that took place during the prolonged civil wars and military dictatorships in Central and South America in the 1970s and the 1980s when women were targeted for rape. Many believed the Juárez murders were not receiving the necessary attention and investigation by law enforcement and the city government. The president of the Mexican Congress’ Committee on Sexual Equality asserted a commonly held position among feminist organizations on both side of the border: “Authorities haven’t cared

Still Coming North from Mexico   357

because the victims are women and they’re poor, and many times they have no family in Ciudad Juárez.73 The narrative of the murders of young, Mexican women in Ciudad Juárez is constructed out of a combination of many factors: the failure by law enforcement and political figures to develop a more effective and sustained effort to solve these cases, the lack of concern by the general public as the murders occurred less frequently, the growing lack of investigative reporting by the media, the ongoing fear of the drug cartels and, sadly, an ideology of blaming the victim—the murdered women. Juárez was considered the most violent city until 2012 when it was surpassed by San Pablo, a city in northwest of Honduras.74 It does, however, remain the city with the highest rate of domestic violence in Mexico. This border city became notorious as being a place of generalized fear and violence for women both at home and in the workplace. The murders spread to other areas close to Juárez and across the border to El Paso, Texas, Referring both to Mexico City and Juárez, one of Mexico’s representative to the United Nations committee for gender equality stated that “Violence against women isn’t an epidemic, it’s a pandemic in Mexico.”75 In addition to the murders, about 4,000 women disappeared in Mexico in 2011–2012, mostly in Chihuahua and the state of Mexico, according to the National Observatory Against Femicide. Studies have shown that violence against women increases in those areas where drug wars escalate. According to a study by Mexico’s National Commission to Prevent and Eradicate Violence against Women, the number of women slain increased dramatically between 2001 and 2010 in northeastern Mexico, an area of intense rivalry and war between drug cartels. Human rights groups say security forces are often involved in sexual abuse and disappearance of women. One-fifth of all murders of women in Mexico in 2012 took place along the U.S.-Mexico border, particularly in Ciudad Juárez where hundreds of women were murdered in the 1990s. In all probability, these statistics represented an undercount as many murders go unreported.76 Even as the murder rate declined beginning in the late 1990s, human rights organizations, the United Nations, a wide range of nongovernmental organizations established national, state, and local feminist organizations and other grassroots organizations that have argued that these violations are genderspecific and violations of human rights. In several reports, the Inter-American Commission’s investigation listed several reasons the commission railed against Mexican authorities and their responses to the murders. First, the Mexican government failed to conduct an expeditious investigation when the murders began. For example, the press reported a case when a woman called the local police to report a woman screaming near a drainage canal, but police did not respond. Similar reports surfaced soon after this event. Although many of the women who were eventually murdered had been listed as missing by

358  North from Mexico

friends and relatives, police delayed their investigations or, in many cases, even failed to act on such reports. At another level, although state officials in Chihuahua arrested some person or persons believed to be responsible for some of the murders, critics soon pointed out that many of them were never prosecuted. The Mexican National Commission on Human Rights called for a review of both administrative and criminal investigations. Lastly, human rights organizations and others criticized authorities for being less than forthright with the families of the murdered women. International pressure over the tide of killings persuaded Mexican lawmakers in 2007 to approve new legislation aimed at preventing violence against women. It formed a national committee to investigate the killings and developed prevention programs for domestic violence, condemning and recognizing the relationship between the two types of violence against women. Due to many factors, including insufficient and sustained funding, the committee fell short of its intentions.77 A subject for more research on violence against women involves the experiences of Mexican lesbians living in the borderlands, where race, ethnicity, social class, gender, and sexual orientation collide. The murdered women were often labeled as lesbians who were engaging in dubious and “sinful” behavior. One researcher explains that Mexican lesbians living along the border “travel” through various geopolitical spaces, navigating along the way, always hoping to reach a point where their self-identification as lesbians, is a space free of violence.78 Not only do undocumented lesbians face deportation they must also deal with a climate permeated by homophobia, one that can easily translate into violence. In addition, these immigrant women often maintain some ties with their families who have not left Mexico. As they gain more sexual autonomy, the price is often a deep rupture with their families in Mexico. Those that knew of their daughter’s sexual orientation usually did not acknowledge this within the family since discussion about sexuality remains a difficult one within Mexican families. Silence is substituted for open discussions surrounding female/male sexuality. Nevertheless, research shows that Latina lesbians, immigrant or not, have created “imagined communities” in which they join together to share their lived experiences, struggle, survival skills, and strategies for publically confronting anti-gay immigration laws and policies: a “space that serves as a home to the marginalized.”79 The story of the Ciudad Juárez murders is not complete without a discussion of the mobilization efforts and activism of Mexican women to raise awareness of the murders and the subsequent flawed investigations. Women’s activism in Latin America has taken many forms: protests, rallies, hunger strikes, petitions, labor strikes, and the formation of grassroots organizations, many of which eventually developed into formal organizations. Women leaders emerged as women’s activism spread through regions and countries in Latin America: for example, Rigoberta Menchú in Guatemala and Elvia

Still Coming North from Mexico   359

Alvarado in Honduras.80 From the beginning, existing Mexican feminist organizations, such as those that focused on improving women’s health and protecting reproductive rights, took up the fight against violence against women. Some groups had already been functioning as lobbying groups for women’s rights but soon turned their specific attention to the Juárez murders, calling for increased investigations and prosecutions. In 1997, a group called the Citizens Committee against Violence worked with other women’s organizations and together they drafted a letter of protest to the governor of Chihuahua. Their demands called for the creation of a fully funded and staffed special prosecutor’s office, the establishment of community programs to address domestic and workforce violence, and the development of mechanisms whereby law enforcement agencies keep the families of the murdered women updated on the status of the investigations. After a surge of murders in 1998, the mothers of some of the victims started the group Voces sin Eco (Voices without Echo) and later, as more murders were discovered in 2002, other mothers started the group Justicia para Nuestras Hijas (Justice for Our Daughters) in an attempt to pressure authorities to increase their investigations. Similar groups emerged and together they bore witness that the women had succeeded in mobilizing to confront these tragic cases.81 Noted filmmaker and director Lourdes Portillo produced Señorita Extraviada, a documentary that brought to life the horrific murders, almost all of which remain unsolved but not forgotten. Portillo describes her goal in producing this film: “With over 270 girls raped and killed and another 450 missing, we felt we had to investigate these disappearances and attacks specifically directed toward young, brown, unprotected, poor women. This film is mostly about deciphering the silence [surrounding these murders]. . .putting an end to the terror.”82 The development of “drop houses” represents still another example of the gender-specific violence experienced by unauthorized Mexican women who make the dangerous border crossing into the United States, primarily along the Arizona border with Mexico. For the most part, the Department of Homeland Security’s policy of increased border security produced an unexpected outcome—the proliferation of such “drop houses.” Drop houses are also prevalent in other cities, including Houston, San Diego, and Los Angeles. The intensification of such border security and local law enforcement has been exacerbated in Arizona’s Maricopa County, where mass “criminal sweeps” took place. Arizona passed a series of measures aimed at intensifying the state’s “zero tolerance” for those crossing the border into Arizona without documents. Proposition 200, passed in 2004, required social service agencies to check a client’s immigration status. Other measures followed in the next couple of years. For example, H.B. 2008 required proof of citizenship for anyone applying for federal, state, or local benefits. In addition, S.B. 1070, perhaps the most vilified bill, provided that law enforcement agencies could stop and

360  North from Mexico

detain anyone to ask for proof of legal residence. It led to nationwide protests and demonstrations. As with other state bills, like California’s Proposition 187, the U.S. Supreme Court overturned the bill based on the right of the federal, not the state, government, to regulate immigration. Nevertheless, a virulent anti-immigrant climate flourished in Arizona that, as in the past, demonized all immigrants and people of color who were automatically assumed to be not only immigrants but also undocumented ones. With increased border security checks along traditional crossing points, such as El Paso, Texas, undocumented migrants started crossing the border at isolated places in the desert where migrants hoped that the likelihood of apprehension was less than in the usual border crossings. The result proved tragic as almost 2,000 bodies have been found in the desert and areas in close proximity since 2005. Drop houses developed as more migrants, in an attempt to avoid desert crossings, turned to paying organized smugglers. Generally, migrants pay about $5,000, a longstanding practice, but one that has become more expensive in recent years. This fee includes being brought across the border by smugglers, usually referred to as coyotes, and “dropped” at a house for pickup by relatives or friends who are forced to pay a ransom in order to have their relatives, mostly women, released. In many cases, payment of ransom still did not secure release.83 While trapped in drop houses, undocumented Mexican women become victims of violence, including torture, rape, and, in some cases, murder. A generalized fear of escaping is not an option for these women because they fear deportation or violence, even murder, at the hands on their kidnappers. Sexual assault or the threat of sexual assault represents a frequent source of terrorization of migrant women who are forced into drop houses. Those women whose relatives pay the ransom for their release or the few who manage to escape, also face social services that “victimize” them for a second time. Understaffed and underpaid, many social service workers have been responsible for “blaming the victim,” in this case those migrant women who report that they were sexually assaulted while at drop houses.84 All this contributes to a “culture of cruelty” that makes the journey in search of a better life north from Mexico a precarious one for all Mexican immigrant women, and, indeed all Mexican immigrants.

22

Mexican Immigrants in a Changing Society Alma M. García

Mexican immigration, unlike immigration from other countries, has been one of continuous replenishment, a fluid one in which the movement of people across the U.S.-Mexico border may wax and wane but never comes to a full stop. The epic story of Mexican immigration to the United States, like the story of other immigrants, is incomplete without an understanding of the unfolding dynamics of second and later generations of Mexican-Americans. Mexican immigrants and their U.S.-born children experience conflicts and adjustments as they navigate through a society in which “nationality and citizenship matter” as well as race, class, gender, sexual orientation, religion, age, and other critical variables. First- and second-generation Mexicans in the United States stand witness to a variety of social and public policy issues dealing with both documented and undocumented immigration and political participation, specifically in presidential elections. A common theme runs through a discussion these issues: the process through which U.S. society reimagines itself as a nation of immigrants.

1. Children of Mexican Immigrants: The Second Generation Children of immigrants, also referred to as the second generation, face a paradox: being Mexican but not from Mexico, and being “American” but children of Mexican immigrant parents. Many of the research studies on the second generation have focused on ethnic identity formation. Children of immigrants and later generations face an evolving sense of self that is constructed from a combination of factors: the timing of their parents’ immigration to the United States; the communities within which they were raised; the maintenance of their parents’ cultural traditions, the adoption of new cultural traditions, and the tension between the two; relations with other immigrant and racial/ethnic

362  North from Mexico

groups they encounter in their communities; and the personal impact of social structural factors, such as social class inequality and access to quality education, including their interaction with students and teachers.1 As a result of this ongoing interaction, research studies of the identity formation of the second generation are informed by the idea that “the construction and adoption of a racial and ethnic social identity. . .[is] an ongoing negotiation between self and other identifications which reflects” who they are, how others see them, and their reaction to how other see them.2 As such, second-generation Mexicans struggle with an evolving, fluid, and often ambiguous sense of identity. Estela, a Mexican-American university student, daughter of Mexican immigrants, describes this struggle: The only reason I call myself Mexican is because that’s the only way that I am absolutely sure of who I am. I’m Mexican because my parents came from Mexico, but I was born here [the United States], but I still am Mexican. My parents remind me all the time that I am Mexican. Yes, I speak to them in Spanish, but I prefer speaking English, but I mean one thing that I am sure of is that I’m Mexican, sort of.3 (italics in the original)

Estela’s description of her identity illustrates a theme that cuts across narratives of other second-generation children. She identifies her sense of self as one shaped by many nuances colored by the recognition that she is both similar to but different from her parents. Second-generation individuals live in two social worlds that collide but often blend through a process of “negotiated boundaries.”4 According to García, “Second-generation Mexican-Americans construct their ethnic identities by (re)negotiating the social boundaries between themselves and their immigrant parents and between themselves and the larger U.S. society.”5 Several research questions tap into the second generation’s sense of self. First, what meanings do second-generation and later individuals attach to their ethnicity and their family’s immigrant past? Second, do women’s and men’s experiences as children of immigrants differ? Third, how do the secondgeneration individuals navigate between their immigrant family backgrounds and the larger society, which consists of second generation and later generations of individuals from different immigrant backgrounds? Fourth, how does such social class shape their experiences? Fifth, what language barriers exist, particularly between the immigrant and second-generation family members? These and other research questions provide insight into the experiences of children of Mexican immigrants, insights that will have public policy implications. For example, there are public policy implications for citizenship in light of the recent movement to repeal the birthright citizenship clause of the Fourteenth Amendment. Reflecting anti-immigrant public attitudes and

Mexican Immigrants in a Changing Society   363

political sentiments, a movement to eliminate this part of the Fourteenth Amendment is shaped by the following question: Do U.S.-born children of undocumented parents have the right to become citizens of the United States? Without an amendment to the Constitution, only one answer exists: yes. Immigration patterns of Mexican immigrants, including time, place of entry, social class, and education have produced families with mixed legal status in which foreign-born children are being raised alongside siblings who were born in the United States. Stories from second-generation Mexican-Americans provide some answers to a specific question: What meanings does the second generation attach to their ethnicity and their family’s immigrant past? María, a university student who is the daughter of Mexican immigrant parents recalls how her parents left a small village outside Durango in the state of Durango, Mexico, to join other family members who had made the trip to El Norte a few years ago. After crossing the border without documentation, her parents eventually joined her uncle and his family in San José, California. Her father had completed three years of elementary school in Mexico, but eventually, when he was sixteen, found work at a local construction company. He married and decided he wanted his children to have a better life in the United States. Soon after his arrival, he joined his brother, who worked as a nonunion construction worker and received his weekly pay “under the table.” María’s mother stayed at home and raised María and her two sisters. Her mother never entered the paid labor force except for some occasional child care for her neighbors. María recalls the times her parents told her siblings and her that they missed their lives in Mexico, but they remained in the United States so that she and her sisters would have a better life, one in which they would be educated and might even become teachers. María relates that her family, her neighborhood, and her friends were all “Mexicans,” and she heard Spanish spoken “everywhere” in her neighborhood. On the one hand, the message she received from her parents was that their lives in the United States would be better than what they would have experienced if they had remained in Mexico. On the other hand, her mother told them many times that even though she and her sisters had been born in the United States they would inevitably be seen as “diferente” [different] by “Americanos.” Interestingly, her daughter said she would tell her mother that she and her sisters were in fact “American”: My mom would tell me that she had heard that Mexicanos had problems here in the United States. One of her cousins returned to home from Mexico and told her that he made a lot of money but also told her that Mexicans are not treated well.6

Diego, son of Mexican immigrant parents from Jalisco, Guadalajara, Mexico, said that his parents also told him that “Americanos” treated “Mexicanos” as if

364  North from Mexico

they were inferior to them. Ironically, when his parents took Diego to visit their relatives in Mexico, he said that his cousins criticized him for being “too American.” Another second-generation young woman related the story about the time her “Mexican aunt” came to visit. I remember this time that my aunt [paternal] came to visit. Before she got here, I heard my mom talking to my dad and sort of complaining. She [her mother] was saying that his sister was always on her case because he had come to the U.S. and had forgotten Mexico and he hadn’t taught his kids that they were Mexicans. So my dad said that he was going to tell his sister that his kids were now born here [the U.S.] but they still knew that they were also Mexican. I guess he wanted my aunt to know that we were born here but we were also Mexicans, but also American.7

Similar stories are found in research studies on second-generation individuals from other immigrant groups, such as Chinese-Americans, VietnameseAmericans, and Japanese-Americans.8 In their search for some kind of identity, the second generation confronts a combination of contradictions as they try to establish their sense of self in juxtaposition to their immigrant parents. Carmen describes her dilemma of trying to make “sense out of all this.” When asked to describe her feelings about being a second-generation Mexican Carmen offered the following: My parents wanted me to be American, but still know that I am Mexican. I have to admit that sometimes it’s pretty confusing. It didn’t always click that I could be both. The way I see it, my parents are Mexican because they were born over there [Mexico] and then came here [United States]. Pretty simple, but then what am I? I was born here so I am American but what about the Mexican part of me? It is so confusing. I know that I am proud of my parents and they are Mexican or Mexicano as they say, because they were born in Mexico. I was born here in San José, California, but I also am Mexican. I am different from them, not better. It’s confusing, don’t you think?9

For Mexican-Americans, the process of blending the “American” with the “Mexican” is a balancing act, one that other second-generation children of immigrants experience. Second- and later-generation Mexican-Americans commonly emerge with a multilayered sense of ethnic identity that facilitates moving between the two worlds of their immigrant background and their roots in the United States. Latina creative writer Pat Mora eloquently captures the identity dilemma of being Mexican-American in her poem, “Legal Alien”:

Mexican Immigrants in a Changing Society   365

Bi-lingual, Bi-cultural, able to slip from “How’s life?” to “Me’stan volviendo loca,” able to sit in a paneled office drafting memos in smooth English, able to order in fluent Spanish at a Mexican restaurant, American but hyphenated, viewed by Anglos as perhaps exotic, perhaps inferior, definitely different, viewed by Mexicans as alien, (their eyes say, “You may speak Spanish but you’re not like me”) an American to Mexicans a Mexican to Americans a handy token sliding back and forth between the fringes of both worlds by smiling by masking the discomfort of being pre-judged Bi-laterally10*

Immigrants and their children and later generations will continue to transform U.S. society. As a result, basic demographic data provides a key to understanding these immigrant experiences that educators, politicians, and all society will need to know to understand this sector of society and its contribution to increased diversity in American society. The number of children aged 17 years and younger who have immigrant parents has been increasing at a high rate. In 1990, the number of children of immigrants stood at about 8 million but increased to 13 million in 2000 and reached 17.4 million in 2013. This represented about 25 percent of the 70 million children aged 17 years and younger in the United States.11 The total number of second-generation children accounts for 88 percent (15.3 million) of all children with immigrant parents. This is significant because this group of U.S.-born children represents future eligible voters. Equally important for policy makers is that of the 31 million children living with parents whose income classifies them as below the poverty level, almost 9.5 million are children living with immigrants. These demographic features will require major adjustments within the social service * Reprinted with permission from the publisher of Chants by Pat Mora © 1985. Arte Publico Press, University of Houston.

366  North from Mexico

and educational system in order to best serve this group of children living in poverty with at least one immigrant parent. Children living with immigrant parents cluster in five states: California, Texas, New York, Florida, and Illinois with Mexican immigrants representing the largest immigrant group in these five states. Together these states account for 58 percent of this population. Studies of the second generation focus on many other demographic features: ages of children and parents, year in which the parent arrived in the United States, parents’ and children’s educational level, and parents’ occupation and income.12 Level of educational attainment is one of the major areas of research in studying the lives of U.S.-born children of immigrants. Generations of Mexican-Americans have experienced problematic relations with the American educational system. Both de jure and de facto educational segregation have blocked the upward mobility of second- and later-generation Mexican-Americans.13 Mexican communities have an equally marked history of struggling against segregation to ensure that their children will have access to their vision for a better life: the American Dream. For example, on May 17, 1984, thirty years after Brown v. Board of Education, parents in San José, California, filed a lawsuit against the San José Unified School District. The case alleged that the school district was operating an unconstitutionally segregated public school system. Parents won the case in 1984, exactly thirty years after Brown v. Board of Education, and, as a result, the city was forced to implement a desegregation policy. The San José decision came two years after the 1982 Supreme Court case of Plyler v. Doe that established the constitutionality of providing public education to undocumented children.14 Despite community activism to reform the educational system, major roadblocks continue to exist.

2. Undocumented Students: Court Cases and Legislation Undocumented Mexican immigrants make the journey to El Norte with their young children who, with their parents, begin a new life in the United States. Unlike their parents, these young children leave Mexico at such an early age that lasting memories of their parents’ country are largely nonexistent, particularly if they arrived at the age of four years or younger. They grow up in Mexican homes and communities, attend school, graduate from high school, and, like their parents, dream about a better life in the United States. For these undocumented children, a college education represents the American Dream turned into a reality. However, they soon confront another reality: the persistent threat of deportation even if they have lived in the United States for most of their lifetime. It is estimated that in 2000 approximately 1 million undocumented youth found themselves in such a distinct immigrant category. They

Mexican Immigrants in a Changing Society   367

came to the United States as very young children without documentation but still possess the constitutional right to a free public school education as guaranteed by the 1982 landmark Supreme Court case of Plyler v. Doe. During the 1970s an increase in unauthorized Mexican immigration to Texas led to a heightened anti-immigrant backlash. The rising numbers of undocumented children attending public school in Texas raised the question of whether or not these children had legal rights to a public education. Texas passed a state law in 1975 that allowed public schools to charge tuition to parents with undocumented children attending schools. Many school supervisors had long been in support of such legislation. In the town of Tyler, Texas, Superintendent James Plyler began charging $1,000 tuition per year per child. Working with local community activists, a lawyer for the Mexican American Legal and Defense Fund (MALDEF) filed suit on behalf of the parents who were being forced to pay tuition. In 1978, the judge ruled the law unconstitutional because it violated the Fourteenth Amendment’s equal protection clause. The Circuit Court of Appeals upheld this decision and, ultimately, in 1982, the U.S. Supreme Court declared the Texas law unconstitutional based on similar grounds. In his writing for the majority, Justice William Brennan described the Texas law as one that imposed “special disabilities upon groups disfavored by virtue of circumstances beyond their control. . . . [suggesting] the kind of ‘class or caste’ treatment that the Fourteenth Amendment was designed to abolish.”15 The period after Plyler v. Doe witnessed attempts to nullify the landmark case. In 1996, Congress passed the Illegal Immigration Reform and Immigrant Responsibility Act, which provided for stricter immigration enforcement protocols and limited the type and amount of government aid that undocumented immigrants could receive. Both laws stipulated that states could not allow undocumented students in postsecondary schools to pay in-state tuition. In addition, both laws denied federal financial aid to these students. In 1996, Congress also passed the Personal Responsibility and Work Opportunity Reconciliation Act, which barred undocumented immigrants from receiving certain federal benefits. Both of these contained sections that would affect the undocumented, specifically school-age children. Following this pattern, Representative Elton Gallegly (R-CA) introduced a bill “to deny public education benefits to certain aliens not lawfully present in the United States.”16 The Gallegly amendment drew swift opposition from several sources that succeeded in ultimately forcing its withdrawal from existing legislation. President Bill Clinton stated that he would veto any legislation with such an amendment that would overrule Plyler v. Doe. Many politicians, educators, and community activists agreed that “by 1996, after more than a decade of living with Plyler, educators had made their peace with its requirements and had come to accept the decision and contributed to educating these children.”17

368  North from Mexico

As a result of Plyler v. Doe the number of undocumented students completing high school increased and with it the number of students seeking higher education. The Plyler v. Doe decision did not apply to post–high school education, however. This set the stage for the enactment of policies at public universities aimed at undocumented high school graduates. Policies varied from state to state. In many states, unauthorized students, just like international students, were required to pay out-of-state tuition fees. California, Illinois, New York, and Texas allowed unauthorized students to establish residency in order to pay the much lower rates of in-state tuition. Although undocumented university students were able to pay in-state tuition, economic constraints made it difficult for them to take advantage of these policies. Undocumented students were also not eligible for federal and state financial aid. Although technically unable to work given their immigrant status, students often found work in the informal labor market where workers are often paid in cash. The revised university tuition policies facilitated the enrollment of undocumented students but could not address the students’ overarching problem: their undocumented legal status, which caused them to live in the shadows where they and their undocumented families lived in fear of deportation. The 2001 Development, Relief and Education for Alien Minors Act (DREAM Act) was based on the possibility of ending this precarious legal status. This bipartisan piece of legislation addressed the distinct status of children who, at a very young age, accompanied their parents who came to the United States without documents. This legislation echoed the sentiments expressed by Supreme Court Justice Brennan who wrote the majority opinion in the Plyler v. Doe: “Education has a fundamental role in maintaining the fabric of our society. We cannot ignore the significant social costs borne by our Nation when select groups are denied the means to absorb the values and skills upon which our societal order rests.”18 In 2001, Orrin Hatch (R-UT) and Dick Durbin (D-IL) introduced the DREAM Act, which called for the repeal of half of the Illegal Immigration Reform and Immigrant Responsibility Act. The DREAM Act’s goal was to extend the rights provided under Plyler v. Doe to include postsecondary education benefits to undocumented university students. It would make it possible for states to provide educational benefits such as in-state tuition to undocumented students. Most significantly, the DREAM Act would provide a path to citizenship for youth who came to the United States as minors when their parents entered the United States without authorization. Several requirements would have to be met before the undocumented youth would be eligible for a conditional lawful permanent resident status. The DREAM Act contained certain conditions that had to be satisfied before undocumented youth could be eligible for lawful permanent resident. (1) younger than sixteen years old when they arrived to the United States with their parents, (2) younger than

Mexican Immigrants in a Changing Society   369

thirty-five years old when the act was passed, (3) had been living in the United States for an entire five-year period before the act’s passage, (4) had graduated from high school or had received a general equivalency diploma (GED), and (5) demonstrated good moral character. Undocumented students who met these requirements could apply for a six-year conditional legal permanent status, permitting them to work, travel, join the military, and attend school. If, within these six years, students completed two years of a four-year degree program, graduated from a two-year community college, or served at least two years in the military they would receive permanent resident status and, as a result, would be eligible for citizenship. Although these students could not receive federal grants, they would be eligible for the federal work study and student loan programs. They would also qualify for in-state tuition and financial aid from states. This legislation echoed the sentiments expressed by Supreme Court Justice Brennan who wrote the majority opinion in Plyler v. Doe: “Education has a fundamental role in maintaining the fabric of our society. We cannot ignore the significant social costs borne by our Nation when select groups are denied the means to absorb the values and skills upon which our societal order rests.”19

3. The DREAMers: Struggles and Mass Mobilization Undocumented youth poured into the streets, occupied the offices of. . .leading politicians, filled up blogs and editorial pages with eloquent arguments, lobbied senators and White House officials, and worked their networks to gain the backing of some of the most powerful unions and rights associations in the country. Their immediate goal was to pressure the Senate to support the Development, Relief and Education for Minors Act (DREAM Act).20

Spanning almost a decade, the political debates over the DREAM Act, combined with the general public debate over undocumented immigrants, produced a mass mobilization of undocumented youth— the DREAMers. The immigrant narrative of the DREAMers is one of aspiring dreams, legal constraints, and, ultimately, a mass movement to overcome the barriers that had turned hopes for attaining the American Dream into an American nightmare. Through their collective efforts and mobilization strategies, DREAMers reshaped the “immigrant question” by focusing on their own specific legal dilemma. The DREAMers share a particular journey to El Norte: Their parents had decided to come to the United States with their children, many of whom were younger than eighteen. So, as the DREAM Act languished in Congress, the DREAMers emerged, cautiously at first, then with all the collective energy

370  North from Mexico

and visibility of other mass movements, immigrant or not. Their mantra became “undocumented, unafraid, and unapologetic.”21 Undocumented youth came together to form a collective presence and speak with a collective voice with mobilizing and organizational skills provided by existing immigrant rights’ organizations. Students and the various organizations that came to their assistance adopted several strategies to raise a collective voice in the public arena. Using a public relations approach, DREAMers crafted slogans that would identify undocumented youth as hardworking children and young adults whose legal status needed to be changed. They wove a narrative that their “illegality” was not of their own making. On the contrary, they had entered the United States with their parents at a young age and, therefore, could not be held responsible for their undocumented status. These undocumented youth saw themselves as “American” and had “the right to have rights.” Immigrant rights organizations portrayed these undocumented youth as deserving of legalization. Immigrant rights associations, such as the National Immigration Law Center, the Center for Community Change, and others had been working to support the passage of a DREAM Act and, as part of these efforts, provided support and established infrastructure for undocumented youth. They were “responsible for introducing the issue of undocumented college students into Congress, deciding what strategy was right for the youth, crafting and controlling their representation in the public sphere, and representing them directly to public officials and the media.”22 Eventually, groups of undocumented youth would, in turn, build on their own organizational infrastructures and coalitions. By 2005, national, regional, and local organizations of DREAMers had emerged on the political scene. Immigrant youth–led organizations, with the assistance of immigrant rights organizations, adopted several mobilization strategies. First, undocumented youth organizations used the American flag and other national symbols to stress that they considered themselves de facto citizens not “illegals.” Youth organizers also used pictures or made references to the Statue of Liberty to provide further evidence of their national allegiance. When they told their family histories to the media, these DREAMers went to great lengths to position themselves as “ordinary American kids” deserving of a chance to fulfill their dreams—beginning with citizenship. The message was simple: “I didn’t ask to be brought here [the United States]. I was brought here [by my undocumented parents].”23 The undocumented student movement countered the image of them as “illegals” with an image of hard-working, intelligent students who would contribute to U.S. society once they graduated from college and began their careers. The DREAM Act would enable them to fulfill their dreams. Supporters of the DREAM Act adopted the theme of “through no fault of their own” to show their solidarity with the undocumented youth.

Mexican Immigrants in a Changing Society   371

The undocumented youth movement became a national one with the emergence of youth-led organizations throughout the United States. The United We Dream (UWD) organization reached a membership of 100,000 youth and had fifty-five affiliate organizations in twenty-six states. UWD started in the mid2000s when the National Immigration Law Center proposed a coalition of support for the DREAM Act. By 2005, groups for undocumented students could be found throughout the United States: the New York State Youth Leadership Council, the Student Initiative Movement in Massachusetts, the University Leadership Initiative in Austin, Texas, and the California Dream Network. Two years later, in 2007, the DREAM Act failed to pass. The organizations continued to meet but were forced to reassess their mobilization strategies, expanding efforts to achieving immigration reform in general, not only through the passage of the DREAM Act. In 2008, working with National Immigration Law Center, leaders from seven youth organizations across the country met in Washington, D.C., to discuss mobilization options, strategies, and ideologies. The formation of UWD signaled the beginning of efforts to create a sustainable infrastructure in the struggle for immigration reform for the DREAMers. From 2009 to 2012, UWD members met in various states and agreed to a regional organizing structure consisting of eight regions: New England, Mid-Atlantic, Southeast, Mid-South, Midwest, Mid-North, Northwest, and Southwest. By 2012, UWD represented the major organization fighting for the legal status for undocumented youth. The formation of the UWD program’s Queer Undocumented Immigrant Project confirmed the organization’s mission to value diversity, including sexual orientation, within its movement, by addressing the specific needs of undocumented lesbian, gay, bisexual, transgender, and queer immigrants. In addition to public protests and lobbying efforts, UWD provides a variety of programs, instructional videos, advocacy tool kits and a news blog on its Web site.24 The UWD’s overarching mission is to provide “a stepping stone for immigrant youth to become politically active, to gain exposure to different streams of social justice work, and to share responsibility for building a movement based on principles of social inclusion and justice.”25 Congressional opposition led Hatch and Durbin to submit several DREAM Act bills, but they never reached the Senate floor. The legislative battles to pass the DREAM Act continued over the years.26 In 2010, editorial pages in newspapers throughout the United States demonstrated support for the DREAM Act: La Opinión, “The Time is Now! (November 16, 2010); Los Angeles Times, “A Path to College” (November 17, 2010); Wall Street Journal, “A Worthy Immigration Bill” (November 29, 2010), and New York Times, “Dreaming of Reform” (November 30, 2010) to name only a few. Ultimately, the Senate voted down the most current version of the DREAM Act.27 Wearing graduation caps, students sat in the gallery watching the act’s defeat. Most of these students represented a variety of student organizations from across the country created

372  North from Mexico

to mobilize students. Lorella Praeli, a student at New Haven’s Quinnipiac University who had not been eligible for a fellowship because of her immigration status, expressed her disappointment that the Senate had failed to pass the bill: “Some senators proudly put their thumbs down, and with each thumb down I felt like my heart was a tighter knot, All I could think was ‘can I continue to live like this?’”28 In 2013, at the end of the 112th Congress, the DREAM Act had still not passed. National polls continued to show that 70 percent of Americans support this type of legislation. Approval ratings were even higher among Hispanics and Asian-Americans. Most significantly, many states have initiated their own DREAM Act. Fourteen states have passed laws under which undocumented high school graduates can pay in-state tuition: California, Connecticut, Illinois, Kansas, Maryland, Nebraska, New Mexico, New York, Oklahoma, Texas, Utah, and Washington.

4. The 2008 Presidential Race and Latinos Although not able to vote, Mexican immigrants have been exposed to U.S. politics at different levels. Many Spanish-language newspapers and radio news programs have covered elections in regions with a large immigrant population. Moreover, children of immigrants have been exposed to U.S. politics through their U.S. history, current events, and civic classes at school. For recently naturalized Mexicans, voter registration efforts, such as the Southwest Voter Registration Project in Texas, have canvased Mexican immigrant communities in door-to-door drives. Volunteers set up voter registration booths in parks, church parking lots, community centers, and other nontraditional venues. In varying degrees, presidential candidates needed to assess how they would fare among Mexican-American voters. The voter turnout rate and voting patterns of naturalized Mexican immigrants and U.S.-born Mexican-Americans is complicated by many factors. The U.S. Bureau of the Census, the American Community Survey, the Pew Research Center, and the Migration Policy Institute represent major sources of information regarding voting patterns but do not always disaggregate the category of “Latino” by national origin, such as Mexican or Mexican-American, the largest group within the Latino category. Studies of ethnic electoral politics have shown that both Latinos and Asians lag behind what would be expected given their population growth. Three measures are used to compare and contrast groups of voters. First, eligibility refers to the share of the group eligible to vote according to age and citizenship. Second, registration refers to the share of eligible voters who actually register to vote. Third, turnout is measured by the number of registered voters who actually vote.29 Demographic and political factors represent general predictors of voting behavior. The number of Latinos,

Mexican Immigrants in a Changing Society   373

primarily Mexicans, who are younger than eighteen, and therefore ineligible to vote, is higher than the general population. In addition, Mexican immigrants have the lowest rates of naturalization among immigrant groups. Only 38 percent of Latino immigrants eligible to become U.S. citizens were naturalized by 2000, in contrast to the 60 percent of other eligible immigrants. Of the 12 million undocumented in the United States, approximately 6 million are Mexican, representing an untapped source of potential voters. It was estimated that about 1 million Latino votes could have been added by the 2004 election if 60 percent had become U.S. citizens. Naturalized citizenship did not translate into voter registration rates. Patterns indicate consistently low voter registration rates at any age. Similarly, Latinos who do register to vote are less likely to vote than other groups. Latino activist groups have targeted Latino communities in an effort to raise naturalization, registration, and voter turnout rates. It is estimated that the pool of potential Latino voters will reach 50 percent by 2025, a number that will most likely prove instrumental for increasing the impact of the Latino vote.30 As such, “twenty-first century American politics will be shaped, in large measure, by how Latinos are incorporated into the political system. . . .Latino population growth will combine with growth in the Latino electorate to present both political parties with new opportunities in their approaches to Latino voting.”31 The Democratic Party primaries leading up to the 2008 presidential election illustrate the importance of the Latino voting bloc. Both Hillary Clinton and Barack Obama developed outreach measures, including voter registration, aimed at Latino voters throughout the United States but particularly in states holding early primaries with a large Latino, specifically Mexican, population. Hillary Clinton surged early in gaining Latino support for many reasons. While addressing Latino audiences, she frequently referred to her work in voter registration drives in Mexican-American communities, stressing her personal interaction with Mexican immigrants and Mexican-Americans. Mexicans transferred their long-standing support and approval for Bill Clinton to Hillary Clinton. Moreover, Hillary Clinton gained the support of more Latino elected officials than any other Democratic or Republican candidate, including Los Angeles mayor Antonio Villaraigosa and Henry Cisneros, former HUD secretary and mayor of San Antonio. Dolores Huerta gave Clinton her personal support and the endorsement of the United Farm Workers of America. Obama experienced a disadvantage in winning the support of Latino elected officials largely due to their limited numbers and their overwhelming early endorsement of Hillary Clinton. Nevertheless, during the course of the primary, Obama’s favorability ratings almost equaled Clinton’s. His campaign worked as hard as Clinton’s to gain Latino support. Together both campaigns spent tens of millions of dollars in Latino outreach and mobilization efforts.32

374  North from Mexico

Bill Richardson, governor of New Mexico, became the only Hispanic, specifically Mexican-American, to enter the Democratic primaries. Richardson’s political experience included his service as a congressman, governor, former secretary of energy, and ambassador to the United Nations. Richardson was also well known for his role in securing the release of U.S. hostages from Cuba, North Korea, and the Sudan. He withdrew from the primaries after he finished fourth in both the Iowa (2 percent) and New Hampshire (less than 5 percent) primaries. He had hoped to do well in primaries in Nevada and California. Richardson endorsed Barack Obama over Hilary Clinton.33 A Pew Research Hispanic Center report listed several findings in their analysis of the Latino vote in the 2008 Democratic primaries. Latinos favored Clinton over Obama by a two to one margin. Latino voters played an important role for Clinton in the mega primaries of California and Texas. In both states, Latino turnout vote increased dramatically from the 2004 primaries. Clinton beat Obama in the Super Tuesday primaries 63 percent to 35 percent and in the Texas primary 66 percent to 32 percent. Latino voters in the Super Tuesday primaries were younger that all other voters. More than one in five Hispanic voters on Super Tuesday were 17 to 29 years old. Latinos across genders, ages, educational levels, and incomes cast their votes for Obama on Super Tuesday. Latinos who voted for Clinton cited several reasons for their support: the economy, the continued problems with race relations, and women’s persistent inequalities at all levels of society.34 Once Obama won the presidential nomination, 75 percent of the voters who said they had voted for Hillary Clinton in the primaries became Obama supporters.35 On November 3, 2008, Barack Obama won the presidential election beating John McCain and becoming the first African-American president. Across the United States and the world front-page headlines captured this historic moment: “Yes We Can,” “Change Has Come,” “History Maker,” “Obama: Racial Barriers Falls in Decisive Victory,” and “Yes He Did.” For many Latinos, particularly those voting for the first time as a result of their age or newly gained status as U.S. citizens, the 2008 presidential election marked a turning point. They had joined in the effort that resulted in an African-American becoming the forty-fourth president of the United States. Obama’s victory led to celebrations in Latino communities throughout the United States and Latin America. Supporters chanted “Sí Se Puede/Yes We Can,” the slogan adopted by Barack Obama from César Chávez and the United Farm Workers.36 Assessing the “Latino vote” became one of the major postelection narratives. According to a report published in 2009 by the Pew Research Center, the electorate in 2008 presidential election proved to be one of most racially and ethnically diverse in history with one in four votes cast by African-Americans, Asians, Latinos, and other racial/ethnic groups. The increase in voters from these groups reflected the larger numbers of eligible voters and the voter

Mexican Immigrants in a Changing Society   375

turnout rates, particularly among African-Americans voters whose turnout rate of 65 percent was an increase of 5 percent from 2004. The Latino voter turnout rate of 50 percent, up 3 percent from 2004, was attributed to an increase in the total number of eligible Latino voters. Two million more Latinos voted in 2008 than in 2004 bringing the total number of Latinos who voted to 9 million. More importantly, the number of Latino voters increased in three battleground states: Colorado, Nevada, and New Mexico. Obama won all three of these important states. Obama would go on to win the election getting 67 percent of Latino voters in contrast to 31 percent for McCain.37 Studies of voting patterns among Latino voters show that the category “Latino” captures overall trends for this group, but patterns within the Latino category reveal within-group differences based on national origins, generation, immigration flows, and other variables.38 Long-standing historical and contemporary differences between Cubans and Mexicans, for example, have produced differences in political ideology, party affiliation, voter turnout, and voting behavior. Cubans have represented a strong base of support for the Republican Party dating back to the first wave of Cuban immigrants as a result of the 1959 Cuban Revolution. In contrast to Mexican immigrants, Cubans experienced significant advantages. The Cold War, the United States’ fear of a communist takeover in Cuba, and the deepening ties between Cuba and the Soviet Union, all combined to create a haven for Cuban immigrants, mostly professionals, who came to the United States and settled in Miami. Cubans remained resentful of the Kennedy administration for its failure to intervene in Cuba and overthrow Castro. These factors set in motion the dynamics that created a solid base of support for the Republican Party and its presidential candidates. Nevertheless, this support waned significantly in 2008. In contrast to Cubans, Mexican immigrants have experienced virulent antiMexican immigrant sentiments, prejudice, discrimination, and other forms of overt and covert racism. Obama framed a specific political narrative intended for Mexican-American voters by repeatedly expressing his shared historical and contemporary experiences with racism, discrimination, and inequality. This narrative succeeded in gaining their support. Obama won the Latino vote in other states with a large Latino population, primarily Mexican-American: Arizona (56 percent), California (74 percent), Colorado (61 percent), Nevada (74 percent), New Mexico (69 percent), and Texas (63 percent).39 Studies analyzing Latino voter turnout and voting identified several issues that bolstered Latino support for Obama.40 The 2003 Iraq War emerged as one galvanizing issue that Latinos identified as fundamental to their support for Obama. Latinos had served with honor in World War II, the Korean conflict, and Vietnam. Although Mexican-Americans, unlike African-Americans and Asians, were not forced to serve in segregated regiments, they nevertheless confronted prejudice and discrimination within the armed forces reflective of

376  North from Mexico

their treatment in their own communities.41 The anti–Iraq War sentiments among Latinos have been compared with the anti-Vietnam protests during the Chicano civil rights movement of the 1960s. In both cases, Latinos protested the disproportionate numbers of Latino casualties in comparison with the overall percentage of eligible Latinos. Critics of the war pointed out that although Latinos made up only 10.5 percent of the total military forces, they composed 17 percent of the combat forces and approximately 11 percent, or even higher according to some estimates, of those killed in action. Ironically, three of the first U.S. soldiers to be killed were noncitizen Latinos who were eventually given posthumous citizenship. Although laws existed to award such citizenship, the Armed Forces Naturalization Act in 2003 ensured that posthumous citizenship would be given to noncitizen U.S. soldiers who died in Iraq. It also provided for expedited naturalization procedures for service personnel who were noncitizens. Latinos’ criticism of the act came from various sectors of their communities even though citizenship could be obtained in this way: “An irony and deep contradiction [is that] if they were to die in combat, these noncitizen soldiers would be awarded posthumous citizenship immediately.”42 During the early phase of the war, Latinos expressed more antiwar sentiments than other groups in the United States. As it became known that the public had been largely deceived into believing Iraq had weapons of Mass destruction, the outcry against the Iraq War increased within Latino communities. Electoral data shows that Latinos with family members in the armed services, whether already deployed to Iraq or not, were more likely to support John Kerry in 2004. By 2007, a large segment of Latinos were positioned to support a candidate in 2008 who would be opposed to the war and pledged to end it if elected president. Barack Obama became this person. The Latino Policy Coalition Survey reported that 46 percent of those Latinos who had registered to vote had a family member or friend who was serving in Iraq or Afghanistan.43 A survey on Latino attitudes on the war in Iraq found that Latinos had greater negative attitudes against the war than the rest of the population; approximately 50 percent of Mexican-Americans favored withdrawing troops from Iraq and 38 percent favored keeping the forces. Since 2004, the negative ratings for George W. Bush had increased among Latinos, both foreign-born and U.S.born.44 Barack Obama made his opposition to the Iraq war a central issue to his 2008 campaign. During the debates and in interviews and political ads, Obama reminded the electorate that in 2003 Hillary Clinton had voted in favor of sending U.S. troops to Iraq. Obama’s pledge to end the war and bring the troops home resonated deeply with Latinos.45 Latinos also identified problems in the economy as another key issue that led to their support for Obama over McCain. Those voters who were worried about economic conditions and their own financial security voted for Obama

Mexican Immigrants in a Changing Society   377

(59 percent). One in three voters expressed their fear that they would not be able to afford health care services and these backed Obama (65 percent).46 Within a day that Lehman Brothers filed for Chapter 11 bankruptcy in 2008 and the Dow Jones Industrial Average lost close to 500 points, McCain issued a statement that the economy was sound. Adding to the economic downward spiral, homeowners across the country were unable to pay their mortgages. All these factors contributed to McCain’s inability to counter Obama, particularly among Latinos.47 Latinos shared this concern with the general electorate, but Mexican-Americans’ persistently greater rates of poverty and unemployment as well as their low educational levels and occupational background compared with the total population exacerbated their concerns about the state of the U.S. economy.48

5. Arizona’s Senate Bill 1070: Criminalization of Immigrants On April 23, 2010, Arizona governor Jan Brewer signed into law Senate Bill 1070, the Support Our Law Enforcement and Safe Neighborhood Act, which declared: The legislature finds that there is a compelling interest in the cooperative enforcement of federal immigration laws throughout all of Arizona. The legislature declares that the intent of this act is to make attrition through enforcement the public policy of all state and local government agencies in Arizona. The provisions of this act are intended to work together to discourage and deter the unlawful entry and presence of aliens and economic activity by persons unlawfully present in the United States.49

The law outlined crimes and penalties related to the enforcement of immigration laws and would be effective as of July 29, 2010. This legislation appeared at a time when Arizona found itself experiencing major demographic changes that ultimately fueled severe anti-immigrant sentiments in general, and more specifically with greater vitriol. Several factors came into play in Arizona leading up to 2010 and the passage of S.B. 1070. In the 1990s, the Immigration and Naturalization Services increased the number of checkpoints in California and Texas and, as a result, border crossings along the border area between Mexico and Arizona became a more frequently used new point of entry. In addition, the passage of North American Free Trade Agreement (NAFTA) in 1994 increased the economic vulnerability of Mexico’s rural farmers who found themselves unable to compete with U.S. agricultural products entering Mexico in the post-NAFTA period and ultimately made the journey north from Mexico, specifically Arizona. Arizona’s Latino population increased from 18 percent in 1980 to 33 percent in 2002. By 2010, Latinos

378  North from Mexico

made up 30 percent of the state’s population; 91 percent of these were Mexican, and U.S.-born Mexican-Americans represented the majority of these. In 2008, the number of Arizona’s unauthorized immigrants numbered between 475,000 and 550,000, or about 6 percent of the state’s population, but those who supported S.B. 1070 believed the percentage was much higher.50 As a result, waves of anti-undocumented fervor, even hysteria, created conditions favorable to such fervor being turned into creating legislation such as S.B. 1070,51 which became a lightning rod for the immigrant rights movement who viewed its provision as manifestations of deeply embedded and, indeed, institutionalized, anti-immigrant political ideology. Anti–S.B. 1070 protest demonstrations and marches were held throughout Arizona and across the nation. Within a month of its signing by the governor, twenty cities or counties issued boycotts of Arizona.52 Criticism focused on a key provisions of S.B. 1070, which stipulated that when a law enforcement official or agency suspected that a person was “an alien who is unlawfully present in the United States, a reasonable attempt shall be made, when practicable, to determine the immigration status of the person.”53 In sum, S.B. 1070 represented a direct threat to civil rights: it allowed for law enforcement officers to determine the legal status of a person they encountered under any circumstances if they had reason to believe the person was undocumented. Fierce opposition centered on a major question: How will a police officer determine who is undocumented? In an effort to highlight the problematic nature of determining legal status from physical appearances, protestors at demonstrations often wore T-shirts with the question “Can you tell if I am legal or not?” Later, S.B. 2162 amended S.B. 1070 by changing the language regarding the power of law enforcement officers to determine a person’s immigration status. H.B. 2162 stipulated that a law enforcement officer is required to ascertain the immigration status of a person “where reasonable suspicion exists that the person is an alien who is unlawfully present in the United States.”54 Such a revision did little to appease opponents of H.B. 1070. Again, the question remained: How do law enforcement officers determine “reasonable suspicion”? Since U.S. citizens are not required to carry proof of their citizenship, determining what is “reasonable suspicion” becomes quite problematic. As a result, the dangerous conflation of physical appearance, particularly skin color, with legal status becomes yet another undermining of a person’s civil rights. The American Civil Liberties Union (ACLU) concluded that “There is simply no unbiased means of determining ‘unlawful presence,’ because . . . there are no observable characteristics of ‘unlawful presence,’ or readily available means by which a police officer could discern ‘unlawful presence’ in any stop, detention or investigative encounter.” S.B. 2162 also provided for Arizona law enforcement officers to issue warrantless arrests if they believed a person had committed an offense that could remove the person from the United States.

Mexican Immigrants in a Changing Society   379

Without the existence of a clear list of offenses that could lead to deportation, this section of S.B. 2162 remained problematic. Many factors, not just the identification of the crime, combine to reach a deportation decision.55 The state law outlined crimes and penalties related to the enforcement of immigration laws and would be effective as of July 29, 2010. Soon after S.B. 1070 passed, other state legislatures introduced a total of two dozen copycat bills, but only five passed; these were in the states of Alabama, Georgia, Indiana, South Carolina, and Utah. The ACLU and several civil rights organizations filed lawsuits in all of these states. The National Coalition for Latino Clergy and Christian Leaders filed the first challenge to these Arizona laws. The coalition based its challenge on equal protection, due process, and preemption under the Supremacy Clause under which federal laws take precedence over state laws. The group believed S.B. 1070 would lead to an increase in racial profiling of Latinos in Arizona.56 The ACLU, MALDEF, and the National Immigration Law Center filed a class action law suit against Arizona seeking a permanent injunction. In addition to its violation of the Supremacy Clause, their lawsuit challenged S.B. 1070 on the grounds that it violated the First Amendment right to freedom of speech, freedom from unreasonable searches and seizure and the Equal Protection Clause and Article II, Section 2 of the Arizona Constitution.57 Before the law could go into effect on July 29, 2010, the U.S. Department of Justice filed a lawsuit on July 28, 2010 in U.S. District Court seeking a permanent injunction. The civil action stated that S.B. 1070 was preempted by federal law (8 U.S.C. 1101, and following sections) and by U.S. foreign policy and that it violated the Supremacy Clause and the Commerce Clause of the U.S. Constitution. Once the judge ordered the injunction, Arizona governor Brewer appealed the injunction. The Ninth U.S. Circuit Court of Appeals heard the appeal but upheld the injunction on April 11, 2011. The Supreme Court’s decision on June 25, 2012, in Arizona v. Supreme Court found that S.B. 1070 violated the Supremacy Clause and therefore was unconstitutional; in its decision the court stressed that “the federal power to determine immigration policy is well settled.”58

6. Latinos, the 2012 Presidential Race, and Beyond According to a 2010 survey of Latinos by the Pew Hispanic Center, Latinos remain divided over many issues regarding immigration, the undocumented, and other key political issues, all of which help set the stage for the 2012 presidential election. Latinos expressed overwhelming support for a policy that allows the undocumented to stay in the United States but pay a fine. Latinos also expressed overwhelming support (76 percent) for the survey’s statement that immigrants make and will make major contributions to the United States.

380  North from Mexico

Throughout this period there was an increased number of anti-immigrant challenges to the Citizenship Clause of the Fourteenth Amendment of the Constitution, which states that “All persons born or naturalized in the United States, and subject to the jurisdiction thereof, are citizens of the United States and of the State wherein they reside.” While a small part of the anti-immigrant population actually agreed that Mexican immigrants come to the United States in order to have U.S.-born children, an overwhelming majority of Latinos did not agree. In fact, 78 percent of Latinos agreed that the U.S. Constitution should not be changed, whereas 56 percent of the general public favored amending the Fourteenth Amendment. More than the majority of the respondents feared the consequences of deportation. Latinos expressed strong concern that the undocumented face the tragic experience of deportation for themselves and their families. The mixed legal status of undocumented families takes this fear of deportation to an even higher level: they fear that not only that will they be deported but that their U.S.-born children will not and, as a result, their families will be separated. Twice as many foreign-born Latinos feared that someone they know will be deported than native-born Latinos. Six in ten Latinos agreed that discrimination is a major problem in the United States, and this number has increased since 2007.59 Latinos in the survey identified several other national issues that became key during the 2012 presidential election between Barack Obama and Mitt Romney. Latinos listed education, jobs, and health care as the three top issues facing Latino communities. Immigration ranked fourth at the time the survey was conducted, but immigration proved to be a more important part of the election narrative at various times during the presidential race. Latinos disapproved of building more fences along the border as a means to keep out the undocumented. Most importantly for presidential elections such as the 2012 election, only 19 percent of Latinos expressed confidence in the Republican Party’s agenda on immigration in contrast to 51 percent of Latinos who supported the policies of the Democrat Party (16 percent listed “neither” and 12 percent listed “don’t know”).60

7. Immigration and the Deferred Action for Childhood Arrivals (DACA) and Its Impact on the 2012 Presidential Election On August 15, 2012, long lines formed at the offices of the U.S. Citizenship and Immigration Services (USCIS) in Chicago, Los Angeles, New York, and many other cities across the United States. Undocumented immigrants waited hours in line to complete their applications for the Deferred Action for Childhood Arrivals (DACA), which would become available when the offices opened. “Deferred action” referred to a limited immigration benefit provided by the Department of Home Security under which immigrants who have deferred

Mexican Immigrants in a Changing Society   381

action are able to apply for work authorizations. A direct path to legal permanent residence or citizenship does not come under the definition of “deferred action.” For many years, minor children who had arrived in the United States with their parents, who were also undocumented, lived at the margins of society, trying to remain in the shadows to avoid deportation. In many cases, these young children were oblivious to their status as undocumented. One young woman recalled the first time she learned that she was undocumented: I thought I was “American.” I never even visited Mexico, but I did only speak Spanish at home with my parents and grandparents. My home has always been here [United States] and not there [Mexico]. I wanted to apply for financial aid for junior college, but a friend told me I couldn’t because I was “illegal” and I told him “Hey what are you talking about. What did I do that was illegal?61

Despite setbacks and the ultimate defeat of the DREAM Act in Congress, DREAMers did not fold in defeat but emerged reenergized and ready to continue their political activism in reaching two major objectives: freedom from fear of deportation and redress from their invisibility through the introduction of opportunities allowing them to emerge from the shadows of marginalization due to their status as unauthorized individuals. DREAMer activists in California had been pressuring California legislatures to pass the state’s own DREAM Act. Different versions of DREAM Acts had reached the floor of the legislature, but despite its passage in 2006, 2007, and 2008, Governor Arnold Schwarzenegger vetoed all three. DREAMers and their supporters continued their struggle for the introduction and ultimate passage of such legislation. On January 11, 2011, Assemblyman Gil Cedillo introduced a new version of the DREAM Act that consisted of A.B. 130 and A.B. 131. A.B. 130 granted access to an estimated $88 million dollars in private financial aid to undocumented students and A.B. 131 allowed for undocumented students who are eligible for in-state tuition to apply for financial aid. Governor Jerry Brown signed this legislation on October 8, 2011. DREAMers had won with their successful strategy of forging a variety of alliances with such groups as the California Dream Network, the Center for Humane Immigrant Rights of Los Angeles, and Dream Team Los Angeles.62 Turning away from their efforts in California, DREAMers began to focus on the federal level, as President Obama had failed to develop a concerted plan to address DREAMer goals. This, in turn, led DREAMers to adopt new mobilization and protest activities. As in their mobilization efforts in California, the group United We Dream succeeded in forming alliances with their supporters. They aimed their protests at a higher level than the state legislature—the presidency—due to its stalling on immigration reform. DREAMers organized protests against the Department of Homeland Security and the Immigration and Customs Enforcement.

382  North from Mexico

With Obama’s 2012 reelection campaign well underway, the DREAMers used traditional and social media to call attention to Obama’s record of rising numbers of deportations. On October 12, after extensive mobilization, DREAMers put into action a time-tested protest strategy: occupation of government offices, in this case, the Los Angeles Immigration and Customs Enforcement office. Some of the demonstrators were arrested, further energizing the protesters. One protester described the scene with this account: We are told that they would be taken away any moment now. . . . We can see the police officers now inside the building and we see glimpses of Alex and Nellie. The whole crowd starts to roar, cheering, whistling, chanting. . . . Then we see Alex, cuffed and escorted by the police, smiling. We all cheer as loud as we can.63

Obama did not respond immediately, and as result, protest activities developed throughout the country. The message to Obama was a succinct one: “Latinos helped you to get elected in 2008, but now it’s 2012.” This rallying cry became the source of political leverage that the DREAMers and their supporters used in their sustained efforts to elicit a response by the Obama administration that would meet their demands. Through their efforts and those of other Latino and immigrant rights’ groups, Obama signed the Deferred Action for Childhood Arrivals (DACA).64 Obama’s decision on June 12, 2012, to formulate a policy under which these children could leave behind their marginalized status as undocumented came at a time in his political reelection campaign when he was being criticized by immigrant rights activists from all national backgrounds, but particularly Latinos. For many, Obama had stalled in his push for immigration reform legislation. Although support for Obama among Latinos remained high, it had started to slip in the months leading up to the November elections, news that alarmed Obama’s campaign team. Obama’s timing represented a crucial juncture for immigrant communities whose patience with Obama was waning. For Republicans, DACA proved to be a political jolt that led to Romney’s increasingly downward slope regarding Latino supporters. In the end, Romney won only 27 percent of the Latino vote. During DACA’s first three months, an estimated 349,000 Mexico-born young adults who had come to the United States as children were provided with a temporary reprieve from deportation under DACA’s provisions. The majority (75 percent) of these applicants were from Mexico.65 DACA contained several provisions that applicants had to meet in order to qualify for the program. Eligibility for DACA would be decided on a case-by-case basis, and qualified persons would not be deported for two years. Those under the age of thirty-one as of June 15, 2012, would receive work authorization if they met

Mexican Immigrants in a Changing Society   383

the requirements. The Department of Homeland Security listed the following requirements that must be met for a person to be eligible to apply for “deferred action for childhood arrivals”66: 1. Were under the age of 31 as of June 15, 2012; 2. Came to the United States before reaching your 16th birthday; 3. Had continuously resided in the United States since June 15, 2007, up to the present time; 4. Were physically present in the United States on June 15, 2012, and at the time of making your request for consideration of deferred action with USCIS; 5. Entered without inspection before June 15, 2012, or had their lawful immigration status expired as of June 15, 2012; 6. Were currently in school, have graduated or obtained a certificate of completion from high school, have obtained a general education development (GED) certificate, or are an honorably discharged veteran of the Coast Guard or Armed Forces of the United States; and 7. Have not been convicted of a felony, significant misdemeanor, or three or more other misdemeanors, and do not otherwise pose a threat to national security or public safety.

A study by the Migration Policy Institute estimated that 1.8 million unauthorized children whose unauthorized parents brought them to the United States might be eligible for a two-year delay in their deportation process if they met the DACA requirements. Those who received deferred action could have both their status and work authorization renewed. Together, California, Florida, Texas, and New York had the largest numbers of undocumented immigrants who might qualify for DACA. Almost three in four of all potential beneficiaries of DACA came from Mexico and Central America. For all areas of the world, approximately half were men and half women. Men, however, outnumber women unauthorized immigrants in the United States.67

8. Latinos, the Issues, and the 2012 Presidential Race The campaigns of Barack Obama and Mitt Romney addressed the issue of immigration both directly and in more nuanced ways. Each candidate developed a political strategy designed to capture support for his party. Romney went after the traditional Republican conservative voters with an extra push toward winning the evangelical vote. Romney’s campaign managers understood, however, that he needed to win nontraditional Republican voters. When the votes came out, despite their efforts to win over Latinos and African-Americans, the Republicans soon discovered the political damage of not successfully reaching

384  North from Mexico

these critical voting blocs. In the end, Republican candidate Romney, at 27 percent, had the second lowest percentage of Latino voters, behind Bob Dole’s 21 percent in 1996 when he challenged the incumbent Bill Clinton. Incumbent president Obama won 71 percent of the Latino vote, second highest to Clinton’s in 1996.68 Speaking on election day, Republican senator Marco Rubio explained why the Republican party did not make political inroads among Latino voters in Florida: “The conservative movement should have particular appeal to people in minority and immigrant communities who are trying to make it, and Republicans need to work harder than ever to communicate our beliefs to them.”69 Ana Navarro, the national Hispanic cochair of McCain’s presidential campaign echoed Rubio’s forecast by commenting that “If we [Republicans] don’t do better with Hispanics, we’ll be out of the White House forever.”70 Latinos composed 10 percent of the electorate—an increase from 9 percent in 2008 and 8 percent in 2002. Those referred to by surveys and polls as “nonwhite” made up 28 percent of the voters, an increase of 2 percent from 2008. Obama carried the Latino vote in battleground states of Colorado, Florida, and Nevada, representing a major political victory. Romney and his campaign mistakenly thought that the traditionally Republican state of Florida would vote “red.” Cubans’ ties to the United States began when the first wave of Cubans left Castro’s Cuba and settled in ‘Little Havana”; for decades they would support candidates at any political level who showed sympathy for their anticommunist movement to regain Cuba. By 2012, the number of Puerto Ricans in Florida increased and with this demographic trend, the state began to lean to the Democratic Party, culminating in a historically significant political victory by Obama: winning 49 percent of the Latino vote to Romney’s 47 percent. Although Romney lost this vote by only 2 percent, the future of the Republican stronghold in Florida would be problematic. In contrast, Obama won the Latino vote by 52 percent, an increase from 2008, when he won the Latino vote by 23 percent. In Nevada, Obama captured 70 percent of the Latino vote, but this was down 6 percent from 2008. In the months leading up to election, Obama and his campaign staff made a tremendous effort to reverse this trend through voter registration drives, voter informational material published in Spanish, and appearances by Obama throughout the state, particularly to Mexican-American community organizations. Mexican-Americans represented the largest Latino group in all three states. Obama also won the Latino vote in North Carolina (68 percent), Wisconsin (65 percent), Virginia (64 percent), and Ohio (53 percent). In addition, Obama won the Latino vote of youth (74 percent), those with a total family income less than $50,000 (82 percent), and those without a college degree (75 percent). Clearly, such voter wins underscored the saliency of issues facing each of these groups. Many young Latino voters were voting for the first time and were the first in their

Mexican Immigrants in a Changing Society   385

families to vote. Polls and surveys indicate that in general Latinos were disillusioned with their family’s economic situation and their own educational experience, but they continued to place their hopes in Obama, far less for Romney. Low-income Latino voters continued to experience stagnant wages and high unemployment but, as with young Latino voters, expressed confidence as Obama’s economic policies proved to be working, albeit slowly. In general, Latino voters continued to view immigration as a key political issue, particularly the plight of undocumented immigrants, with an overwhelming 77 percent agreeing that the undocumented should receive the opportunity to apply for legal status. The percentage increased among Latino voters who knew someone who was undocumented or had been deported.71 The narrative of the Latino vote continues with an overview of Latino voter turnout: who voted and who didn’t. While a record 11.2 million Latinos voted in 2012, 12.1 million did not vote even though they were eligible to do so. Overall, among the nation’s 82.1 million nonvoters in 2012, 15 percent were Latinos. The Pew Research Center’s study of the 2012 Latino electorate found that of the total number of eligible Latino voters, 48 percent cast a ballot, down 2 percent from 2008. Black and white voter turnout surpassed that of Latinos by 66 percent and 64 percent, respectively.72 The Latino vote is one of diversity and differences that will provide insights into the various historical and contemporary experiences of specific groups of Latinos, such as Mexican-Americans. Even though the overall participation decreased between 2008 and 2012, differences among Latino groups reveal the importance of comparing and contrasting political behavior, such as voting, among all Latinos eligible to vote, a category that includes U.S.-born Latinos and Latinos who are naturalized U.S. citizens. Latino college graduates and those of Cuban origin were the Latino groups with the highest rates of voting turnout rates, surpassing the rate for the total U.S. population. For women and men, the turnout rate equaled just about 50 percent. The four Latino groups with lowest numbers included in descending order: those of Mexican origin, high school graduates, 18- to 29-year-olds, and those with less than a high school diploma. As a result of specific social and economic factors, Mexicans are overrepresented in the last three groups, specifically in the lowest levels of educational attainment. One of the most significant political trends for future candidates, particularly presidential candidates in the 2016 election, is that two-thirds of all the voters of Mexican origin in 2012 did not vote. The Pew Research Center, however, estimates that the number of eligible Latino voters will account for 40 percent of the growth in the electorate between 2012 and 2030. This estimate will increase significantly if naturalization rates increase among those Latinos who are in the United States as permanent legal residents. The passage of legislation that could establish a pathway to citizenship for the unauthorized population of Latinos represents yet another factor

386  North from Mexico

with the possible impact of increasing the Latino electorate. The Pew Research Center’s study also revealed the major reasons why Latinos who were eligible for naturalization did not apply. These include a lack of knowledge about the process and low levels of English proficiency. In contrast, those Latinos who wanted to become naturalized citizens identified several factors for this decision. Future naturalized Latinos want to have more legal and civil rights, specifically the right to vote. This political reality will prove essential to future political candidates and their campaign organizations.73

9. Assessing DACA In 2014, two years after DACA began, various nonpartisan research institutes, such as the Migration Policy Institute, undertook an assessment of DACA’s success as well as providing policy recommendations. By July 20, 2014, close to 681,000 had turned in their applications for protection under DACA’s provision, representing, however, only 37 percent of the 1.8 million of those eligible for DACA. Several factors have produced this low number. The $465 application fee is a major roadblock because those eligible are also most likely to find themselves in a precarious economic situation. The ability to have verifiable documentation that an applicant had resided continuously for five years in the United States also accounts for the low application rates. Although the unauthorized live in constant fear of deportation for themselves and their families, many of those eligible feared that applying could lead to the detection of those undocumented family members who are not eligible for DACA. Still others did not have all the necessary information regarding DACA in order to determine whether or not they are eligible. Lastly, assessment reports stressed the need to provide legal advice and information about the basic application procedures that potential applicants could use to apply for DACA.74 Meeting the requirement that an applicant must currently be in school, have graduated from high school, or earned a GED, presents a more complicated roadblock. States’ adult education programs are related to the expansion of the applicant pool for DACA. Those adults who do not have a high school diploma or GED traditionally turn to adult education programs. Unfortunately, many states, such as California and Texas, have slashed these programs or have allowed school districts to reallocate funds set aside for adult education services. As the Migration Policy Institute’s report concludes, “in the short term the decimation of adult education programming in many parts of [California] has made it more difficult for potentially eligible youth to enroll in education programs that would allow them to qualify for DACA.”75 Despite implementation problems, some states initiated specific recommendations in their efforts to overcome factors that have hindered potential immigrants from applying for DACA. The development of better outreach and

Mexican Immigrants in a Changing Society   387

marketing programs intended to keep immigrant youth in school remains one of the best recommendations that can be adopted. In many cases, the private sector has contributed to such efforts by sponsoring student internships and increasing funding for such needed educational tools as computers and highspeed Internet servers for schools and students in at-risk neighborhoods. Adult education programs are now designing more innovative programs to address the needs of “nontraditional adult students” to facilitate a large pool of adults who earn their GED certificates and as a result, meet one of the DACA requirements. Assessment reports also recommend that cities increase their funding for public libraries, specifically their computer and public information resources as well as English proficiency classes and tutoring. Still, DACA provided needed relief. The story of one Mexican immigrant who came to the United States with his parents when he was five years old represents the story of countless Mexican-born children of immigrants. He considers himself a “New Yorker, Brooklyn-raised.” This young man’s life has been one lived with fear of his deportation and that of his parents. He progressed through the educational system, graduated from law school at the City University of New York, started his own lobbying firm, and became an advocate for undocumented youth. After qualifying for DACA he expressed the impact it had on his daily life by saying “that [DACA] really provided that strength to know that I no longer could be deported any day.” His story is one of thousands upon thousands of children of immigrants who were born in Mexico, brought to the United States as young children, and, for all purposes, can be considered a special type of second-generation individuals.76

10. Obama’s Executive Order Facing sustained criticism from immigration reform groups, leaders in DREAMers organizations, and the Mexican and Mexican-American community as well as a Congress unwilling to pass immigration reform, Obama issued an Executive Order on November 20, 2014. Not lost among Mexican immigrants and Mexican-Americans was the fact that November 20 is the date of the official Mexican national holiday commemorating the Mexican Revolution of 1910. In his televised announcement on the evening of November 20, President Obama expressed his frustration that House Republicans did not bring a comprehensive immigrant reform bill to a vote. Therefore, Obama concluded that it was incumbent on him to issue an executive order on immigration and, after outlining its components, concluded with a passionate statement regarding the immigrant and the American Dream: My fellow Americans, we are and always will be a nation of immigrants. We were strangers once, too. And whether our forebears were strangers who

388  North from Mexico

crossed the Atlantic, or the Pacific, or the Rio Grande, we are here only because this country welcomed them in, and taught them that to be an American is about something more than what we look like, or what our last names are, or how we worship. What makes us Americans is our shared commitment to an ideal—that all of us are created equal, and all of us have the chance to make of our lives what we will.77

Obama outlined the major components of the executive order, all of which would ignite a scathing attack by the anti-immigration Congress and the public in general. The Department of Homeland Security lists the following eligibility for requirements78: 1. Expanding the population eligible for the Deferred Action for Childhood Arrivals (DACA) program to people of any current age who entered the United States before the age of 16 and lived in the United States continuously since January 1, 2010, and extending the period of DACA and work authorization from two years to three years. 2. Allowing parents of U.S. citizens and lawful permanent residents to request deferred action and employment authorization for three years, in a new Deferred Action for Parents of Americans and Lawful Permanent Residents program, provided they have lived in the United States continuously since January 1, 2010, and pass required background checks. 3. Expanding the use of provisional waivers of unlawful presence to include the spouses and sons and daughters of lawful permanent residents and the sons and daughters of U.S. citizens. 4. Modernizing, improving and clarifying immigrant and nonimmigrant visa programs to grow our economy and create jobs. 5. Promoting citizenship education and public awareness for lawful permanent residents and providing an option for naturalization applicants to use credit cards to pay the application fee.

Although opponents of President Obama’s use of an executive order on immigration argued that he had overstepped his authority, former presidents have in fact issued similar types of executive orders dealing with immigrants. President John F. Kennedy’s executive order, issued in 1961, is a classic example. At the time, thousands of Cubans were fleeing the Cuban revolution. Kennedy’s order created the Cuban Refugee Program, which provided Cuban refugees with federal assistance. Obama’s executive order on immigration followed other executive orders targeted to specific groups of immigrants, such as Ethiopians who, in the 1970s, were fleeing from the Marxist military dictatorship; Liberians who were escaping their country’s civil wars in the 1980s; and students who lost their academic eligibility as a result of Hurricane Katrina

Mexican Immigrants in a Changing Society   389

in 2005.79 President Obama received even more opposition from those who argued that in contrast to executive orders issued by other presidents, Obama’s executive order would protect unprecedented numbers of unauthorized immigrants in the United States. Approximately half of the total unauthorized immigrant population would qualify for relief from deportation. Mexican immigrants would constitute the largest group that would benefit (44 percent). Several reasons accounted for this. Obama’s executive order focused on length of residence and family status, and Mexicans are more likely to have been living in the United States longer than other immigrants and are more likely to have a family.80 On December 3, 2014, Texas attorney general Gregg Abbott (now governor) filed a lawsuit on behalf of the state of Texas. Soon after, twenty-five states joined Texas: Alabama, Arizona, Arkansas, Florida, Georgia, Idaho, Indiana, Kansas, Louisiana, Maine, Michigan, Mississippi, Montana, Nebraska, Nevada, North Carolina, North Dakota, Ohio, Oklahoma, South Carolina, South Dakota, Tennessee, Utah, West Virginia, and Wisconsin. These states considered Obama’s order to be unconstitutional, but the White House remained steadfast in maintaining its support for the executive order and its constitutionality. On February 15, 2015, a Federal District Court judge in Brownsville, Texas, ruled in the states’ favor concluding that the state of Texas had a legal right to bring a lawsuit against Obama’s executive order. The judge issued the injunction after determining that Texas’ lawsuit had shown that the state would incur direct and serious damages if the executive order were implemented. On May 26, 2015, the U.S. Court of Appeals for the Fifth Circuit in New Orleans let stand the injunction granted by a Federal District Court judge in Brownsville, Texas. The court ruled in the states’ favor citing their legal right to bring a lawsuit against Obama, adding that a delay in its implementation would not pose a problem. The court also denied Obama’s request that the injunction would only apply to those states that brought the lawsuit.81 Ultimately, Obama’s executive order led to state lawsuits that fueled opposition to his overall immigration policies. Nevertheless, fourteen states and the District of Columbia filed papers in the appeals court arguing that Texas and its coalition of 24 states did not understand that Obama’s plan had the potential for economic growth by the increase in tax revenues. According to the Department of Homeland Security, as a result of the federal court order, USCIS [U.S. Citizenship and Immigration Services, a component of the U.S. Department of Homeland Security] did not begin accepting requests for the expansion of DACA as originally planned and suspended implementation of the Deferred Action for Parents of Americans and Lawful Permanent Residents. However, the court’s temporary injunction, did not affect the existing DACA.82 On November 9, 2015, the Obama administration experienced another setback. The Fifth Circuit Court of Appeals blocked Obama’s

390  North from Mexico

executive order by upholding the Texas federal judge’s decision to uphold the injunction aimed at preventing the implementation of the executive order. The Department of Justice immediately announced that it would appeal to the U.S. Supreme Court.83 As President Obama nears the end of his second term, the immigration debate over his executive order will not be resolved but will be left to the new president to be elected in 2016.

11. Social Changes and the Future of Mexican Immigrants and Their U.S.-born Children Mexican immigrants, their U.S.-born children, and later generations will continue to experience a variety of social changes. Many issues will continue to evolve, including transformations in the lives of children of Mexican immigrants; public sentiment and public policies that address the future of undocumented children, many of whom came to the United States at a very young age with their undocumented parents; struggles by these children—the DREAMers—to provide sustained pressure on future presidents and the Congress to redefine citizenship and, in so doing, open doors to higher education for the DREAMers; the ongoing introduction of state laws aimed at curbing unauthorized immigration that too often affect U.S.-born Mexicans; the impact of the Latino vote, specifically the Mexican-American vote, on local, state, and presidential elections; and legal battles over presidential executive orders dealing with unauthorized immigration. All of these issues will remain fluid, contentious, contested, and capable of producing mass mobilization and protests. The historical record shows the inevitability of such anti-immigrant backlashes over attempts to produce social changes. New issues are looming on the horizon that will affect Mexicans, MexicanAmericans, and, indeed, all Latinos. Children and later generations of Mexican immigrants will face several new realities. Educational attainment levels will most likely continue to increase and with these will come the occupational and income-level advancement of the second generation and their children. Ironically, levels of overt and covert prejudice and discrimination will remain. For example, the U.S. government is once again dealing with the discriminatory practice of “redlining.” Banks, mortgage companies, and other lending institutions limit, or even refuse, mortgages to specific (“redlined”) geographic areas, particularly inner-city neighborhoods that are overwhelmingly Latino and African-American.84 The influx of new groups of Mexican and Central American immigrants will also present new challenges. While the total number of Mexican immigrants coming to the United States has stabilized, increases in other groups have been producing new dynamics for social change. For example, since 2012, unaccompanied children aged twelve years and

Mexican Immigrants in a Changing Society   391

younger are the fastest-growing group of unauthorized and apprehended persons crossing the border, although teenagers continue to be the largest group to enter the United States. High rates of poverty and widespread violence attributed to gangs and drug trafficking in Central America represent the major reasons for this new influx of immigrants. Differences exist among countries of origin for these unaccompanied children. Honduras has the highest number, followed by El Salvador, Guatemala, and Mexico.85 Complicating this development, the number of young, unaccompanied girls, particularly from Honduras, who are crossing the border is rising. Girls experience greater rates of sexual violence and murder by gangs in their countries of origin leading them to opt for unauthorized immigration as a survival strategy.86 The rise in border crossings of unaccompanied youth and their detention in often less than acceptable facilities will continue to be a humanitarian challenge that will affect the total U.S. population, but specifically Latinos living in the United States. In addition, young U.S.-born children will, like their parents before them, be forced to respond to situations in which they are mistaken for undocumented, a historical trend that is exacerbated with this new development. The surpassing of Asian immigrants over Latino immigrants is another aspect of the type of social change developing in the United States.87 Asian immigrants are now predicted to be the largest immigrant group coming to the United States, and the decline in Mexican immigration is a major factor in this shift. A slight upward swing in the Mexican economy, declining birthrates in Mexico, and increased deportation and border enforcement are the major factors leading to the decline in Mexican immigration. This demographic shift will most certainly lead to a comparison between the two immigrant groups. Although within-group differences exist, Asian immigrants, particularly Asian Indians, have higher incomes, occupations, years of education, levels of English proficiency, and business ownership than Latinos. Nevertheless, Asian and Mexican immigrants will continue to experience racial discrimination, and their U.S.-born children, like U.S.-born Mexicans, will have to deal with challenges to their citizenship as long as they are perceived as “foreigners.”

Epilogue Alma M. García

Mexican immigrants will continue to make their journey north from Mexico to El Norte, a place of anticipated opportunities, but also a site of antiimmigration public opinion and periods of anti-immigrant public policy. Mexican immigrants bring with them cultural traditions while also adopting new ones as they navigate life in the United States and the immigrant communities within which they most frequently have lived for a period of time. Mexican immigrant women and men, along with their children, have confronted and, indeed contested inequality through social justice movements as the Immigrant Spring of 2006 and the DREAMers movement. Nevertheless, lingering social problems and new and emerging issues pose challenges. A record of these serve as the basis for this new edition of McWilliams’s North from Mexico. These include, but are not limited to, the following: demographic changes of immigrants in the United States, rates of immigration from Mexico, policies on border security and deportations, experiences of women crossing the border, and trends and movements for social change. Mexican immigration will always be fueled by social, political, and economic changes in Mexico and the United States. The U.S.-Mexico border will continue to be a fluid one that is crossed by immigrants and their children. Rates of both authorized and unauthorized immigration have remained stable over the past few years due to a slight improvement in the Mexican economy and, perhaps more importantly, the slowing down of the U.S. economy and the greater enforcement of the border leading to increased deportations. Immigrants cross the border and find a combination of the American Dream and the American nightmare. With changes in the globalized economy in the United States, Mexicans have been coming and migrating to new destination states such as Georgia, Tennessee, and Vermont, where they have found work opportunities; however, communities in these states have had limited contact with Mexican immigrants, producing the potential for social conflict,

394   Epilogue

including violence, directed at them. Mexicans, like other immigrants, often find themselves caught in a nexus of exploitation in the paid labor force, antiimmigrant hostilities, and limited access to higher education for their children. The demonization of immigrants has been transformed into anti-immigrant legislation such as the Immigration Act of 1990, state initiatives such as California’s Proposition 187, and challenges to birthright citizenship protection found in the Fourteenth Amendment. In the decades after Carey McWilliams wrote North from Mexico and Matt Meier updated this classic book, Mexican immigrants, U.S.-born Mexicans, and their supporters from all walks of life have joined in collective activism to contest social injustices. Mexican women have been active in these struggles, particularly those that address specific hostilities against women, such as the sexual exploitation, even murders, of immigrant women who come to the border to work in maquiladoras, many of whom later cross the border at isolated and dangerous entry points. Mexican women have contested these conditions by forming grassroots organizations to combat sexual violence, including violence against lesbians. Undocumented students, most of whom have only known life in the United States, having arrived as very young children with their undocumented parents, have transformed themselves from marginalized individuals into organizers. For example, undocumented students have formed such collective organizations as United We Dream, an activist group aimed at providing unauthorized children—the DREAMers—with information on the DREAM Act and DACA as well as functioning as a political interest group that continues to lobby for the passage of the DREAM Act. Unauthorized Mexican immigrants continue to face “demonization” as witnessed in deportation rates and enforcement practices of the Department of Homeland Security. Deportations increased under the Obama administration, and, as a result, his support among Mexican immigrants and MexicanAmericans, usually taken for granted, has become problematic. Immigration reform has continued to be stalled during President Barack Obama’s two presidential terms and a Republican Congress. Though they still lag behind those of other groups, the voter registration and voter turnout rates for naturalized Mexicans and U.S.-born MexicanAmericans have continued to increase, particularly when political candidates seek the support of “the Latino vote.” Obama’s win in 2008 over John McCain and his reelection in 2012 over Mitt Romney revealed the importance of building outreach strategies to groups such as Mexicans-Americans, particularly first-time voters. Future presidential candidates as well as candidates for other offices will need to build innovative strategies to gain their support. Mexican immigrants, their U.S.-born children, and later generations have had opportunities for upward mobility, albeit limited. Mexican-Americans continue to face issues such as obstacles to higher education, limited job

Epilogue   395

opportunities with low wages and limited benefits, and, in general, ongoing prejudice and discrimination. Nevertheless, over the past twenty years, more Mexican-Americans are graduating from high school and are entering and completing college; a limited number are even continuing to postcollege education. In addition, Mexican immigrants have a long history as entrepreneurs who have started small “mom-and-pop stores” and restaurants, particularly in immigrant communities. More recently, the U.S. Census Bureau, particularly American Community Survey, has recorded a small increase in the number of midsize businesses among Mexicans and U.S.-born Mexicans, including hightech industry startups in Silicon Valley. The number of Mexican-Americans in elected office has increased over the past twenty years. Some of these will most certainly become national figures, even presidential candidates. All of these developments point to a gradual upward mobility, one in which some children of Mexican immigrants are becoming better off than their immigrant parents. Much work needs to be done to improve socioeconomic and political opportunities for Mexicans. Anti-immigrant sentiments, however, exist side by side with ones that welcome immigrants, support social reforms, and promote movements for social justice. As the historical and contemporary record shows, Mexicans in the United States have raised their voices in protest through mass mobilization and organization building. Indeed, echoes of César Chávez’s slogan of si se puede will be heard throughout Mexican and MexicanAmerican communities that continue to struggle for social justice.

Acknowledgments

In preparing this volume I have received invaluable assistance from my greatly esteemed friend, Dr. Eshref Shevky, who served as director of the Tewa Basin Study in New Mexico and is now with the John R. Haynes Foundation in Los Angeles. I also wish to acknowledge my indebtedness to the following individuals: the late Arthur Shapiro, who helped me with some of the research; Fred Ross, formerly with the American Council on Race Relations and now directing the important work of the Industrial Areas Foundation among Spanish-speaking people in Los Angeles; Louise Evans and Charles Graham of the Colorado Civic Unity Council; Dr. Norman Humphreys of Wayne University; Dr. Ruth D. Tuck of the University of Redlands; and Alice Greenfield, who served as secretary of the Sleepy Lagoon Defense Committee. For help in preparing the manuscript, I am indebted to Ross B. Wills and Margaret O’Connor.

Acknowledgments to the New Edition Alma M. García

I am grateful to many colleagues and friends who have provided me with constant support, helpful criticism, and insights that together have helped me write this third update to Cary McWilliams’s North from Mexico: The SpanishSpeaking People of the United States. My most sincere and heartfelt thanks to the following: Marian Perales, Manager, Editorial Development, Race and Ethnicity/ American History, ABC-CLIO, for her support and encouragement during this project. Marian, a special thanks for your patience and understanding when my work on this project was temporarily interrupted. Francisco Jiménez, for his lifelong mentorship and friendship. Without his support I could not have completed this project. Francisco, a special thank you for always believing in me; Matt Meier (1917–2003), a pioneer in Chicano/Mexican American Studies, for his mentorship, support, and friendship. I am truly honored to have my name appear with mi amigo Matt in this third update of North from Mexico. Diane Jonte-Pace, Senior Vice Provost for Academic Affairs, Provost’s Office, Santa Clara University, for providing me with a stipend while I completed this book and, more importantly, for always supporting my research projects; Gloria Hofer, Michael Gilkinson, and Marc Ramos, Technical Training Specialists, Santa Clara University Library, for their technical assistance; Anna Sampaio and Mario T. García for providing me with their expertise in Chicano/Mexican-American history; Juan Velasco, Associate Professor, English Department, Santa Clara University, for introducing me to the spectacular artwork by Consuelo Jiménez Underwood, whose work Undocumented Border Flowers graces the cover of this book; Swati and Harish Negi and their wonderful staff at Copy Craft, Santa Clara, California, for their technical assistance; and Hilda Lopez and Karen Vigil, Santa Clara Public Library, Mission Branch, for their help in locating many materials I needed for this project.

400   Acknowledgments to the New Edition

Writing is too often a solitary endeavor. Without the friendship of the following I could not have “survived” the months of writing, often under exceptional personal circumstances. Special thanks to: Ed Garcia, my brother and dear friend, for his daily phone conversations with me that helped make the difficult times that I was experiencing more bearable. Ed, you are my rock of support; the Takamatsu family—Will, Tiff, Megan, and Kenji—and their extended family who have taught me what neighborly love is. To my former professor Dr. John Haddox, Philosophy Department, University of Texas at El Paso, for introducing me to North from Mexico when I was an undergraduate at the University of Texas at El Paso. His mentorship during my college years provided me with a solid foundation for my life as an academic. He introduced me to Chicano Studies and the philosophers Antonio Caso, Jean Paul Sartre, and Albert Camus. Dr. Haddox, thank you from the bottom of my heart. All students should be as fortunate and blessed to have someone like you as their professor. I would like to express my deepest and most sincere thanks to my lifelong friend Josie Rosas, my treasured amiga para siempre and always “rational” advocate who has taught me the meaning of friendship and spirituality. Josie, thank you for always sustaining me when I was “not in a good place.” A special thanks to my colleague Professor Laura Robinson, Department of Sociology, Santa Clara University, for her invaluable technical assistance and encouragement during the final phase of this book. And lastly, I would also like to express my thanks to the hundreds of thousands of Mexican immigrants who came and are coming to the United States in search of a better life for themselves and their children. Mexican immigrants have endured many injustices but have always struggled for social justice. They have maintained their sense of dignity and pride in their Mexican heritage, one that they have proudly inculcated in their children. Their presence in the United States strengthens our country as Pope Francisco so eloquently stressed in many of his speeches during his visit to the United States in 2015. Referring to his own family roots, he proudly stated in a simple, but forceful way, “También soy inmigrante.” (“I am also an immigrant.”) In memoriam: My father, Amado García Rodarte (1909–1986), who came “North from Mexico” and taught me to enjoy life’s every moment; My mother, Alma Araíza García (1920–2010), a daughter of Mexican immigrant parents who spent her lifetime inculcating in me the value of higher education.

Notes on Sources

For general bibliographic materials about cultural relations in the Southwest and the Spanish-speaking people, the best sources are: Spanish-Speaking Americans in the United States: A Selected Bibliography by Lyle Saunders (1944), and A Guide to Materials Bearing on Cultural Relations in New Mexico also by Lyle Saunders (University of New Mexico Press, 528 pp., 1944). Mr. Saunders has kept the last-mentioned guide current by bibliographic notes which have regularly appeared in the excellent New Mexico Quarterly Review. In view of the existence of these exhaustive, thoroughgoing bibliographies, I have deemed it unnecessary to include a bibliography in this volume. A word or two about the sources, however, may not be amiss. On the general character of the Southwest as a region there are several valuable sources. Under this heading, the works of Dr. Walter Prescott Webb, particularly The Great Plains (1931) and The Texas Rangers (1935), are of major importance. Nor can I commend too highly A Vaquero of the Brush Country by J. Frank Dobie (1929), and, by the same author, Coronado’s Children (1931) and The Longhorns (1941). Special mention should also be made of The Greater Southwest by Rupert Norval Richardson and Carl Coke Rister (1934), a firstrate general source; and, in this same category, Our Southwest (1940), by Erna Fergusson, and the well-known writings of Charles Fletcher Lummis and Mary Austin. The indispensable source, however, is Dr. Webb’s The Great Plains, one of the finest volumes ever written about Western America. There are many excellent books about New Mexico but my preferences, in this field, are the following: Forgotten People: A Study of New Mexicans by George I. Sanchez (1940), a most valuable source; Old Santa Fe by Ralph Emerson Twitchell (1925); Piñon Country by Haniel Long (1941); Sky Determines by Ross Calvin (1934); Rio Grande by Harvey Fergusson (1933); and Caballeros by Ruth Laughlin Barker (1931), the best discussion of the New Mexico arts and crafts. In the chapter on New Mexico, I have also drawn heavily upon two excellent unpublished dissertations: “Relations Between the Spanish-Americans and Anglo-Americans in New Mexico” by Dr. Carolyn Zeleny (Yale University, 1944); and “A Study of Isolation and Social Change in

402  Notes on Sources

Three Spanish-Speaking Villages in New Mexico” by Dr. Paul Walter (Stanford University, 1938). On the Spanish-speaking in Texas, in addition to the sources indicated in the notes, I have drawn upon an unpublished dissertation by Jovita Gonzales entitled “Social Life in Cameron, Starr, and Zapata Counties” (University of Texas, 1930); Cultural Conflict in Texas, 1821–1835 by Dr. Samuel Harman Lowrie (1932); and Texas: A Contest of Civilizations by Dr. George P. Garrison (1903). Mention should also be made, on this score, of Latin Americans in Texas by Pauline R. Kibbe (1946). Perhaps the most interesting material on the Spanish-speaking of Colorado is to be found in yet another unpublished dissertation: “The Spanish Heritage of the San Luis Valley” by Olibama López (University of Denver, 1942). Of the Spanish-speaking in California, the best single source is Not with the Fist: Mexican-Americans in a Southwest City by Ruth D. Tuck (1946), a fine, perceptive study, notable for its sympathetic understanding and interpretation of Mexican-American life and culture. The outstanding authority on Mexicans in the United States, of course, is Dr. Paul S. Taylor of the University of California. Representing years of study, research, and field investigations, the series of monographs making up his study of Mexican labor in the United States are of the utmost importance. Included in the series are the following: Volume I: Imperial Valley (1928); Volume II: Valley of the South Platte (1929); Volume III: Migration Statistics; Volume IV: Racial School Statistics California (1927); Volume V: Dimmit County, Winter Garden District, South Texas (1930); Volume VI: Bethlehem, Pennsylvania (1931); and Volume VII: Chicago and the Calumet Region (1932). Dr. Taylor’s volume, An American-Mexican Frontier (1934), is the finest single volume on Anglo-Hispano relations in print today. In 1939 the Soil Conservation Service in New Mexico released its famous Tewa Basin Study which, with the supplemental studies that accompanied the major two-volume report, constitutes an invaluable piece of research on the Indian pueblos and the Spanish-speaking villages of New Mexico.

Additional Notes on Sources Matt S. Meier

Since Carey McWilliams wrote his “Notes on Sources” in North from Mexico, great strides have been made in the researching, writing, and publishing of historical materials on Mexican-Americans and in the conceptualizing of their history. In the past forty years possibly as many works have been published about the history of this group as in all the years before 1948. In this brief addition to McWilliams’s “Notes” only the most notable authors and works can be listed. Additional important works are cited in the notes to chapters XVII, XVIII, and XIX. Following McWilliams’s trailblazing work, which has become the point of bibliographic departure, nearly two decades passed before the task of exploring, organizing, and committing to print the experience of Mexican-Americans was resumed in earnest. Early in the 1960s the University of California, Los Angeles, undertook a massive Mexican-American Study Project with a Ford Foundation grant. This resulted in eleven published advance reports and, in 1970, a comprehensive sourcebook, The Mexican American People: The Nation’s Largest Minority, edited by Leo Grebler, Joan Moore, and Ralph Guzmán. Concurrent with this project a growing interest in Mexican-Americans, accelerated by the civil rights movement, helped spawn many useful books of selected readings: Julian Samora, La Raza: Forgotten Americans (1966); in 1970 two collections: Manuel Servín, The Mexican American: An Awakening Minority; and John Burma, Mexican Americans in the United States: A Reader; Wayne Moquin, A Documentary History of the Mexican American (1971); and in 1973, three works: Julian Nava, ¡Viva La Raza!: Readings on Mexican Americans; Livie Durán and H. R. Bernard, Introduction to Chicano Studies: A Reader; and Renato Rosaldo, et al., Chicano: The Evolution of a People. In the early 1970s, immediately following Joan Moore’s excellent sociological study, Mexican American (1970), there appeared two popular general

404   Additional Notes on Sources

histories of la raza, the first since McWilliams’s North from Mexico. The Chicanos: A History of Mexican Americans (1972) by Matt S. Meier and Feliciano Rivera, following in the steps of McWilliams, stressed the interrelated themes of immigration, conflict of economic interests, and cultural adjustment. In the same year Rodolfo Acuña published his Occupied America: The Chicano Struggle toward Liberation, which set forth an internal colonialism model. This was followed in the 1980s by a second (1981) and third edition (1988). In his revisions, Acuña abandoned the colonialism theme for a thematic approach, but one which still emphasized victimization. These two survey accounts were soon followed by a number of similar general works which added little to Chicano historiography. However, in 1979 Mario Barrera’s more theoretical and heavily economic Race and Class in the Southwest: A Theory of Racial Inequality, appeared. In 1985 Alfredo Mirandé stressed Chicanos’ indigenous roots in his highly interpretive The Chicano Experience: An Alternative Perspective. Meanwhile there had come into existence a number of excellent regional monographic works. In 1966 Leonard Pitt had published his outstanding Decline of the Californios: A Social History of the Spanish Speaking Californians. At the end of the sixties The Spanish Americans of New Mexico: A Heritage of Pride (1969), by Nancie González, building on the earlier work of George I. Sánchez, depicted the changes taking place within this subgroup. A dozen years later Robert Rosenbaum’s Mexicano Resistance in the Southwest (1981), while broad in its coverage, devoted over half of its pages to Nuevo Mexicanos’ struggle against Anglo domination. In the Texas area Lynn Perrigo led the way with his Texas and Our Spanish Southwest (1968), a broad survey of southwestern history stressing Texas and the Mexican contribution. Three years later Félix D. Almaraz’s Tragic Cavalier: Governor Manuel Salcedo of Texas, 1808–1813, focused on the early period of the Tejano revolt for independence from Spain. A full decade later Arnoldo de León published The Tejano Community, 1836–1900 (1982). This study of the interaction of Mexican and United States cultural influence was followed by his They Called Them Greasers: Anglo Attitudes toward Mexicans in Texas, 1821–1900 (1983). More recently David Montejano published his Anglos and Mexicans in the Making of Texas, 1836–1986 (1987), which has continued the stress on regional rather than general Mexican-American topics. By the mid-1970s there was, in addition, a new generation of young Chicano historians completing doctoral dissertations and beginning to write and publish monographic studies in their various areas of expertise in social history and particularly pioneering in urban history. In 1976, for example, Louise Año Nuevo Kerr completed her University of Chicago dissertation, “The Chicano Experience in Chicago,” one of the first and frequently cited studies of Mexican-Americans in the Midwest. A year earlier Oscar Martínez had

Additional Notes on Sources   405

published Border Boom Town: Ciudad Juarez Since 1848, followed by Richard Griswold del Castillo’s The Los Angeles Barrio, 1850–1890: A Social History, and Albert Camarillo’s Chicanos in a Changing Society: From Mexican Pueblos to American Barrios in Santa Barbara and Southern California, 1848–1930, both in 1979. Two years later Mario T. García published his well-documented Desert Immigrants: The Mexicans of El Paso, 1880–1920, a detailed study of urban life; and in 1982 Francisco Balderama’s In Defense of La Raza: The Los Angeles Mexican Consulate and the Mexican Community, 1929 to 1936 was published followed in the next year by East Los Angeles: History of a Barrio (1983), by Ricardo Romo, who emphasized immigration to Los Angeles up to the Great Depression. In one of the newest and therefore least-developed fields, the history of the Chicana, Marta Cotera published Profile of the Mexican-American Woman (1977); two years later La Chicana: The Mexican-American Woman, by Evangelina Enriquez and Alfredo Mirandé, appeared. In 1984 Vicki Ruiz and Susan Tiano published Woman on the U.S.-Mexico Border: Responses to Change; and Richard Griswold del Castillo came out with La Familia: Chicano Families in the Urban Southwest, 1848 to the Present. Three years later came Patricia Zavella’s Women’s Work and Chicano Families: Cannery Workers of the Santa Clara Valley. In 1986 the National Association for Chicano Studies (NACS) issued Chicana Voices: Intersections of Class, Race, and Gender. This collection of essays, a majority written by Chicanas, was the first NACS publication devoted exclusively to the Chicana experience. Besides these monographs in the newer social historical fields, we have the more traditional labor history works of Ernesto Galarza, Juan Gómez Quiñones, Mark Reisler, Vernon Briggs, David Macías, Luis Arroyo, and others. In political history there are also the valuable works of Ralph Guzmán, F. Chris García, Rodolfo O. de la Garza, Tony Castro, and John Shockley; and in the vast area of immigration and borderland studies are the publications of people like Jorge A. Bustamante, Wayne Cornelius, Julian Samora, María Herrera-Sobek, Gilbert Cárdenas, Alejandro Portes, Arthur Corwin, Ellwyn Stoddard, Abraham Hoffman, Lawrence Cardoso, and Stanley Ross. In the field of demography Los chicanos: geografía histórica regional (1976), by geographer Richard Nostrand, is unique in presenting geographic and demographic details of Mexicanos in the Southwest in historical perspective. Also of continuing great importance are two major reprint series, the twenty-one-volume The Mexican American (1974) and the fifty-five-volume The Chicano Heritage (1976), both edited by Carlos Cortés. Likewise, of exceptional value to understanding the historical background to the MexicanAmerican experience are the works of David J. Weber and Oakah Jones dealing with the Spanish colonial and Mexican national periods on the northern Mexican frontier. Not strictly historical in organization but extremely useful to

406   Additional Notes on Sources

the student reader are the works of Octavio Romano V. in cultural anthropology and Clark Knowlton in sociology. In addition to the important publications already cited, there are a number of bibliographic works which can lead the inquiring student and scholar to many more books describing, analyzing, and interpreting the MexicanAmerican experience. Among the more extensive and useful are, in chronological order, Mexican American Bibliographies (1974), a reprint of five earlier bibliographies, edited by Carlos Cortés; Frank Pino’s two-volume multidisciplinary Mexican Americans: A Research Bibliography (1974); Arnulfo Trejo’s broad work, Bibliografía Chicana: A Guide to Information Sources (1975); The Mexican American: A Critical Guide to Research Aids, by Barbara J. and J. Cordell Robinson (1980), a well-annotated guide to bibliographies; and Bibliography of Mexican American History, with over 4,000 entries, partially annotated, by Matt S. Meier (1984). Besides full-length books, there have been numerous important journal articles published in leading social science periodicals such as Aztlán: International Journal of Chicano Studies Research, International Migration Review, Journal of Borderlands Studies, Journal of Ethnic American History, Journal of the West, New Scholar, Social Science Quarterly, and regional journals like the New Mexico Historical Review. All of this is not to mention the large number of works in sociology, cultural anthropology, literature, theater, and the arts which serve to complement the more strictly historical publications. Obviously in a short overview of such a large field as this some important works may get overlooked.

Chapter Notes

Introduction to the Third Edition 1.  Carey McWilliams, “Introduction,” in North from Mexico: The Spanish-Speaking People of the United States (New York: Praeger, 1948). 2.  Oscar Handlin, The Uprooted: The Epic Story of the Great Migrations that Made American People (New York: Little, Brown and Company, 1951).

Chapter 1 1.  New Mexico Historical Review, Vol. 16, p. 1. 2. Harry Bernstein, “Spanish Influences in the United States: Economic Aspects,” Hispanic American Historical Review 18, no. 1 (February 1938): 43–65. 3. Arthur P. Whitaker, “The Spanish Contribution to American Agriculture,” Agricultural History 3, no. 1 (January 1929): 1–14. 4.  Claude B. Hutchinson, ed. California Agriculture (Berkeley: University of California Press, 1946). 5.  William R. Shepherd, “The Spanish Heritage in America,” Modern Language Journal 10, no. 2 (November 1925): 75–85.

Chapter 2 1.  See picture-spread Los Angeles Daily News, May 7, 1947. 2.  American Planning and Civic Manual (1940): pp. 260–266. 3.  Arthur Campa, Spanish Folk-Poetry in New Mexico (Albuquerque: University of New Mexico Press, 1946): p. 13. See Helen Zunser, “A New Mexican Village,” Journal of American Folklore, 48, no. 188 (April–June 1935): 141. 4.  New York Times, June 1, August 3, 4, 8, and September 4, 1947. 5. Stetson Kennedy, Palmetto Country (Florida Historical Society Press, 1942): 269–296. 6.  The Civilization of the Americas (1938): p. 116. 7.  Arizona Quarterly (1946): Vol. 2, p. 34.

408   Chapter Notes

Chapter 3 1.  The Greater Southwest (1934): p. 26. 2.  J. Fred Rippy, The United States and Mexico (New York: Alfred A. Knopf, 1926): p. 382.

Chapter 4 1.  Ruth Barker Laughlin, Caballeros (New York: Appleton-Century, 1931); see also the article by Frank Applegate, Survey-Graphic, May 1, 1931. 2.  Note the article by Juan B. Rael, California Folklore Quarterly, Vol. I, p. 83. 3.  Allan G. Harper, Andrew R. Cordova [and] Kalvero Oberg, Man and Resources in the Middle Rio Grande Valley (Albuquerque: University of New Mexico Press, 1943), in which the process of deterioration in natural and human resources is discussed in great detail. 4.  Survey-Graphic, May 1, 1931, p. 142.

Chapter 5 1.  Sylvester Mowry, Arizona and Sonora: The Geography, History, and Resources of the Silver Region of North America (New York: Harper & Brother, 1864). 2.  Harper’s, July, 1890. 3.  The Southwest Political and Social Science Quarterly, December 1929, p. 267. 4.  Frank W. Blackmar, Spanish Institutions of the Southwest (Baltimore, Johns Hopkins Press, 1891). 5.  John L. McCarty, Maverick Town: The Story of Old Tascosa (Norman: University of Oklahoma Press, 1946).

Chapter 6 1. Rippy, supra, p. 296. 2. Mody C. Boatright, ed. Mexican Border Ballads (Austin: Texas Folklore Society, 1946). 3. Tracy Hammond Lewis, Along the Rio Grande (New York, Lewis Publishing Company, 1916). 4.  Manuel Ugarte, The Destiny of a Continent (New York: Alfred A. Knopf, 1925). 5.  Frederick Law Olmsted, A Journey Through Texas (New York: Dix, Edwards and Company, 1857): 455.

Chapter 7  1. New Mexico Guide, p. 14.  2. Sixty Years in Southern California, p. 140.  3. See the article by Arthur E. Hyde, Century, March 1902, Vol. 63, p. 690.  4. Recollections of a Western Ranchman (1884).

Chapter Notes   409  5. New Mexico Historical Review, Vol. 20, p. 202.  6. See article by C. P. Loomis, Sociometry, February 1943; and “Race Relations in New Mexico,” pp. 208–216, in Mexican Immigration to the United States by Manuel Gamio (New York: Arno Press, 1930).  7. Paul I. Wellman, The Trampling Herd (New York: Carrick, 1939).  8. For an account of a lynching in Arizona, see New Mexico Historical Review, Vol. 18; and, for Colorado, see Gamio, supra, p. 213.   9.  Owen Cochran Coy, Gold Days (Los Angeles, Powell Pub. Co., 1929), 204. 10.  Walter M. Fisher, The Californians (London: Macmillan, 1876).

Chapter 8  1. California Gold by Rodman W. Paul (Cambridge: Harvard University Press, 1947): 36.   2.  Report on the Mineral Resources of the United States by J. Ross Browne, 1867, p. 21. See also T.A. Rickard, A History of American Mining (New York: McGraw-Hill, 1932).  3. Arizona by James H. McClintock (1916), Vol. I, p. 101.  4. Harper’s, June, 1863; see also “Down in the Cinnabar Mines” by J. Ross Browne, Harper’s, October, 1865.   5.  Owen Cochran Coy, Gold Days (Los Angeles, Powell Pub. Co., 1929): 165.   6.  Will H. Robinson, The Story of Arizona (Phoenix: Berryhill Company, 1919).  7. McClintock, supra, Vol. II, p. 421.  8. Charles Wayland Towne and Edward Norris Wentworth, Shepherd’s Empire (Norman: University of Oklahoma, 1945).   9.  Winifred Kupper, The Golden Hoof (New York: Alfred A. Knopf, 1945). 10. Campa, Spanish Folk-Poetry in New Mexico, 97. 11.  Dane Coolidge, Old California Cowboys (E. P. Dutton, 1939), Chapter X. 12.  Arizona Quarterly, Summer, 1946, p. 24; see also “Mexican Color Terms for Horses” by W. H. Whatley, in Mustangs and Cow Horses, publication Texas Folklore Society (1940): p. 241. 13.  Journal of Political Economy, Vol. 20, p. B07; Journal of Economic History, December 1942, article by Sanford A. Mosk. 14. Edith Nicholl Ellison, The Desert and the Rose (Boston: Cornhill Company, 1921): 37. 15. French, supra, p. 133. 16.  Journal of American Folklore, Vol. IX, p. B1.

Chapter 9  1. The White Scourge by Edward Everett Davis (1940).  2. Survey-Graphic, May 1, 1931.  3. Economic Geography, January 1931, p. 1.  4. The New Republic, April 7, 1947, p. 14.  5. Harry Schwartz, Seasonal Farm Labor in the United States by Harry Schwartz (New York: Columbia University Press,1945), 29.

410   Chapter Notes  6. Commercial Survey of the Pacific Southwest, Dept. of Commerce (1930): 224–254.  7. Lawrence Leslie Waters, “Transient Mexican Agricultural Labor” Southwestern Social Science Quarterly 22, no. 1 (June, 1941): 49–66.  8. See article by Dr. Max Handman, American Journal of Sociology, January 1930.  9. See testimony of Emelio Flores before the Industrial Relations Commission, 1915. 10. Alvin T. Steinel, History of Agriculture in Colorado (Colorado State Board of Agriculture, 1926). 11.  Hearings, Committee on Agriculture and Forestry, U.S. Senate, Both Congress, March 1947, p. 24.

Chapter 10 1. Stuart Jamieson, Labor Unionism in American Agriculture (Washington, D.C.: United States Department of Agriculture, 1945). 2. Selden C. Menefee and Orin C. Cassmore, The Pecan-Shellers of San Antonio (Federal Works Agency, Works Progress Administration, 1940). 3.  The Nation, May 1, 1935. 4. Clarence Raymond Niklason, Edwin Bates, Fowler Wesley Barker, Commercial Survey of the Pacific Southwest (Washington, D.C.: United States GPO, 1930): 322. 5.  The Outlook, February 2, 1916; Survey, October 27, 1917, p. 97. 6.  In the Matter of Miami Copper Company, Non-Ferrous Metals Commission, National War Labor Board, February 5, 1944; also, PM, July 3, 1944, p. 3. 7.  James Monroe Patton, “The History of Clifton,” unpublished thesis, University of Arizona, 1945. 8.  New Mexico Historical Review, Vol. 13, p. 415; also Organized Labor in Mexico by Marjorie Ruth Clark (1934): p. 11. 9. Gamio, supra, p. 44.

Chapter 11 1.  See The Shadows of the Trees, by Jacques Ducharme (New York: Harper and Bros., 1943); many of his observations about French-Canadians in New England could be applied, without modification, to Mexicans in the Southwest. 2. Robert Redfield, “The Folk Society,” American Journal of Sociology 52, no. 4 (January, 1947): 293–308. 3. Fred W. Ross, Community Organization in Mexican-American Communities by Fred W. Ross (Los Angeles, American Council on Race Relations, 1947). 4.  “They Fenced Tolerance In” by Dallas Johnson, Survey, July 1947, pp. 398–400. 5.  American Journal of Sociology, Vol. 22, p. 391.

Chapter 12 1.  See comments by Dr. George Sanchez, Common Ground, Autumn 1943, pp. 13–20. 2.  See comments by Albert Deutsch, PM, June 14, 1943; Racial Digest, July 1943, pp. 3–7; New York Times, June 11, 1943.

Chapter Notes   411

Chapter 13 1.  Los Angeles Herald-Express, June 5, 1943. 2.  “Imported Mexican Workers Save Millions in Citrus Crops,” reads a headline, Los Angeles Times, June 30, 1943. 3.  Los Angeles Times, June 18, 1943.

Chapter 14 1.  Congressional Record, April 24, 1945. 2.  The New Republic, September 30, 1946, p. 412. 3.  See Betty Kirk, “Mexico’s Social Justice Party” The Nation, June 12, 1943; PM, May 21, 1944, p. 3. 4.  Los Angeles Times, October 15, 1942. 5.  Los Angeles Times, December 20, 1942. 6.  See article by Heinz H. F. Eulau, The Inter-American, March 1944, pp. 25–28. Mr. Eulau, during the war, was chief of the Division of Propaganda Analysis, Department of Justice. 7.  Los Angeles Times, September 22, 1947, Part II, p. 1. 8.  Common Ground, Autumn, 1943, pp. 13–20. 9.  New York Herald Tribune, February 16, 1944.

Chapter 15 1.  Interpreter Releases, March 22, 1943. 2.  Lloyd H. Fisher, The Problem of Violence (American Council on Race Relations, 1945): p. 18. 3.  See the series of articles by Agnes E. Meyer in the Washington Post, April 22–29, 1947. 4.  Nutrition and Certain Related Factors of the Spanish-American in Northern Colorado (Denver, 1943); Inter-American Short Papers, no. VII (University of New Mexico, 1943). 5.  Common Ground, Spring 1947, pp. 80–83. 6. J. T. Reid, It Happened in Taos (Albuquerque: University of New Mexico Press, 1946). 7.  Lloyd Spencer Tireman, La Comunidad (Albuquerque: University of New Mexico Press, 1943).

Chapter 16  1. California by Josiah Royce (1897): p. 226.  2. Herbert Eugene Bolton, The Pacific Ocean in History (New York: Macmillan, 1917), the chapter on “Speech Mixture in New Mexico” by Aurelio M. Espiñosa.  3. Hispania, Vol. 28, pp. 505–507.   4.  From a list given me by Ruth D. Tuck; see also “The Pachuco Patois” by Beatrice Griffith, Common Ground, Summer 1947, pp. 77–84.

412   Chapter Notes  5. Arizona Quarterly, Summer 1946, p. 24.   6.  See also, Emerson Hough, The Story of the Cowboy (New York: Appleton, 1908): 26; Ramon F. Adams, Cowboy Lingo (New York: Houghton Mifflin, 1936).  7. Mary Austin, Starry Adventure (Boston: Houghton Mifflin, 1931): 62; by J. Frank Dobie, Coronado’s Children (Dallas: Southwest Press, 1930): 361–367; Laughlin, Caballeros, 403–410.   8.  Papers, Conference on Educational Problems in the Southwest, Santa Fe, August 19–24, 1943.  9. Joaquin Ortega, The Compulsory Teaching of Spanish in the Grade Schools of New Mexico (University of New Mexico Press, 1941): 9. 10.  New York Times, August 10, 1947, article by Gladwin Hill.

Chapter 17   1.  The terms Mexican-American and Chicano are used completely interchangeably to refer to persons of Mexican descent who were born in the United States or have made it their home. The word Chicano avoids the ambiguity of both Mexicano and Latino/Hispanic as well as the hyphenate implication some find in Mexican-American. The terms Latino and Hispanic, which refer to all persons of Spanish cultural background, are used because many statistics are collected and organized under those categories.   2.  For greater detail of this experience, see Raul Morin, Among the Valiant (Los Angeles, Borden Publishing, 1963).  3. Despite the movement out of the Southwest, the 1980 census indicated that eighty-five percent of Mexican-Americans still lived in the five southwestern states.   4.  Manuel P. Servin, “The Mexican-American Awakens: An Interpretation, Journal of the West 14 (October 1975): 121–130. Ricardo Romo, “The Urbanization of Southwestern Chicanos in the Early 20th Century,” New Scholar 6 (1977): 183–207.   5.  For more details about this important Mexican-American organization see Allsup, Carl, The American G.I. Forum: Origin and Evolution (Austin: University of Texas Press, 1982).  6. Carlos Larralde, Mexican American: Movements and Leaders (Los Alamitos, CA: Hwong Publishing, 1976). There is a chapter on Bert Corona.   7.  I do not see leadership as an elitist or cacique issue. If la raza is to obtain its due in American society, it must develop organization, and organization means leadership.   8.  For the early impact of the Delano Grape Strike on Chicanos see Ralph Guzmán and Joan Moore, “The Mexican Americans: New Wind from the Southwest,” Nation, Vol. CCII, May 30, 1966, pp. 645–648.   9.  The UFW charges that in the recent past over three hundred field workers have been poisoned by toxic sprays, which also pose a health threat to consumers. 10.  For an excellent general survey of the unionizing of farm labor see Anne Loftis and Dick Meister, A Long Time Coming: The Struggle to Unionize America’s Farm (New York: Macmillan, 1977). Ronald Taylor, Chávez and the Farm Workers (Boston: Beacon Press, 1975). One of a number of good accounts of Chávez and the grape strike. 11. For a sympathetic account of Gonzales’s early career see Christine Marin, A Spokesman of the Mexican American Movement: Rodolfo “Corky” Gonzales and the Fight for Chicano Liberation, 1966–1972 (San Francisco: R & E Research, 1977).

Chapter Notes   413 12. Patricia Bell Blawis, Tijerina and the Land Grants: Mexican Americans in the Struggle for Their Heritage (New York, International Publishers, 1971). Tijerina has also written a personal account, Mi lucha por la tierra (1978). 13.  Gutiérrez discusses his political ideas in La Raza and Revolution (San Francisco: R & E Associates, 1972). For a more detailed factual account of La Raza Unida party see Richard Santillan, Chicano Politics: La Raza Unida (Los Angeles: Tlaquilo, 1973).

Chapter 18   1.  Note the change in statistical basis. Because many government statistics and those of private organizations are collected and organized in terms of Hispanics/Latinos rather than Mexican-Americans, it is necessary at times to use this broader category. The reader is cautioned to note these differences in statistical base and to remember that overall MexicanAmericans make up about two-thirds of all Latinos. If the statistics pertain only to the Southwest, they are a much larger percentage, of course.  2. For greater detail see Alejandro Portes and John W. Curtis, “Changing Flags: Naturalization and Its Determinants among Mexican Immigrants,” International Migration Review 21, no. 2 (Summer 1987): 352–371.   3.  F. Chris García and Rudolph O. de la Garza, The Chicano Political Experience: Three Perspectives (North Scituate, MA: Duxbury Press, 1977).  4. Leonard Dinnerstein and David M. Reimers, Ethnic Americans: A History of Immigration (New York, Dodd and Mead, 1988): p. 123.  5. El Plan de Santa Bárbara (1969).   6.  For a broad account of the student movement, especially the 1968 walkouts, see “Chicano Student Militancy: The Los Angeles High School Strike of 1968,” by Kaye Briegel, 215–225, in Manuel P. Servín, An Awakened Minority: The Mexican Americans (Beverly Hills: Glencoe, 1974).  7. Armando Morales gives an intensely personal account of the riots in Ando Sangrando (La Puente, CA: Perspectiva Publications, 1973).   8.  For an account by the founder of the Brown Berets, see Expedition Through Aztlan by David Sánchez (La Puente, CA: Perspectiva Publications, 1978). Chris Marín gives an excellent account in “Go Home, Chicanos: A Study of the Brown Berets in California and Arizona,” pp. 226–246 in An Awakened Minority: The Mexican-Americans by Manuel P. Servín (1974).   9.  For greater detail see The Chicano and the Church by Antonio R. Soto (1975). 10. “Manuela Solís Sager and Emma Tenayuca: A Tribute,” by Roberto Calderón and Emilio Zamora, pp. 30–41, in Chicana Voices: Intersections of Class, Race, and Gender (1986). 11. See Occupied America: A History of Chicanos by Rodolfo Acuña (1988): pp. 332– 333 and 345. 12.  In addition to the books about Chicanas in the Additional Notes Section of this book, the reader might look at Chicanas Speak Out: Women, the New Voice of La Raza by Mirta Vidal (1971). 13. See The Identification and Analysis of Chicano Literature, Francisco Jiménez, editor, for a detailed study of the literary movement (1979).

414   Chapter Notes 14.  The outstanding work on Chicano art is Mexican American Artists by Jacinto Quirarte (1973). 15.  For greater detail of the bilingual issue see “Chicano Bilingualism,” by Rosaura Sánchez pp. 209–225 in New Scholar, Vol. VI (1977). Armando Rodríguez presents a forceful argument for bilingual and bicultural education in Speak Up, Chicano: The MexicanAmerican Fights for Educational Equality (1968). 16.  For a discussion of this change see “The Chicano Movement and the Mexican American Community, 1972–1978: An Interpretative Essay,” by Richard A. García, Socialist Review, Vol. VIII, July–October 1978, pp. 117–136. See also Anglos and Mexicans in the Making of Texas by David Montejano (1987). Chapter 13, “A Time of Inclusion” summarizes the changes that have taken place in Texas, especially in the past decade and a half.

Chapter 19   1.  This figure is based on numbers apprehended and returned to Mexico, not discrete persons. This is true of all similar estimates of undocumenteds.   2.  As with most words, the term bracero has several meanings. In its narrowest sense it refers to a Mexican laborer contracted to work in the United States under various government-to-government agreements, especially between 1942 and 1964. In a wider sense it may be applied to any Mexican national recruited to work in the United States, and its broadest meaning may include any Latino worker in the United States.  3. Coyotes refers to persons who organize smuggling operations at the border. A patero (“duck herder”) is a person who assists and guides people across the Rio Grande River, and a “pollero” (“chicken herder”) is a person who does the same across the land border.   4.  For a scholarly sociological point of view see Los Mojados: The Wetback Story by Julian Samora (1971).   5.  The reader is reminded that these numbers may include the same person several times.  6. See Operation Wetback: The Mass Deportation of Mexican Undocumented Workers in 1954 by Juan Ramón García (1980). It includes background to the issue.   7.  For an outstanding study of the braceros by a man who was deeply involved, see Merchants of Labor: The Mexican Bracero Story by Ernesto Galarza (1965).   8.  For a more detailed picture of the commuters, see The Border Crossers: People Who Live in Mexico and Work in the United Stater by David S. North (1970).  9. The Golden Door: International Migration, Mexico, and the United States by Paul Ehrlich, et al. (1979). 10. For greater detail see chapters 5, 6, and 7 in Ethnic Americans: A History of Immigration by Leonard Dinnerstein and David M. Reimers (1988). 11. Wayne Cornelius is an outstanding specialist in Mexican migration studies. See his Mexican Migration to the United States: Causes, Consequences, and the U.S. Response (1978). 12.  The reader is cautioned that U.S. Census statistics are not strictly comparable, as the decennial censuses have been based on different criteria to identify Mexican-Americans. Statistics on the immigration of Mexicans to the United States vary from source to source, even among U.S. government agencies, and must be used with caution.

Chapter Notes   415 13.  For further discussion of commuters, see Hired Hands: Seasonal Farm Workers in the United States by Steven H. Sosnick (1978). 14.  Estimates of the I&NS and other agencies have gone as high as twelve to fifteen million. The 1980 census counted approximately two million. A 1978–1979 Mexican Labor Ministry poll indicated that there were about one million sojourners, plus 250,000 to 350,000 “permanent” undocumenteds in the United States, about seventy-five percent of them in California and Texas. 15. For an excellent survey of undocumented migration see Building the Cactus Curtain: Mexican Migration and U.S. Responses, from Wilson to Carter by Wayne Cornelius (1980). See also Mexican Illegal Alien Workers in the United States by Walter Fogel (1978). 16.  The reader is again cautioned that the same person may be apprehended several times and thereby become three or four people in the statistics. 17.  See chapter 6 in The Border Economy by Niles Hanson (1981). 18.  For interesting details of the law and its passage, see “The Albatross of Immigration Reform: Temporary Worker Policy in the United States,” by Vernon Briggs, International Migration Review, Vol. XX, Winter 1986, pp. 995–1019. 19.  For greater detail, see Through the Maze: An Interim Report on the Alien Legalization Program by David S. North (1988).

Chapter 20   1.  Although Canada shares a border with the United States, the number of Canadian immigrants is vastly smaller than that of Mexicans.   2.  Mexicans often refer to the United States as El Norte—“the North.”   3.  Preceding this period of immigration, Mexicans lived in the upper regions of Mexico [the Southwest] until the United States defeated Mexico in the U.S.-Mexican War of 1848. Most Mexicans stayed, but many lost their lands and access to political power.  4. Tomás R. Jiménez, Replenished Ethnicity: Mexican American, Immigration and Identity (Berkeley: University of California Press, 2010): 31–65. The term “native-born” refers to the general non-Mexican population in the United States.  5. Oscar Handlin, The Uprooted: The Epic Story of the Great Migrations That Made the American People (Philadelphia: University of Pennsylvania Press, 2002): 3. The Uprooted was first published in 1951 and won the 1952 Pulitzer Prize in history.   6.  The U.S. Census Bureau defines “foreign born” as anyone who is not a U.S. citizen at birth, a naturalized U.S. citizen, a legal permanent resident, a temporary migrant, a humanitarian migrant or an unauthorized migrant. The term “unauthorized immigrant” replaces terms such as “illegal” to refer to immigrants entering a country, in this case, the United States, without legal papers. The term native refers to people residing in the United States who are U.S. citizens in one of three categories: (1) people born in one of the 50 states or the District of Columbia, (2) people born in the U.S. insular areas such as Puerto Rico or Guam, or (3) people born abroad of a U.S. citizen parent.  7. Alma M. García, Narratives of Second Generation Mexican American Women: Emergent Identities of the Second Generation (Walnut Creek, CA: AltaMira Press, 2004): 45.   8. “The Foreign-Born Population in the United States, 2010,” U.S. Census Bureau, accessed August 10, 2014, http://www.census.gov/prod/2012pubs/acs-19.pdf.

416   Chapter Notes   9.  Mark Granovetter, “The Strength of Weak Ties,” American Journal of Sociology 78, no. 6 (1973): 1360–1380. For a critique of Granovetter’s theory of social capital see Alejandro Portes, “The Downside of Human Capital,” Proceedings of the National Academy of Science 111, no. 52 (2014): 18407–18408. 10. García, Narratives,103–104. 11.  Richard Sennett and Jonathan Cobb, The Hidden Injuries of Class (New York: Knopf, 1972). 12.  “The Rise of Asian Americans,” Pew Research Center, June 19, 2013, accessed July 10, 2015, http://www.pewsocialtrends.org/2012/06/19/the-rise-of-asian-americans/; Xue Lan Ron and Russell Ron, Asian American Education: Identities, Racial Issues and Languages (Charlotte, NC: Information Age Publishing, 2011). 13. Jie Zong and Jeanne Batalova, “Mexican Immigrants in the United States,” Migration Policy Institute, October 9, 2014, accessed January 4, 2015, http://www.migration policy.org/article/mexican-immigrants-united-states/. 14.  Zong and Batalova, “Mexican Immigrants.” 15.  Sierra Stoney and Jeanne Batalova, “Mexican Immigrants in the United States,” Migration Policy Institute, February 28, 2013, accessed October 2, 2014, http://www.migrati onpolicy.org/article/mexican-immigrants-united-states-2. 16.  Stoney and Batalova, “Mexican Immigrants in the United States.” 17.  Lyndon B. Johnson, The State of the Union Address, January 8, 1964, accessed on August 12, 2014, http://www.lbjlib.utexas.edu/johnson/archives.hom/speeches.hom/6401 08.asp. 18.  Steven A. Camarota, “Immigrants in the United States, 2010: A Profile of America’s Foreign-Born Population,” Center for Immigration Studies, August 2012, accessed February 10, 2015, http://www.cis.org/sites/cis.org/files/articles/2012/immigrants-in-the-united -states-2012.pdf. 19.  Drew Desilver, “Who’s Poor in America? 50 years into the ‘War on Poverty,’ a Data Portrait,” Pew Research Center, January 13, 2014, accessed February 1, 2015, http://www .pewresearch.org/fact-tank/2014/01/13/whos-poor-in-america-50-years-into-the-war-on -poverty-a-data-portrait/. 20. Aaron Terrazas, “Immigrants in New-Destination States,” Migration Policy Institute, February 8, 2011, accessed October 2, 2014, http://www.migrationpolicy.org /article/immigrants-new-destination-states#15. 21.  Terrazas, “Immigrants in New-Destination States.” 22.  “Garret Warner, “Working Hands in Vermont’s Borderlands: A Guide to Resources for Migrant Workers in Addison County, Vermont,” Migrant Resources, accessed December 18, 2014, http://sites.middlebury.edu/migrantresources/. 23.  “Migration to Vermont,” Peace and Justice Center, August 24, 2014, accessed July 8, 2015, http://vtdigger.org/2014/08/24/peace-justice-center-statement-response-shooting -michael-brown-racism-vermont/. 24.  Marcela Mendoza, David H. Ciscel, and Barbara Ellen Smith, “Latino Immigrants in Memphis, Tennessee: Their Local Economic Impact,” Working Paper Series Center for Research on Women 15 (2001): 1–16; N. E. Cantu, “Report on Latino Culture and Traditional Arts in Tennessee,” Tennessee Arts (1999): 40–51.

Chapter Notes   417 25.  Nestor Rodríguez, “New Southern Neighbors: Latino Immigration and Prospects for Intergroup Relations between African-Americans and Latinos in the South,” Latino Studies 10 (Spring 2012): 18–40. 26. Mary E. Odem and Elaine Cantrell Lacy, eds., Latino Immigrants and the Transformation of the U.S. South (Athens: University of Georgia Press, 2009): ix. 27. Carl l. Bankston, “New People in the New South: An Overview of Southern Immigration,” Southern Cultures 13 (Winter 2007): 24–44. 28.  Bankston, “New People,” 35. 29.  Ronald Bayor, Race and the Shaping of Twentieth Century Atlanta (Chapel Hill: University of North Carolina Press, 1996). 30.  Mary E. Odem, “Our Lady of Guadalupe in the New South: Latino Immigrants and the Politicas of Integration in the Catholic Church,” Journal of American Ethnic History 24 (Fall 2004): 26. 31.  The Catholic Church faced a similar situation when the Irish arrived in the East in such places as New York and Boston, a migration triggered by the Irish potato blight beginning around 1850 and continuing well into the early twentieth century. 32.  Odem, “Our Lady of Guadalupe in the New South,” 27. 33.  Interestingly, the majority of African-Americans were Protestants but a small minority were Catholics. They were forced to practice their religion in racially segregated churches and schools; Odem, “Our Lady.” 34. Odem, “Our Lady,” 30; Daniel Ramirez, “Borderlands Praxis: The Immigrant Experience in Latino Pentecostal Churches,” Journal of the American Academy of Religion 67 (September 1999): 573–596; Arlene M. Sánchez-Walsh, Latino Pentecostal Identity: Evangelical Faith, Self, and Society (New York: Columbia University Press, 2003). 35.  Odem, “Our Lady,” 50. 36.  Jens Manuel Krogstad and Jeffrey S. Passel, “5 Facts About Illegal Immigration in the U.S.,” Pew Research Center, July 24, 2015, accessed October 10, 2015, http://www .pew­research.org/fact-tank/2015/07/24/5-facts-about-illegal-immigration-in-the-u-s/. 37.  Stoney and Batalova, “Mexican Immigration in the United States.” 38.  Jeffrey H. Passel and D’Vera Cohn, “Why Wave of Mexican Immigration stopped,” Online CNN Special Report, April 26, 2012, accessed May 2, 2015. http://www.cnn .com/2012/04/26/opinion/passel-cohn-mexican-immigration/. 39. “Testimony of Jeffrey H. Passel–Unauthorized Immigrant Population,” Pew Research Center, March 26, 2015, accessed October 15, 2015. http://www.pewhispanic .org/2015/03/26/testimony-of-jeffrey-s-passel-unauthorized-immigrant-population/. 40.  Mark Hugo Lopez and Paul Taylor, “Latino Voters and the 2012 Election,” Pew Research Center, November 7, 2012, accessed April 10, 2015, http://www.pewhispanic .org/2012/11/07/latino-voters-in-the-2012-election/. 41. Ana Gonzalez-Barrera and Jens Manuel Krogstad, “U.S. Deportations of Immigrants Reach Record High in 2013,” Pew Research Center, October 2, 2014, accessed March 25, 2015, http://www.pewresearch.org/fact-tank/2014/10/02/u-s-deportations-of -immigrants-reach-record-high-in-2013/. 42.  Gonzalez-Barrera and Krongstad, “U.S. Deportations.” 43.  Stoney and Batalova, “Mexican Immigration in the United States.”

418   Chapter Notes

Chapter 21  1. La Frontera (“the frontier”) is used by Mexicans to refer to the border between the United States and Mexico.   2.  William Shakespeare, The Tempest, Act V, Scene I: O, wonder! How many goodly creatures are there here! How beauteous mankind is! O brave new world. That has such people in’t!  3. Méjico is Spanish for Mexico. El Norte translates as “the North,” but is used by Mexicans to refer geographically to the United States.  4. Oscar Handlin, The Uprooted: The Epic Story of the Great Migrations That Made the American People (New York: Grosset & Dunlap, 1959).  5. Ronald Takaki, Strangers from a Distant Shore: A History of Asian Americans (Boston: Little and Brown, 1998).  6. Ernesto Galarza, Barrio Boy (South Bend, IN: University of Notre Dame Press, 1971): 183.  7. Francisco Jiménez, The Circuit: Stories From the Life of a Migrant Child (Albuquerque: University of New Mexico Press, 1997): 1.  8. Gloria Anzaldúa, Borderlands/La Frontera: The New Mestiza (San Francisco: Aunte Lute Books, 1987); Ana Castillo, The Mixquiahuala Letters (Binghamton, NY: Bilingual Press, 1986).   9.  Prior to landing, in 1630, John Winthrop addressed the others aboard the Arabella in an effort to regain their enthusiasm after the Atlantic crossing that had sorely tested them. Winthrop used the term “city on a hill,” which is originally found the Christian gospel’s story of Christ’s Sermon on the Mount. Winthrop used this as a metaphor rally to capture the spirit of the Puritan’s goal: to bear witness to the world that they were creating a new society to be emulated by others. 10.  See Kenneth Stampp, The Peculiar Institution: Slavery in the Ante-Bellum South (New York: Knopf, 1956); Guadalupe San Miguel, Let Them All Take Heed: Mexican Americans and the Campaign for Educational Equity in Texas (Austin: University of Texas Press, 1987); Sucheng Chan, Asian Americans: An Interpretive History (New York: Twayne Publishers, 1991); George Horse and Duane Champagne, American Indian Nations: Yesterday, Today, and Tomorrow (Walnut Creek, CA: Altamira Press, 2007); Ellen Carol Dubois, Harriot Stanton Blatch and the Winning of Woman Suffrage (New Haven, CT: Yale University Press, 1997). 11.  Norma Iglesias, The Most Beautiful Flower of the Maquiladoras: Life Histories of Women Workers in Tijuana, 2nd ed., trans. Michael Stone and Gabrielle Winkler (Austin: University of Texas Press, 1997). 12. Iglesias, The Most Beautiful Flower of the Maquiladoras. 13. Douglas S. Massey, Jorge Durand, and Nolan J. Malone, Beyond Smoke and Mirrors: Mexican Immigration in an Era of Economic Integration (New York: Russell Sage Foundation, 2003); Lisa Breglia, Promises, Peaks and Declines on Mexico’s Gulf Coast, (Austin: University of Texas Press, 2013).

Chapter Notes   419 14.  Alma M. García, The Mexican Americans (Westport, CT: Greenwood Press, 2002): 47–49. 15.  Warren R. Leiden and David L. Neal, “Highlights of the U.S. Immigration Act of 1990,” Fordham International Law Journal 14 (1990): 328–339; Massey, Beyond Smoke and Mirrors, 91–93. 16. Massey, Smoke and Mirrors, 91–93. 17.  George Lipsitz, The Possessive Investment in Whiteness: How White People Profit from Identity Politics, revised and expanded ed. (Philadelphia: Temple University Press, 2006). 18.  “Plyler v. Doe,” Legal Information Institute, accessed March 10, 2015, http://www .law.cornell.edu/supremecourt/text/457/202. 19. Jerry Brown Jr. served as governor of California during 1975–1983 and was elected again in 2011. He won reelection in 2014 to an unprecedented fourth term. 20.  Martin Miller, “Proposition 187: Fund-Raiser by Supporters Draws 100 in Orange County,” Los Angeles Times, October 29, 1994. 21. B. Drummond Ayres Jr., “The 1994 Campaign: California; Minorities Join California Fight.” The New York Times, November 1, 1994. 22.  Patrick J. McDonnell and Robert López, “L.A. March Against Prop. 187 Draws 70,000: Immigration: Protesters Condemn Wilson for Backing Initiative That They Say Promotes ‘Racism, Scapegoating,’” Los Angeles Times, October 17, 1994. 23.  McDonnell and López, “L.A. March Against Prop. 187.” 24.  McDonnell and López, “L.A. March Against Prop. 187.” 25.  Mario T. García and Sal Castro, Blowout: Sal Castro & the Chicano Struggle for Educational Justice (Chapel Hill: University of North Carolina Press, 2011). 26.  Amy Pyle and Beth Shuster, “10,000 Students Protest Prop. 187,” Los Angeles Times, November 3, 1994. 27.  Brett Tam, “Marchers Call “Save Our “State Initiative Racist,” Daily Bruin, March 1, 1994. 28.  David G, Gutiérrez, Walls and Mirrors: Mexican Americans, Mexican Immigrants, and the Politics of Ethnicity (Berkeley: University of California Press, 1995); Lina Yvette Newton, “Why Latinos Supported Proposition 187: Testing the Economic Threat and Cultural Identity Hypotheses,” Social Science Quarterly 81 (2000). 29. See the organization’s website at http://lulac.org/about/history; See also San Miguel, 1987; Cynthia E. Orozco, No Mexicans, Women, or Dogs Allowed: The Rise of the Mexican American Civil Rights Movement (Austin: University of Texas Press, 2009). 30. García, The Mexican Americans, 120–121. 31. Ignacio M. García., Viva Kennedy: Mexican Americans In Search of Camelot (College Station: Texas A&M Press, 2000). 32. Melanie Mason and Patrick McGreevy, “Latino Lawmakers Move to Reverse Decades of Anti-Immigration Legislation,” Los Angeles Times, June 22, 2014. 33. García, The Mexican Americans, 120–121. 34. “Unauthorized Immigrant Population Trends for States, Birth Countries and Regions.” Pew Research Center, December 11, 2014, accessed March 24, 2015, http://www .pewhispanic.org/2014/12/11/unauthorized-trends/. 35.  The Yearbook of Immigration Statistics, Department of Homeland Security, last modified September 3, 2015, http://www.dhs.gov/yearbook-immigration-statistics#.

420   Chapter Notes According to the Department of Homeland Security website, “The Yearbook of Immigration Statistics is a compendium of tables that provides data on foreign nationals who, during a fiscal year, were granted lawful permanent residence (i.e., admitted as immigrants or became legal permanent residents), were admitted into the United States on a temporary basis (e.g., tourists, students, or workers), applied for asylum or refugee status, or were naturalized. The Yearbook also presents data on immigration enforcement actions, including alien apprehensions, removals, and returns. The Yearbook tables are released as they become available. A final PDF is released in September of the following fiscal year.” 36.  For the complete text of the Fourteenth Amendment to the Constitution, see http://www.loc.gov/rr/program/bib/ourdocs/14thamendment.html. 37.  Jens Manuel Krogstead and Jeffrey S. Passel, “5 Facts about Illegal Immigration in the U.S.,” Pew Research Center, November 18, 2014, accessed January 30, 2015, http://www .pewresearch.org/fact-tank/2014/11/18/5-facts-about-illegal-immigration-in-the-u-s/. 38. David Gutierrez, Jeanne Batalova, and Aaron Terrazas, “The 2012 Mexican Presidential Election and Mexican Immigrants of Voting Age in the United States,” Migration Policy Institute, April 26, 2012, accessed November 10, 2014, http://www .migrationpolicy.org/article/2012-mexican-presidential-election-and-mexican -immigrants-voting-age-united-states#4. 39.  “Mexico Passes Law on Dual Citizenship,” New York Times, December 12, 1996, accessed June 10, 2015, http://www.nytimes.com/1996/12/12/world/mexico-passes-law -on-dual-citizenship.html. 40.  Jennifer Mena, “Mexican Election Heating Up in the U.S.,” Los Angeles Times, May 13, 2004, accessed February 26, 2015, http://articles.latimes.com/print/2004/may/13/local /me-mexte13. 41.  Jennifer Mena, “Mexican Election,” Los Angeles Times, May 13, 2004, accessed February 10, 2015, http://articles.latimes.com/print/2004/may/13/local/me-mexvote13. 42.  Mena, “Mexican Election.” 43  Mena, “Mexican Election.” 44. Aurelia Fueros, “Absentee Voter Registration Begins for Mexicans in the U.S.,” Huffington Post Latino Voices, October 5, 2011, accessed July 14, 2015, http://www .huffingtonpost.com/2011/10/04/mexico-absentee-voter-registration_n_995188.html. 45.  “Mexico Election: Mexican Immigrants Shocked, Wary of PRI Victory,” Fox News Latino, July 3, 2012, accessed July 10, 2015, http://latino.foxnews.com/latino/politics /2012/07/03/immigrants-express-shock-at-return-mexico-pri/. 46.  Gutiérrez, Batalova, and Terrazas, “The 2012 Mexican Presidential Election.” 47.  “H.R. 4437 Border Protection, Antiterrorism, and Illegal Immigration Control Act of 2005,” Congressional Budget Office, December 13, 2005, accessed December 12, 2014, https://epic.org/privacy/surveillance/spotlight/0406/cbo4437.pdf. 48.  F. Arturo Rosales, Chicano: A Documentary History of the Mexican American Struggle for Civil Rights (Houston: Arte Público Press, 1997); Carlos Muñoz, Youth, Identity, Power: The Chicano Movement (New York: Verso, 1989); Alma M. García, Francisco Jiménez, and Richard A. Garcia, eds., Ethnic Community Builders: Mexican Americans’ Struggle for Justice, Power and Citizen Rights (Walnut Creek, CA: Altamira Press, 2007). 49.  Ted Wang and Robert C. Winn, “Groundswell Meets Groundwork: Building on the Mobilization to Empower Immigrant Communities,” in Rallying for Immigrants Rights:

Chapter Notes   421 The Fight for Inclusion in 21st Century America, ed. Kim Voss and Irene Bloemraad (Berkeley: University of California Press, 2011): 44–59. 50.  Roberto Suro, “Out of the Shadows into the Light: Questions Raised by the Spring of 2006,” in Rallying for Immigrants Rights: The Fight for Inclusion in 21st Century America, ed. Kim Voss and Irene Bloemraad (Berkeley: University of California Press, 2011). 51.  Gary McGoin, Sanctuary: A Resource Guide for Understanding and & Participating in the Central America Refugees’ Struggle (Harper & Row, San Francisco, 1985): 14–29. 52.  Roger Mahony, “A Nation That Should Know Better,” Los Angeles Times, June 1, 2005. 53.  García and Castro, Blowout. 54.  Wang and Winn, “Groundswell Meets Groundwork.” 55.  Irene Bloemraan, Kim Voss, and Taeku Lee, “The Protests of 2006: What Were They, How Do We Understand Them, Where Do We Go?” in Rallying for Immigrants Rights: The Fight for Inclusion in 21st Century America, ed. Kim Voss and Irene Bloemraad (Berkeley: University of California Press, 2011), 3–43. 56.  Bloemraan, Voss and Lee, “The Protests of 2006,” 32. 57.  García and Castro, Blowout. 58. Alfonso Gonzales, Reform Without Justice: Latino Migrant Politics and the Homeland Security State (New York: Oxford University Press, 2014). 59. Gonzales, Reform Without Justice. 60. Gonzales, Reform Without Justice. 61.  See Jonathan Fox and Xóchil Bada, “Migrant Civic Engagement,” Research Paper Series on Latino Immigrant Civic and Political Participation, no. 3 (Washington, D.C.: Woodrow Wilson Center for Scholars, 2009), http://www.wilsoncenter.org/sites/default/ files/Fox%20%26%20Bada%20-%20Migrant%20Civ%20Engagement%202008.pdf. 62.  Denise A. Segura and Patricia Zavella, eds., Women and Migration in the U.S.Mexico Borderlands (Durham, NC: Duke University Press, 2007); Vicki L. Ruiz, From Out of the Shadows: Mexican Women in Twentieth Century America (New York: Oxford University Press, 2008); Patricia Zavella, I’m Neither Here Nor There: Mexicans’ Quotidian Struggles with Migration and Poverty (Durham, NC: Duke University Press, 2011). 63. Monica Boyd and Elizabeth Griego, “Women and Migration: Incorporating Gender into International Migration Theory,” Migration Policy Institute, March 1, 2003, accessed March 10, 2015, http://migrationpolicy.org/article/women-and-migration-incorp orating-gender-international-migration-theory, 2. 64.  Eithne Luibheid, “Looking Like a Lesbian: The Organization of Sexual Monitoring at the United States-Mexican Border,” in Women and Migration in the U.S.-Mexico Borderlands, ed. Denise A. Segura and Patricia Zavella (Durham, NC: Duke University Press, 2007), 106–133. 65.  The word bracero derives from brazo, the Spanish word for “arm,” and is used to refer to a Mexican laborer admitted to the United States especially for seasonal contract labor in agriculture. 66.  Pierrette Hondagneu-Sotelo and Ernestine Avila, “I’m Here, but I’m There: The Meanings of Latina Transnational Motherhood,” In Women and Migration in the U.S.Mexico Borderlands, ed. Denise A. Segura and Patricia Zavella (Durham, NC: Duke University Press, 2007), 388–412. Similar trends can be seen in the migration of domestic workers in the Middle East and Europe.

422   Chapter Notes 67.  Boyd and Griego, “Women and Migration.” 68.  Denise A. Segura and Patricia Zavalla, eds., Women and Migration in the U.S.Mexico Borderlands (Durham, NC: Duke University Press, 2007); Silvia Pérez, “Women and Migration: The Social Consequences of Gender,” Annual Review of Sociology 17 (1991): 303–325; Marta Tined and Karen Booth, “Gender, Migration and Social Change,” International Sociology 6 (1991): 51–72. 69.  Hondagneu-Sotelo and Avila, “I’m Here, but I’m There.” 70.  Hondagneu-Sotelo and Avila, “I’m Here, but I’m There.” 71. Mark Ensalco, “Murder in Ciudad Juárez: A Parable of Women’s Struggle for Human Rights,” Violence Against Women 12 (2012): 41–440. 72.  Jessica Livingston, “Murder in Juárez: Gender, Sexual Violence, and the Global Assembly Line,” Frontiers: A Journal of Women Studies 25 (2004): 60–61. 73.  Livingston, “Murder in Juárez.” 74.  Emma O’Connor, “Mexico’s Cuidad Juárez Is No Longer the Most Violent City in the World,” Time, October 15, 2012, accessed April 21, 2015, http://newsfeed.time. com/2012/10/15/mexicos-ciudad-juarez-is-no-longer-the-most-violent-city-in-the -world/. 75.  O’Connor, “Mexico’s Cuidad Juárez.” 76. Anahi Rama and Lizbeth Díaz, “Violence against Women Is a Pandemic in Mexico,” Reuters, May 7, 2014, accessed April 20, 2015, http://www.reuters.com/article /2014/03/07/us-mexico-violence-women-idUSBREA2608F20140307. 77.  Livingston, “Murder in Juárez.” 78.  Katie L. Acosta, “Lesbians in the Borderlands: Shifting Identities and Imagined Communities,” Gender & Society 22 (2008): 639. See also Emma Pérez, “Sexuality and Discourse: Notes from a Chicana Survivor,” in Chicana Lesbians: The Girls our Mothers Warned Us About, ed. Carla Trujillo (Berkeley, CA: Third Woman Press, 1991). 79.  Acosta, “Lesbians in the Borderlands.” 80. Rigoberta Menchu, Elisabeth Burgos-Debray, I, Rigoberta Menchu: An Indian Woman in Guatemala, 2nd ed. (New York: Verso, 2010); Elvia Alvarado, Don’t Be Afraid, Gringo: A Honduran Woman Speaks from The Heart: The Story of Elvia Alvarado, trans. Medea Benjamin (New York: Harper Perennial, 1989). 81.  Ensalco, “Murder in Ciudad Juárez.” 82.  “Señorita Etraviada: Film Description,” PBS, accessed March 25, 2015, http://www .pbs.org/pov/senoritaextraviada/film_description.php. 83. William Paul Simmons, Cecelia Menjivar, and Michelle Téllez, “Violence and Vulnerability of Female Migrants in Drop Houses in Arizona: The Predictable Outcome of a Chain Reaction of Violence,” Violence Against Women (2015): 552–570. 84.  Simmons, Menjivar and M. Téllez, “Violence and Vulnerability.”

Chapter 22  1. Alma M. García, Narratives of Second Generation Mexican American Women: Emergent Identities of the Second Generation (Walnut Creek, CA: AltaMira Press, 2004).  2. Mary Waters, Ethnic Options: Choosing Identities in America (Berkeley: University of California Press, 1990).

Chapter Notes   423  3. García, Narratives.   4.  Joane Nagel, “Constructing Ethnicity: Creating and Recreating Ethnic Identity and Culture.” Social Problems 41(1994): 152–176.  5. García, Narratives  6. García, Narratives.  7. García, Narratives.  8. Nazli Kibria, Family Tightrope: The Changing Lives of Vietnamese Americans (Princeton, NJ: Princeton University Press, 1993); Pyong Gap Min, Second Generation: Ethnic Identity among Asian Americans (Walnut Creek, CA: AltaMira Press, 2002); Rubén Rumbaut and Alejandro Portes, eds. Ethnicities of Children of Immigrants (Berkeley: University of California Press, 2001).  9. García, Narratives, 56 10.  Vicki L. Ruiz, From Out of the Shadows: Mexican American Women in Twentieth Century America (New York: Oxford University Press, 1998). Used by permission. 11. “The Mexican-American Boom: Births Overtake Immigration,” Pew Research Center, July 14, 2011, http://www.pewhispanic.org/files/reports/144.pdf. 12.  Jie Zong and Jeanne Batalova, “Frequently Requested Statistics on Immigrants and Immigration to the United States,” Migration Policy Institute, February 26, 2015, accessed July 27, 2015, http://www.migrationpolicy.org/article/frequently-requested-statistics -immigrants-and-immigration-united-states. 13. Guadalupe San Miguel, Let Them All Take Heed: Mexican Americans and the Campaign for Educational Equality in Texas, 1910 to 1981 (Austin: University of Texas Press, 1971). 14. Michael L. Olivas, No Undocumented Child Left Behind: Plyler v. Doe and the Education of Undocumented School Children (New York: New York University Press, 2012). 15.  Michael A. Olivas, “Plyler v. Doe: Still Guaranteeing Unauthorized Immigrant Children’s Right to Attend U.S. Public Schools,” Migration Policy Institute, September 9, 2010, accessed September 1, 2015, http://www.migrationpolicy.org/article/plyler-v-doe -still-guaranteeing-unauthorized-immigrant-childrens-right-attend-us-public. 16.  Olivas, “Plyer v. Doe.” 17.  Olivas, “Plyler v. Doe.” 18.  “Plyler v. Doe,” Legal Information Institute, Cornell University, accessed September 10, 2015, https://www.law.cornell.edu/supremecourt/text/457/202#writing-USSC_CR_04 57_0202_ZO. 19.  “Plyler v. Doe,” Legal Information Institute. 20. Walter J. Nicholls, The DREAMers: How the Undocumented Youth Movement Transformed the Immigrant Rights Debate (Stanford, CA: Stanford University Press, 2013). 21. Nicholls, The DREAMers, 7. 22. Nicholls, The DREAMers, 13–14. 23. Nicholls, The DREAMers, 53. 24.  United We Dream, http://unitedwedream.org. 25.  United We Dream, http://unitedwedream.org. 26.  William A. Schwab, Right to Dream: Immigration Reform and America’s Future (Fayetteville: University of Arkansas Press, 2013).

424   Chapter Notes 27.  The White House, “The Dream Act: Good for the Economy, Good for our Security, Good for our Nation,” accessed September 12, 2015, https://www.whitehouse.gov/sites/ default/files/DREAM-Act-WhiteHouse-FactSheet.pdf. 28. Elise Foley, “Dream Act Fails in Senate,” Huffington Post, December 18, 2010, accessed October 1, 2015. 29. Jeffrey Passel, “Latino and Asian Voters in the 2004 Election and Beyond,” Migration Policy Institute, November 1, 2004, accessed September 17, 2015, http://www .migrationpolicy.org/article/latino-and-asian-voters-2004-election-and-beyond. 30.  Passel, “Latino and Asian Voters.” 31.  Matt Barreto and Gary M. Segura, Latino America: How America’s Most Dynamic Population Is Poised to Transform the Politics of the Nation (New York: Public Affairs, 2014): 3 32.  Barreto and Segura, Latino America. 33. Leslie Wayne, “Richardson Drops Out of Democratic Race,” New York Times, January 10, 2008, accessed September 17, 2015, http://www.nytimes.com/2008/01/10/us /politics/10cnd-richardson.html?_r=0. 34. Susan Minushkin and Mark Hugo Lopez, “The Hispanic Vote in the 2008 Democratic Presidential Primaries,” Pew Research Center, March 7, 2008, accessed September 18, 2015, http://www.pewhispanic.org/2008/03/07/the-hispanic-vote-in-the -2008-democratic-presidential-primaries/. 35. Mark Hugo Lopez, “The Hispanic Vote in the 2008 Elections,” Pew Hispanic Center, November 7, 2008, accessed September 18, 2015, http://www.pewhispanic .org/2008/11/05/the-hispanic-vote-in-the-2008-election/. 36. For a slide show of newspaper headlines after Barack Obama won the 2008 presidential election, see http://www.huffingtonpost.com/2008/11/05/obamas-victory-on -newspap_n_141311.html. 37.  Lopez, “The Hispanic Vote.” 38.  Luis R. Fraga, John A. Garcia, Rodney E. Hero, Michael Jones-Correa, Valerie Martinez-Ebers, and Gary M. Segura, “Su Casa Es Nuestra Casa: Latino Politics Research and the Development of American Political Science,” American Political Science Review 100, no. 4 (2006): 515–522. 39.  Lopez, “The Hispanic Vote.” 40.  Barreto and Segura, Latino America; Ricardo Ramírez, Mobilizing Opportunities: The Evolving Latino Electorate and the Future of American Politics (Charlottesville: University of Virginia Press, 2013). 41. Maggie Rivas-Rodriguez, ed., Mexican Americans and World War II (Austin: University of Texas Press, 2005); Richard Griswold del Castillo, ed., World War II and Mexican American Civil Rights (Austin: University of Texas Press, 2010). 42.  Hector Amaya, “Dying American or the Violence of Citizenship: Latinos in Iraq,” Latino Studies 5 (2007): 3–24. 43.  Barreto and Segura, Latino America. 44. “Survey on Latino Attitudes on the War in Iraq,” Pew Research Hispanic Trends Project, February 7, 2005, accessed September 17, 2015, http://www.pewhispanic .org/2005/02/07/survey-on-latino-attitudes-on-the-war-in-iraq/. 45.  Barreto and Segura, Latino America.

Chapter Notes   425 46. “Inside Obama’s Sweeping Victory,” Pew Research Center, November 5, 2008, accessed September 20, 2015, http://www.pewresearch.org/2008/11/05/inside-obamas -sweeping-victory/. 47.  Barreto and Segura, Latino America. 48.  Barreto and Segura, Latino America. 49. Arizona, Senate Bill 1070 (2010), http://www.azleg.gov/legtext/49leg/2R/bills/SB 1070S.pdf. 50. Arizona, Senate Bill 1070. 51. Rogelio Saenz, “Latinos, Whites, and the Shifting Demography of Arizona.” Population Reference Bureau, September 20, 2010, accessed September 23, 2015, http:// www.prb.org/Publications/Articles/2010/usarizonalatinos.aspx. 52.  Tanya Maria Golash-Boza, Immigration Nation: Raids, Detentions in Post-9/11 America (Boulder, CO: Paradigm Publishers, 2012). 53. Arizona, Senate Bill 1070. 54. Golash-Boza, Immigration Nation. 55. Golash-Boza, Immigration Nation. 56.  Jeremy Duda, “Clergy Group Files First Suit Against SB 1070,” Arizona Capitol Times, April 29, 2010, accessed September 23 2015, http://azcapitoltimes.com /news/2010/04/29/clergy-group-files-first-suit-against-s1070/. 57.  Article 2 Section 2 of the Arizona Constitution reads, “All political power is inherent in the people, and governments derive their just powers from the consent of the governed, and are established to protect and maintain individual rights.” accessed September 3, 2015, http://www.azleg.gov/Constitution.asp?Article=2 58. “Arizona v. United States,” Legal Information Institute, Cornell University, accessed September 24, 2015, https://www.law.cornell.edu/supremecourt/text/11-182. 59.  Mark Hugo Lopez, Rick Morin, and Paul Taylor, “Illegal Immigration Backlash Worries, Divides Latinos,” Pew Hispanic Center, October 28, 2010, accessed September 22, 2015, http://www.pewhispanic.org/2010/10/28/illegal-immigration-backlash-worries -divides-latinos/. This report does not disaggregate Latino groups but, given all demographic profiles, Mexicans and Mexican-Americans make up the majority of undocumented immigrants. 60.  Lopez, Morin and Taylor, “Illegal Immigration Backlash.” 61.  Interview with high school students (anonymous) by Alma M. García, August 10, 2014. 62. Nicholls, The DREAMers. 63. Nicholls, The DREAMers. 64. “Top 10 of 2012—Issue #2: Obama Administration Action Benefitting DREAMERers a Game-Changer in U.S. Immigration Debate,” Migration Policy Institute, December 1, 2012, accessed September 29, 2015, http://www.migrationpolicy.org/article /top-10-2012-issue-2-obama-administration-action-benefitting-dreamers-game-changer -us/. 65. “Top 10 of 2012—Issue #2: Obama Administration Action Benefitting DREAMERers.” 66.  “Deferred Action for Childhood Arrivals,” Department of Homeland Security, accessed October 3, 2015, http://www.dhs.gov/deferred-action-childhood-arrivals.

426   Chapter Notes 67.  Jeanne Batalova and Michelle Mittelstadt, “Relief from Deportation: Demographic Profile of the DREAMers Potentially Eligible Under the Deferred Action Policy,” Migration Policy Institute, August, 2012, accessed October 1, 2015, http://www.migrationpolicy.org /research/DACA-deferred-action-DREAMers. 68.  Mark Victor Hugo, “Latino Voters in the 2012 Election,” Pew Research Center, November 7, 2012, accessed September 14, 2015, http://www.pewhispanic.org/2012/11/07 /latino-voters-in-the-2012-election/. 69. Cindy Y. Rodriguez, “Latino Vote Key to Obama’s Re-election,” CNN online, November 9, 2012, accessed July 10, 2015, http://www.cnn.com/2012/11/09/politics /latino-vote-key-election/. 70.  Rodriguez, “Latino Vote.” 71.  Hugo, “Latino Voters.” 72.  Hugo, “Latino Voters.” 73.  Hugo, “Latino Voters”; Mark Hugo Lopez and Ana Gonzalez-Barrera, “Inside the 2012 Latino Electorate: II. Dissecting the Latino Electorate,” Pew Research Center, May 31, 2013, accessed October 1, 2015, http://www.pewhispanic.org/2013/05/31/ii-dissecting-the -latino-electorate/. 74. Margie McHugh, “Diploma, Please: Promoting Educational Attainment for DACA- and Potential Dream Act-Eligible Youth,” Migration Policy Institute September 14, 2014, accessed October 1, 2015, http://www.migrationpolicy.org/research/diploma -please-promoting-educational-attainment-daca-and-potential-dream-act-eligible-youth. 75.  McHugh, “Diploma Please.” 76.  Jerry Markon and Sandhya Somashekhar, “Obama’s 2012 DACA Move Offers a Window into Pros and Cons of Executive Action,” Washington Post, November 11, 2014, accessed September 10, 2015, https://www.washingtonpost.com/politics/obamas -2012-daca-move-offers-a-window-into-pros-and-cons-of-executive-action/2014/11 /30/88be7a36-7188-11e4-893f-86bd390a3340_story.html. 77. “Fixing the System: President Obama Is Taking Action on Immigration,” White House, November 20, 2014, accessed October 4, 2015, https://www.whitehouse.gov/issues /immigration/immigration-action#. 78.  “Executive Actions on Immigration, Department of Homeland Security, accessed October 11, 2015, http://www.uscis.gov/immigrationaction. 79. Drew Desliver, “Executive Actions on Immigration Have Long History,” Pew Research Center, November 11, 2014, accessed October 29, 2015, http://www.pewresearch .org/fact-tank/2014/11/21/executive-actions-on-immigration-have-long-history/. 80.  Eileen Patten, “How Obama’s Executive Action Will Impact Immigrants, By Birth Country,” Pew Research Center, November 21, 2014, accessed October 29, 2015, http://www .pewresearch.org/fact-tank/2014/11/21/how-obamas-executive-action-will-impact -immigrants-by-birth-country/. 81. Julia Preston, “Federal Panel Lets Injunction Against Obama’s Immigration Actions Stand,” International New York Times, May 26, 2015, accessed October, 8, 2015, http://www.nytimes.com/2015/05/27/us/fifth-circuit-court-of-appeals-rules-on-obama -immigration-plan.html. 82.  “Executive Actions on Immigration.”

Chapter Notes   427 83.  Matt Ford, “A Ruling Against the Obama Administration on Immigration,” The Atlantic, November 10, 2015, accessed November 11, 2015. http://www.theatlantic.com /politics/archive/2015/11/fifth-circuit-obama-immigration/415077/. 84.  Rachel L. Swarns, “Long Banned, Mortgage Bias Is Back an Issue,” New York Times, October 31, 2015. 85.  Jens Manuel Krogstad, “Children 12 and Under Are Fastest Growing Group of Unaccompanied Minors at U.S. Border,” Pew Research Center, July 22, 2014, accessed November 1, 2015, http://www.pewresearch.org/fact-tank/2014/07/22/children-12-and -under-are-fastest-growing-group-of-unaccompanied-minors-at-u-s-border/. 86.  Jens Manuel Krogstad, Ana Gonzalez-Barrera, and Mark Hugo Lopez, “At the Border, a Sharp Rise in Unaccompanied Girls Fleeing Honduras, Pew Research Center, July 25, 2014, accessed November 1, 2015, http://www.pewresearch.org/fact-tank/2014/07/25 /at-the-border-a-sharp-rise-in-unaccompanied-girls-fleeing-honduras/. 87. “The Rise of Asian Americans,” Pew Research Center, June 19, 2012, accessed November 1, 2015, http://www.pewsocialtrends.org/2012/06/19/the-rise-of-asian-americans/.

Index

Abbott, Gregg, 389 accommodation, bicultural, 149 acculturation. See assimilation and acculturation Acosta, Dan G., 206 Act of Chapultepec (1946), 213 Adair, Doug, xxi Adams, Frank, 12–13 African Americans: discrimination and segregation, 176, 178, 185, 186, 191, 218–219, 220, 228, 229–230; political influence, 175, 176; post Civil War, 133–134, 153; slavery, 77, 111; violence against, 185, 199, 200, 201; World War II, 156; Zoot suits, 194 Agricultural Labor Relations Act (1975), 256–257 agriculture: in Arizona, 96; bracero program, 214–219, 289, 290–296; brands, livestock, 121–122; cattle industry, 65, 118–122; César Chávez, protests by, 256–257, 269, 278; Chicanas in business, 273–274; in Colorado, 68; cotton, 133–138; coyotes and mansnatchers, 140–142; demographics (2000 and beyond), 319–320; fruit and vegetable farms, 138–140; irrigated farming, 122–126, 138–140; migratory labor, impact of, 146–148, 289; in New Mexico, 46, 51–53, 91; in the New South, 321; patterns of employment, 171–172; post World War II, changes in, 250–251; sheep, raising of, 112–118;

Spanish influence on, 12; sugar-beet production, 142–145. See also tradeunion movement Aguilar, José, 219 Alabama, 318–323 Alamo, 73–74 Alianza Federal de Mercedes, 259–260 Alianza Federal de Pueblos Libres, 259–260 Alpine Club, 195 Alurista, 277 Alvarado, Elvia, 358–359 American Civil Liberties Union (ACLU), 339–340, 378–379 American Civil War, 58, 59, 77–78, 131 American Coordinating Council of Political Education (ACCPE), 255 American Council on Race Relations, 227 American G.I. Forum (AGIF), 253 American Smelting and Refining Company, 157 Amezcua, Consuelo González, 279 amnesty, 307–308 Anaya, Rudolfo, 276 Anaya, Toney, 263 Angel, Frank, 266 Antonio, Don, 112 Anza, Juan Bautista de, 6, 7–8 Anzaldúa, Gloria, 330 Apache Indians, 9–10, 31–32, 44, 58, 110–112, 116 Apodaca, 263 Arbadaos Indians, 3

430  Index Archaeological Institute of America, 233 architecture, Spanish influence on, 49, 201, 240, 245–246 Arias, Ron, 276 Arid Region Doctrine, 124 Arizona: American Coordinating Council of Political Education (ACCPE), 255; Battle of Cananea, 160–161; border disputes, 36–39; Chicano leadership in, 263; Clifton, forty blonde babies, 158–159; cotton industry, 137–138; drop houses, 359–360; economic systems and classes, 59; exploration and settlement of, 6, 10, 27–29; gringo/greaser conflicts, 95–97; Hispanic element of culture, 19–20, 55; immigration data, 128, 314; irrigated farming, 123; labor strikes, 153, 154–157; La Niña de Cabora, 157–158; lynchings in, 96–97; migratory labor, impact of, 146–148; mining industry development, 104–112; pack trains, 131; property rights, 125; S. B. 1070, 359–360, 377–379; sheep industry, 115; as Spanish colony, 57–59; Spanishspeaking population, 27, 32–36; Treaty of Guadalupe Hidalgo (1848), 29–32. See also demographic profile of immigrants; trade-union movement Arizonac Indians, 57–58 Arizona Mining Company, 111–112 Arizona v. Supreme Court (2012), 379 Arkansas, 318–323 Armenta, Catalina, 203–204 Armenta, Esther, 203–204 Armijo, Manuel, 89 Arnaz, Don José, 99 arrastra, 107 Arreguín, Alfredo, 279 arrieros, 130 Arte Público Press, 278 arts and crafts: Chicano culture, 275–279; of New Mexico, 48–50 assimilation and acculturation: bicultural accommodation, 149; colonias and, 172–176; culture conflict and, 168–170;

employment patterns and, 170–172; in Los Angeles, 177–180; movimiento, radical or reformist, 282–285; nativeborn buffer group, 166–168, 170; in northern settlements, 176–177; Southwestern cultural fusion, 235–248; “the Mexican Problem,” 163–165, 220–221, 225–226; University of New Mexico, study of, 230–232. See also culture Austin, Mary, 54 Ayres, E. Duran, 186–187, 189 Baca, Elfego, 91 Baca, Ezequiel de, 94 Baca-Barragan, Polly, 274 Báez, Joan, 278 Baker, Ray Stannard, 140 Bancroft, Hubert H., 232–233 Bandelier, Adolph F. A., 232–233 banditry, 99–100 Bañuelos, Romana Acosta, 274 Barela, Casimiro, 69 barreteros, 109 Barrio, Raymond, 276, 278 Bartlett, John R., 91 Basque immigrants, 23 Basta! La Historia de Nuestra Lucha, xx batea, 106 Battle of Cananea, 159–161 Bell, Horace, 100 Bell, Katherine M., 105 Bell Town Improvement League, 229–230 Bent, Charles, 89 Bentley, Harold W., 237, 239 Bermúdez, Enrique, 160 Berninghaus, Oscar, 233 bicultural accommodation, 149. See also assimilation and acculturation Bilingual Education Act (1968), 281 bilingualism, 280–282 Bilingual Press/Editorial Bilingüe, Arizona State University, 278 Bisbee Mines, 111–112

Index   431 Bishops Committee for the SpanishSpeaking, 269 Bishops Committee on Farm Labor, 269 Blackmar, Frank W., 233 Blanchard, Sarah E., 116 blanket trade, 115 Blumenschein, Ernest, 233 Boddy, Manchester, 201–202 boletas, 111 bonanza, 108 Borajo, Father, 82 Border Industrialization Program, 295–296 borderlands: border of, 36–39; Californios, 63–67; Cortino War, 78–79; cultural fusion in, 235–248; dead-line of sheriffs, 79–81; demographic changes, post World War II, 250–251; feminist perspective of, 330; immigration data (1900–1930), 128; immigration from Mexico, 35–36; Mexican-American War, 74–77; New Mexico, conquest of, 89; Pimeria Alta (Arizona), 57–59; Salt War, 81–82; settlement of, 27–29; Sinarquista movement, 213–214; Spanish colonial plans for, 57; Spanishspeaking population, 27, 32–36; study of history and culture, 232–234; Tejanos (Texas), 59–63; Texas Revolution, 73–74; three waves of immigration, overview of, 287–288; Treaty of Guadalupe Hidalgo (1848) and, 31–32; violence (1908–1925), 82–85; women and border violence, 355–360. See also demographic profile of immigrants; Southwest; individual state names Border Patrol, 37, 297, 303, 305, 326–327. See also U.S. immigration policy Border Protection, Antiterrorism and Illegal Immigration Control Act of 2005 (H.R. 4437), 347–353 Bork, A. W., 25 borrasco, 108 Bowron, Fletcher, 200, 206 boxcar labor camps, 132–133 bracero program, xx, 214–219, 289, 290–296

brands, livestock, 121–122 Braverman, Harry, 184, 189 Breakenridge, W. M., 96 Brennan, William, 367, 369 Brewer, Jan, 377–379 broadcasting, Spanish language use, 279–280 Brown, Jerry, 381 Brown, Kathleen D., 337 Brown Berets, 268–269 Browne, J. Ross, 58, 106–109, 111 Brownell, Herbert, 293 Brown v. Board of Education (1954), 365 buffalo hunting, 70 Burciaga, Cecilia Preciado de, 275 Burlin, Natalie Curtis, 54 Burns, Walter Noble, 97–100 burro transport, 129–131 Bush, George H. W., 265, 335 Bush, George W., 327, 341–342, 376 Cabora, Teresa de, 157–158 Cabrillo, Juan Rodríguez, 5 Calderón, Felipe, 346 California: agriculture, 12, 123, 137–140, 142–145; Basque immigrants, 23; border disputes, 36–39; Californios, 17–20, 63–67, 114; cattle and sheep industry, development of, 114–115, 118–122; César Chávez, protests by, 256–257; colonias, 171, 172–176; Community Service Organizations, 252–253; demographics (2000 and beyond), 318–320; DREAM Act, 381; exploration and settlement of, 5–6, 10, 27–29; fantasy heritage, 17–20; farmworker strikes, xx–xxi, 256–257, 269, 278; gold, discovery and mining, 29, 34, 58, 74–75, 97–100, 103–112; immigration data, 128, 314; invasion by Magonistas, 161–162; lawlessness in, 97–100; lynching of Juanita, 97–100; migratory labor, impact of, 146–148; mining industry development, 103–112;

432  Index property rights, 125; Proposition 187 (1994), 336–342; railroads, 133; Sinarquista movement, 213–214; Spanish influence on, 11–14; Spanishspeaking population, 27, 32–36; Treaty of Guadalupe Hidalgo (1848), 29–32; vermilion cave and New Almaden, 108–109; Zoot Suit riots, 187–191, 195–207, 214. See also demographic profile of immigrants; Los Angeles, California; trade-union movement California Fruit Growers Exchange, 171 Californios, 17–20, 63–67, 114 Calvin, Ross, 48 calzonaires blancos, 105 Camacho, Manuel Avila, 205–206, 218 Campa, Arthur L., 49, 50 Canales, Don Antonio, 74 Canales, J. T., 84 Cananea, Battle of, 159–161 Candelaria, Nash, 276 Cannery and Agricultural Workers Industrial Union, 151 caporal, 114 Carillo, Alejandro, 218 Carr, Harry, 20–21 Carr, Vicki, 278 Carrasco, José, 110–111 Carrillo, Leo, 204 Carson, Kit, 90 Carter, Jimmy, 265, 274, 305 Carter, Lynda Córdoba, 274 Cart War, 77–78 Carvajal, José M., 76 Castillero, Andrés, 108 Castillo, Ana, 330 Castro, Raúl, 263 Castro, Sal, 267 Catholic Church: Bishops Committee for the Spanish-Speaking, 269; Bishops Committee on Farm Labor, 269; Chicano activists and, 269–270; culture conflict and, 94, 169, 322–323; discrimination by, 219, 225; forty blonde babies, 158–159; La Niña

de Cabora and, 157–158; National Catholic Welfare Conference, 225; patron-peon relationship and, 43–44; religion in the New South, 320–323; Saint of Cabora, 157–158; support of Chicano movements, 269–270; support of undocumented immigrants (2006), 348–350, 351, 352 Católicos por la Raza, 269–270, 272 cattle ranching: in Arizona, 95–96; in California, 65–67; in Colorado, 68; development of, 118–122; in New Mexico, 51–53, 90–91; in Texas, 61, 62 Caughey, John W., 64 Cavazos, Lauro, 265 Cedillo, Gil, 381 census data. See U.S. Census Center for Mexican-American Studies, University of Texas, 278 Cervantes, Lorna Dee, 277 Chaboya, Louis, 108 Chavarría, Jesús, 278 Chaves, Francisco, 93–94 Chávez, César, xx, 252, 256–257, 283, 285, 294 Chavez, Dennis, 94 Chávez, Linda, 274, 282 Chavez, Manuel, 116 Chicana Service Action Center, 273 Chicanismo, 282–285 Chicano Associated Students Organization, 266–268 Chicano leadership, post World War II: American Coordinating Council of Political Education (ACCPE), 255; American G. I. Forum, 253; Brown Berets, 268–269; Chicana movements and organizations, 271–275; Community Service Organizations, 252–253; late 1970s–1980s, 263–271; Mexican-American Political Association (MAPA), 254–255; movimiento, radical or reformist, 282–285; national organizations, 270–271; organizational development, 252; overview, 249–250;

Index   433 Political Association of SpanishSpeaking Organizations (PASSO), 255; protest movements, 255–262; student movement, 265–268 Chicano National Welfare Rights Organization, 272 Chicano Studies Center Publications at the University of California, Los Angeles, 278 Chicano Studies programs, origins of, 268 Chicano Youth Association (CYA), 266–268 Chicano Youth Conference, 272 Chicano Youth Liberation Conference, 258 Chichorana, Francisco, 2 children, unaccompanied immigrants, 327 chispa, 103–106 cholos, 98 churros, 113 cigar industry, 24 Cinco de Mayo, 17–20 Cisneros, Henry, 263 citizenship: advocacy for, 175; children of undocumented parents, 342–344, 380; criteria for (2000), 341; Deferred Action for Childhood Arrivals (DACA), 380–387; dual nationality, 345–347; English-proficiency test, 282; political participation and, 264; secondgeneration immigrants, 363–364; Treaty of Guadalupe Hidalgo (1848), 29–32 Ciudad Juárez, violence against women, 356–360 Civil Rights Act, 281 Civil War, American, 58, 59, 77–78, 131 Clanton gang, 95–96 class system: in California, 63–67; in New Mexico, 44–47; in Texas, 62–63 Clements, George P., 149–150, 151, 152 Cleveland, Grover, 52 Clifton Mine, 112 climate, 11–14 Clinton, Hillary, 372–377 Clinton, William (Bill) J., 336, 338, 341 Club Liberal de San Antonio, 271

Coalition for Immigrant Reform, 341 Cobo, Bernabe, 9 Cochise, 58 Coello, Renán Almendárez, 351 Cold War policies, 214, 333, 334 Colegio César Chávez, 268 colonias, 171, 172–176 colonization, 1–14 Colorado: community service clubs, 226–227; exploration and settlement of, 27–29; labor strikes, 153; lost provinces of, 67–70; railroads, 133; Spanishspeaking population, 27, 32–36; sugarbeet production, 142–145. See also trade-union movement Colorado La Raza Unida party (LRU), 258 Comanche Indians, 10, 31–32, 44, 70 Comité Mexicano Contra el Racismo, 213 Committee on Opportunities for Spanish Speaking People, 265 Common Ground, 220 Communism, 214, 333, 334 communities of Mexican-Americans: colonias, 172–176; in Los Angeles, 177–180; northern settlements, 176–177 Community of Action Patrol, 269 Community Organization for Public Service (COPS), 269 Community Service Clubs, Inc., 226–227 Community Service Organizations (CSO), xx, 252–253 Comstock, 107–108 Confederación de Uniones Obreras Mexicanas (CUOM), 150 Congressional Medal of Honor, 209–211 Congress of Mexican American Unity, 269 contract-labor law, 140–143. See also migratory labor copper mines, 110–112, 154–157 Corona, Bert, 252, 254 Coronado, Francisco Vásquez de, 4–5 Cortez, 9 Cortina, Juan Nepomuceno, 78–79 Cortino War, 78–79 cotton industry, 133–138, 148

434  Index Council of Civic Unity, 204 Council of Inter-American Affairs, 223–228 cowboys, 118, 120–121 coyotes, 44, 140–142, 290, 302, 360 Crawford, Remsen, 134 crops, Spanish influence on, 12, 122–126, 138–140. See also agriculture Crusade for Justice, 258, 272 Cuban immigrants, 23–24 culture: arts and literature, 275–279; birth of stereotype, 100–101; of Californios, 63–67; conflict of, 168–170; future trends, 390–395; of lost provinces (Colorado and Texas), 67–70; movimiento, radical or reformist, 282–285; of New Mexico, 41–48; of Pimeria Alta (Arizona), 57–59; secondgeneration immigrants, 361–366; Southwest cultural fusion, 235–248; study of Southwest, 232–234; of Tejanos (Texas), 59–63; Texas, early conflicts in, 71–74. See also assimilation and acculturation; demographic profile of immigrants Curtis, F. S. Jr., 9 Cutting, Bronson, 94 Daly, H. W., 130, 131 Dana, Richard Henry, 100 Dannemeyer, William, 338 Davilla, José, 211–212 Davis, Edward Everett, 134–135 dead-line of sheriffs, 79–81 Deferred Action for Childhood Arrivals (DACA), 380–387 Deganawidah-Quetzalcoatl University (DQU), 268 Delano Grape Strike, xx–xxi, 256, 269, 278 Delaware, 318–320 Delgado, Abelardo, 277 Delgado, Manuel, 183 Delgado v. Bastrop Independent School District, 265

Del Rio Manifesto, 267 demographic profile of immigrants: education, occupation, and poverty, 314–318; emergent immigrant communities, 318–320; future trends, 325–328, 390–395; overview, 311–314; presidential race (2008), 372–373; religion in the New South, 320–323; second-generation immigrants, 361–364; unauthorized immigrants, 324–325 Denman, William, 229 Department of Homeland Security, 342 deportations: Border Protection, Antiterrorism and Illegal Immigration Control Act of 2005 (H.R. 4437), 347– 353; Deferred Action for Childhood Arrivals (DACA), 380–387; future trends, 327; Immigration Act (1990) and, 335–336; labor strikes and, 152, 153, 155–156, 157; parents of U.S. born children, 342–344; presidential politics and, 380 De Quille, Dan, 107–108 Development, Relief and Education for Alien Minors Act (DREAM, 2001), 368–372, 381–387, 390–391 Diaz, Ernesto Felix, 214 Díaz, José, 182–183 Dillingham, William P., 296, 297 Dillingham Commission, 296 Dingley Tariff, 142 discrimination: in Arizona, 59; bilingualism, 280–282; Captain Ayres report, 186–187; Chicana liberation, 272–275; in education, 265–268; movimiento, success of, 283–285; national organizations to protect legal rights, 270–271; in New Mexico, 54–55; segregation practices, 153–154; U.S. history of, 330–331; Westminster, California schools, 227–230; World War II, effects of, 250; during World War II, 212–214, 217–221 Displaced Persons Act (1948), 298

Index   435 Dixon, Compton, 198 Dobie, J. Frank, 80–81, 118, 120, 139, 232, 241 Doctrine of Appropriation, 124 don system, 92–95 Douglas, Walter, 156 Drake, Jim, 256 drapes, 194 DREAM Act (2001), 368–372, 381–387, 390–391 drop houses, 359–360 dual nationality, 345–347 Durbin, Dick, 368, 371 economic systems and classes: in Arizona, 59; in California, 63–67; cattle industry, development of, 118–122; cotton industry, 133–138; don system, 92–95; of lost provinces, Colorado and Texas, 68–70; maquiladoras, 331–332; migratory labor, impact of, 146–148; mining industry, 103–112; in New Mexico, 46–47, 50–55; sheep, raising of, 112–118; stereotypes, birth of, 100–101; in Texas, 60–63; U.S. and Mexican policies and, 333 education: advancements in Spanishspeaking communities, 231–232; bilingualism, 280–282; Chicano student movement, 265–268; in colonias, 174–175; Deferred Action for Childhood Arrivals (DACA), 386–387; demographics (2000 and beyond), 314–318; DREAM Act (2001), 368–372; future trends, 325–326, 390–395; language issues, 243–245; levels of, xxii–xxiii, 174–175, 264; movimiento, success of, 283–285; Proposition 187 (1994), California, 337–340; scholarly journals, 277–278; second-generation immigrants, 365–369; segregation, Westminster case, 227–230; University of New Mexico, 230–232; World War II, effects of, 250 El Enganchado, 179–180

Elizalde, Daniel S., 211 Elizalde Anti-Discrimination Committee, 211 Elizondo, Sergio, 276 Ellison, Edith Nicholl, 123 El Paso incident, 290–292 El Renagado, 166–167 Emigrant Agent Law (1929), 143–144, 146 employment: bracero program, 289, 290–296; Chicanas in business and government, 273–275; commuters, 301–302; demographics (2000 and beyond), 314–318; green carders, 300–301; maquiladoras, 296, 331–332, 356–360; patterns, assimilation and, 170–172, 175, 176–177; rates and occupations (1967), xxii–xxiii; Simpson-Rodino Immigration Reform and Control Act (1986), 304–309; undocumented immigrants, 289, 302–304; World War II, braceros, 214–219; World War II, effects of, 250–251 English Only, 281–282, 336–342 Equal Employment Opportunity Commission, xxii Escalante, Alicia, 272 escalera, 109 escoria, 108 Esparza, Moctezuma, 276 Espiñosa, Aurelio M., 49, 237, 242 Estevánico, 3–4 explorers, Spanish, 1–8 Fair Share Law (1960), 298 fantasy heritage: Californios, 16–20; Spanish and Mexican-Indian heritage, 20–22; Spanish immigration, 22–25 Farm Labor Contractor’s Registration Act, 305 Farm Security Administration, 215 farmworkers: César Chávez, protests by, 256–257, 269, 278; strikes by, xx–xxi; World War II, braceros, 214–219, 289, 290–296. See also agriculture; labor

436  Index Farm Workers Press, xx Federal Alliance of Free Towns, 259–260 Federal Alliance of Land Grants, 259–260 Federation for American Immigration Reform (FAIR), 302 feminism: feminist movement, 272–275; perspective of borderlands, 330; women and border violence, 355–360; women and transnationalism, 343–355 Fender, Freddy, 278 Fergusson, Erna, 53, 54, 73 Fergusson, Harvey, 51 Field, Stephen J., 97 First Liberal Congress, 159–160 Fisher, Walter M., 98 Flores, Francisca, 272 Florida, 23–24, 318–320 folk traditions: Californios, 64; culture conflict, 168–170; of New Mexico, 48–50; San Luis Valley, 68–69; sheepherding songs, 116 forced repatriations, 152, 153, 155–156, 157 Ford, Gerald, 265, 305 Ford Foundation, 270 Forgotten People, 231 forty blonde babies, 158–159 Fox Quesada, Vicente, 346 Fraijo, Manuel, 156–157 Frankfurter, Felix, 155–156 French, William, 91 fruit farms, 12, 138–140, 146–148 Gaarder, Alfred Bruce, 237–238 Gadsden, James, 36 Gadsden Purchase, 104 Galarza, Ernesto, 215, 217, 294, 330 Gallegly, Elton, 367 Gallegos, Father, 90 gambussinos, 104–105 Gamio, Manuel, 166, 168, 214–215, 237 gangs: pachuquismo, 191–194; Sleepy Lagoon case, 182–186; Zoot Suit riots, 187–191, 195–207, 214

García, Alma M., 311, 329, 361 García, Héctor Pérez, 252, 253, 255 García, Macario, 210, 218 Garner, John Nance, 140 Garrison, George P., 60 Garza, Albert, 80 Garza, Catarina, 81 gente de razón, 63–67, 98 Georgia, 318–323 Gerra, Don Manuel, 61 Gill, Irving, 49 Glavecke, Adolph, 79 gold: discovery of, 29, 34, 58, 74–75; lynching of Juanita, 97–100; mining industry development, 103–112 Goldman, Emma, 161 Gómez, Lauriano, 130–131 Gominguez, Adolofo G., 218 Gonzales, Helen, 272 Gonzales, Jovita, 83 Gonzales, Manuel Pedro, 24 Gonzales, Rodolfo “Corky,” 252, 257–258, 267, 275, 285 González, Henry B., 283 González, Raúl, 263 Good Neighbor Commission, 218, 219, 232 Good Neighbor Policy, 212–213, 218, 223–228, 246 grain farms, Spanish influence on, 12 grape pickers strike (1965), xx–xxi, 256, 269, 278 grassroots organizations, 226–227. See also Chicano leadership, post World War II greasers, 87–88, 100–101 Great Depression, 53–54, 146, 233 Great Recession (2007–2009), 324–325 green carders, 300–301 Gregg, Josiah, 44, 100 gringo, 87–88, 100–101 Gruening, Ernest, 8, 83 Guaymas, 36–37 guest workers, 295. See also U.S. immigration policy Guinn, J. M., 99

Index   437 Gutiérrez, José Angel, 252, 258, 260–262, 267, 285 Gutiérrez, Juan José, 339 H-2 guest-workers, 295 hacienda, 10, 169 Handlin, Oscar, 311–312 Harby, Lee C., 61 Harding, Warren G., 297 Harrington, Michael, 316 Hart-Cellars Immigration and Nationality Act (1965), 298–299, 307 Harwood Foundation, 231 Hatch, Orrin, 368, 371 Hayes, Rutherford, 80 health insurance, rates of, 317 Heintzelman, Major, 79 Heintzelman Mine, 111–112 Hernández, Antonia, 341 Hernández, María L., 271, 272 Hernández, Pedro, 271 Herraras, Sylvester, 210 Hicks, Joe, 338 Hick’s Camp, 173 hijos del pais, 65 Hinojosa-Smith, Rolando, 276 Hispanic, New Mexico culture and, 55 Hopi Indians, 30–32 horses: cowboys and, 120–121; value of, 9–10 housing: colonias, 172–176; in Los Angeles, 177–180; in northern settlements, 176–177 Hubbel, Santiago, 130 Huerta, Dolores, 256, 272 Hull, Cordell, 206 Humphrey, Norman, 169–170 Hynes, William (Red), 151 “I Am Joaquin,” 258, 275, 277 Idaho, 133, 153, 318–320. See also tradeunion movement Illegal Immigration Reform and Immigrant Responsibility Act (1996), 367 Illinois, 145–146, 227, 314, 318–320

immigrants, demographic profile of: dual nationality and voting in Mexican elections, 345–347; education, occupation, and poverty, 314–318; emergent immigrant communities, 318–320; future trends, 325–328, 390– 395; overview, 311–314; religious space in the New South, 320–323; secondgeneration immigrants, 361–366; unauthorized immigrants, 324–325 immigration: amnesty, 307–308; Arizona S. B. 1070, 377–379; bicultural accommodation and, 149; boxcar labor camps, 132–133; in California, 66; commuters, 301–302; coyotes and mansnatchers, 140–142; culture conflict, 168–170; data on (1900–1930), 33, 128; employment patterns, 170–172, 217; forced repatriations, 152, 153, 155–156, 157; green carders, 300–301; industrial laborers, 145–146; to Los Angeles, 177– 180; from Mexico, 35–36; migratory labor, impact of, 146–148; movement and population at borderlands, 37; native-born buffer group, 166–168, 170; northern settlements and, 176–177; permanent visas, 300; quota system, 297; refugee legislation, 297–298; from Spain, Cuba, Puerto Rico, 23–24; sugarbeet production and, 142–145; “the Mexican Problem,” 163–165, 220–221, 225–226; three waves of, overview, 287–288; undocumented immigrants, 286, 289, 302–304, 342–345; women and transnationalism, 343–355; World War II, effects of, 250. See also U.S. immigration policy Immigration Act (1921), 297 Immigration Act (1924), 141, 142–143, 163 Immigration Act (1965), 335–336 Immigration Act (1986), 335–336 Immigration Act (1990), 335–336 Immigration and Nationality Efficiency Act (1978), 305

438  Index Immigration and Naturalization Service (I&NS), 303, 307–308, 326–327. See also U.S. immigration policy Immigration Reform and Control Act (IRCA, 1986), 331, 332–336 Indiana, 145–146, 318–320 Indians: Chicano culture, influence on, 275; cultural fusion of Southwest and, 245– 247; Hispanic element and, 55; irrigated farming practices, 123; Mexican-Indian heritage, 20–22, 25; New Mexico, culture of, 44–48; New Mexico isolation, effect of, 44; role in “the Mexican Problem,” 165; Spanish exploration and settlements, 1–11; Treaty of Guadalupe Hidalgo (1848) and, 29–32, 76. See also specific tribe names industrial labor, 145–148 Industrial Workers of the World (IWW), 271 Inscription Rock, 7 Institute of International Education, 225 Inter-Agency Committee on MexicanAmerican Affairs, 264–265 International Smelting and Refining Company, 156 International Union of Mine, Mill and Smelter Workers, CIO, 156, 157 irrigated farming, 122–126, 138–140 Itliong, Harry, 256 Jamieson, Stuart, 153 Jaramillo, Marí-Luci, 274 Jennings, N. A., 80 Jessup, Roger, 200 Jester, Beauford H., 291 Jiménez, Francisco, 276, 330 Jiménez, Francisco de P., 219 Jiménez, Luis, 279 John, Chileno, 130 Johnson, Lyndon B., xxii, 253, 264, 270, 279, 298, 316 jornaleros, 62 journals, expansion of, 277–278

Juanita, lynching of, 97–100 Juárez, Benito, 159 Juárez-Lincoln University, 268 Julian, George W., 52 Justa Publications, 278 Justicia para Nuestras Hijas (Justice for Our Daughters), 359 Keglar Hill, 134–135 Kendall, George, 89 Kennedy, Edward, 305, 335 Kennedy, John F., ix–xx, 298 Kennedy, Mifflin, 76 Kenny, Robert W., 204, 229 Kentucky, 318–323 Kibbe, Pauline, 136 King, Richard, 76 Kino, Father Eusebio Francisco, 6 Kupper, Winifred, 113 labor: bracero program, xx, 214–219, 289, 290–296; Chicana organizing leadership, 271; Chicanas in business and government, 273–275; commuters, 301–302; demographics (2000 and beyond), 314–318; green carders, 300–301; maquiladoras, 296, 331–332, 356–360; patterns of employment, 170–172, 175, 176–177; SimpsonRodino Immigration Reform and Control Act (1986), 304–309; undocumented immigrants, 289, 302–304; World War II, effects of, 250–251. See also trade-union movement labor agents, 141–142 labor importation, xx La Causa Publications, Inc., 278 La Comisión Femenil Nacional, 273 LaFollette Committee, 151 la Huelga, 256 Lamy, Bishop, 90 land ownership: Alianza Federal de Mercedes, 259–260; in California,

Index   439 65–66; in New Mexico, 51–53, 91; system for, 125–126; in Texas, 62 language: bilingualism, 280–282; demographics (2000 and beyond), 315–316; English only, 281–282, 336–342; media and broadcasting, 1960s–1980s, 279–280; patterns in Southwest, 236–245; second-generation immigrants, 361–366 La Niña de Cabora, 157–158 La Orden Caballeros de América, 271 La Orden Hijos de América, 271 La Raza Unida, 272 l’Archevéque, Sostenes, 91–92 Larrazolo, Octaviano A., 93–94 Lathers, Swift, 212 Latin-American, use of term, 54 Latino Policy Coalition Survey, 376 Latinos Unidos por la Justicia, 269 Laughlin, Ruth, 54, 115, 245–248 Lau v. Nichols, 280–281 law of mines, 109–110 League of United Latin-American Citizens (LULAC), 62–63, 271, 339–340 Leon, José de, 130 León, Ponce de, 2 Lesinsky, Henry, 112 Lewis, Tracy Hammond, 82, 83–84 Leyvas, Henry, 182–186 Liga Mexicanista Femenil, 271 Liga Obrera de Habla Española, 154 literature, 275–279 Little, Wilson, 231–232 Llanes, Balton, 211 Lloyd, Lewis, 75 Lockwood, Frank C., 11 Long, Haniel, 45, 47, 48 Longoria, Félix, 253 López, Francisco, 104 López, Ignacio, 227 López, José Mendoza, 211 López, Olibama, 68–69 López, Trini, 278 Los Angeles, California: Californios and fantasy heritage, 17–20; Captain Ayres

report, 186–187, 189; Community Service Organizations, 252–253; immigrant population data (2011), 314; pachuquismo, 191–194; settlement history, 177–180; Sleepy Lagoon case, 182–186; Zoot Suit riots, 187–191, 195–207, 214. See also California Los Lobos, 278 Los Voluntarios, 258 Louisiana, 318–323 Louisiana Purchase, 7 Lowrie, Samuel Harman, 72 Lubin, Simon J., 150 Lucy, Robert E., 269 Luján, Manuel, 265 LULAC (League of Latin-American Citizens), 62–63, 271, 339–340 Lummis, Charles Fletcher, 129, 130, 233 lynching of Juanita, 97–100 magazines, expansion of, 277–278 Magnon, Enrique, 159–160, 161 Magnon, Ricardo Flores, 159–160, 161 Magoffin, James W., 89 Mahony, Roger, 349, 350 Malagrán, Isabel, 271 Mandujano, Telesforo, 144 man-snatchers, 140–142 maquiladoras, 296, 331–332, 356–360 Marcos, Fray, 3–4 Marín, Richard “Cheech,” 276 Marshall, James W., 104 Martí, José, 24 Martinez, Joe, 209–210 Martínez, José Antonio, 89–90 Martínez, Vilma, 270, 275 Marvin, George, 83 Massachusetts, 318–320 Matuz, John, 184 mayordomo, 123–124 McCain, John, 376–377 McCarran, Patrick, 298 McCarran-Walter Immigration Act (1952), 292, 295, 298, 307

440  Index McClintock, James H., 107 McCormick, Cyrus, 231 McCormick, Paul J., 228–229 McGovern, George, 294 McGroarty, John Steven, 19 Meade, George C., 75 Medellin, Roberto, 218 media, Spanish language use, 279–280 Medina, Bartolomé de, 108 Meier, Matt S., 249, 263 Melendes, Edward, 212 Melendez, Jack, 184 Menchú, Rigoberta, 358 Méndez, Gonzalo, 227–230 Méndez, Miguel, 276 Mendez v. Westminster School District (1945), 227–230, 265 merinos, 113 Mesilla Riots, 93 Mexican, use of term, 54 Mexican-American Business and Professional Women, 273 Mexican-American Legal Defense and Education Fund (MALDEF), 270, 339–340, 367 Mexican-American Political Association (MAPA), 254–255 Mexican-American Student Association (MASA), 266–268 Mexican-American Student Confederation (MASC), 266–268 Mexican-American Studies and Research Center, University of Arizona, 278 Mexican-American War, 74–77 Mexican-American Women’s National Association (MANA), 273 Mexican-American Women’s Political Caucus, 273 Mexican-American Youth Organization (MAYO), 260–262, 266–268 Mexican as census category, 32 Mexican banditry, 99–100 Mexican Civic Committee, 227 Mexican-Indian heritage, 20–22, 25 Mexican Liberal Party, 160

Mexican National Commission on Human Rights, 358 Mexican Revolution, 82–85 Mexico: antidiscrimination meetings (1944–1945), 213–214; austerity programs (1980s and 1990s), 313; border with United States, 36–39; bracero program, 289, 290–296; cultural fusion of Southwest and, 245–247; dual nationality and voting in elections, 345–347; economic challenges (1950–1990s), 333; historical influence on Southwest, 1; labor movement in, 159–162; maquiladoras, 296, 331–332, 356–360; oil industry, 333–334; Sinarquista movement, 213–214; Treaty of Guadalupe Hidalgo (1848), 29–32; women and border violence, 356–360; World War II, braceros, 214–219 Michigan, 142–146, 153. See also tradeunion movement Midwest, 142–145, 176–177, 318–320 Migration Policy Institute, 383, 385 migratory labor: César Chávez, protests by, 256–257; cotton industry, 133–138; coyotes and man-snatchers, 140–142; economic impact of, 146–148; fruit and vegetable farms, 138–140; in Midwestern industries, 145–146; sugar-beet production, 142–145; undocumented immigrants, 289; World War II, braceros, 214–219, 289, 290–296. See also demographic profile of immigrants; labor Migratory Labor Agreement (1951), 292 military service: Iraq, 375–376; World War II, 209–221, 250, 253 mines and mining: in Arizona, 57–58; Battle of Cananea, 159–161; in bordertowns, 38; California gold rush, 58, 74–75, 97–100, 103–112; industry development, 103–112; labor strikes, 154–157; laws and regulations, 109–110; lynching of Juanita, 97–100; pack trains, 130–131; Salt War, 81–82; vermilion cave, 108–109

Index   441 Minnesota, 143–145 Mireles, Jovita Gonzales de, 20 Mission Play, 19 missions, Spanish: in Arizona, 57–59; in California, 10, 12–13, 34, 63–67, 119, 126; fantasy heritage, 15, 19, 20, 21; lasting influence of, 11–14, 49; in New Mexico, 6, 10, 47, 94–95; in Texas, 59–60; as tourist promotion, 21 Mississippi, 318–323 Molina, Gloria, 275 Molina, María Elba, 274 Montana, 133 Montenegro, Eugene Chavez Jr., 203 Monteros, Antonio Espiñosa de los, 221 Montgomery, Robert H., 134 Montoya, José, 277 Montoya, Pablo, 90 Mora, Pat, 364–365 Moreno, Dalia, 346 Morrison, Bruce, 335 Morrison visas, 335–336 Morton, Carlos, 276 Mount, Julia, 272 Mounted Watchmen, 37 movimiento, radical or reformist, 282–285. See also Chicano leadership, post World War II Movimiento Estudiantil Chicano de Aztlán (MEChA), 267 Mowry, Sylvester, 58, 154 Mujeres Activas en Letras y Cambio Social (MALCS), 273 mule transport, 129–131 mural art, 279 Murieta, Joaquin, 97 Murray, Churchill, 206 Murrow, Edward R., 316 music, Chicano culture, 278–279 Musquiz, Virginia, 272 Narvaez, Panfilo de, 3 National Alliance for Human Rights, 348

National Association of Latino Elected and Appointed Officials (NALEO), 265 National Capitol Immigration Committee, 350 National Catholic Welfare Conference, 225 National Chicano Moratorium Committee, 269 National Chicano Political Conference, 273 National Chicano Research Network, 277 National Coalition for Latino Clergy and Christian Leaders, 379 National Council of La Raza, 270–271 National Farm Workers Association (NFWA), 256–257 National Hispanic University, 268 National Immigration Law Center, 371, 379 National Observatory Against Femicide, 357 National Organization of MexicanAmerican Services (NOMAS), 270 National Origins Act (1924), 297 National War Labor Board, 156 Native American Party, 75 native-born minority, 166–168, 170 Navajo Indians, 31–32, 44 Navarro, Ana, 384 Nebraska, 142–145 Negroes. See African Americans Nelson, Eugene, xxi Nevada, 318–320 New Almaden Mine, 108–109 New Deal, 233 Newmark, Harris, 87 New Mexico: Alianza Federal de Mercedes, 259–260; border disputes, 36–39; Chicano leadership in, 263; conquest of, 88–89; don system, 92–95; economic classes, 50–55; economic system, 46–47, 92–95; exploration and settlement of, 5–6, 10, 27–29, 42; folk traditions, 48–50; Hispanic element of culture, 55; immigration data (1900–1930), 128; irrigated farming, 123–124; isolation, effect on culture, 41–44; labor strikes, 154; land ownership, 51–53; Martínez,

442  Index José Antonio, 89–90; political system, 45–46; population, characteristics of, 44–48; property rights, 125; School of Inter-American Affairs, 230–232; sheep, raising of, 114–118; Socorro, conflict in, 90–92; Spanish-speaking population, 27, 32–36; Treaty of Guadalupe Hidalgo (1848), 29–32. See also demographic profile of immigrants; trade-union movement newspapers, Spanish language use, 279–280 New York, 318–320 Nixon, Richard, 265, 274 North Carolina, 318–320, 320–323 North Dakota, 143–145 North Town, 173 Noyola, Ricardo, 210 Obama, Barack, 327, 340–341, 348, 372–377, 380, 381–382, 383–390 O’Brien, Robert O., 177 Ochoa, Estevan, 131 Odem, Mary, 322 Ohio, 143–146, 176–177 oil industry, Mexico, 333–334 Olivares, Trueba, 213 Olivárez, Graciela, 275 Olmstead, F. L., 84–85, 100 Oñate, Juan de, 5–6, 7, 44–45 O’Neil, Owen, 66 Operation Jobs, 303 Operation Wetback, 293–296, 302 Ord, General, 80 Oregon, 133 organized labor. See trade-union movement Orozco, José Clemente, 279 Ortega, Joaquin, 24, 41, 230–232, 243, 245 Ortega, Katherine, 274 Otero, Manuel, 91 Oviedo, 3 pachuquismo, 191–194 pack-trains, 129–131

Padilla, Angel, 183 Padilla, Fray Juan de, 5 PADRES (Padres Asociados Para Derechos Religiosos, Educativos, y Sociales), 270 Padua Institute, 19 Pajarito Publications, 278 Pallares, Jesús, 154 Pan-American Union, 215 Papago Indians, 58 Paredes, Ignacio, 107 Parsons, Lucia González, 271 partido system, 117 pastores, 65, 114, 115–118 patio, 108 patrón, 114, 117 Paul, Rodman W., 104, 105–106 Paz, Frank M., 227 Pearce, T. M., 240 Peña, Federico, 263 peon-patron relationships: don system, 92–95; in New Mexico, 43–46; in sheephearding, 117–118; in Texas, 60–61, 77–78 Perea, Francisco, 90–91 Perez, Manuel, 210 Perry, George Sessions, 121 Personal Responsibility and Work Opportunity Reconciliation Act (1996), 367 Peterson, Helen, 226 Pew Research Hispanic Center, 373, 379–380 Pfaelzer, Mariana, 340 Phelps-Dodge Company, 156 Pijoan, Jane W., 224–225 Pima Indians, 30–32, 58 Pimeria Alta, 58 Pino, Don Pedro, 46 placer, 108 Plyler v. Doe (1982), 340, 365, 367 pobres, 45–47, 59, 60–61, 88 Point 13, 213 Political Association of Spanish-Speaking Organizations (PASSO), 255

Index   443 political influence: Alianza Federal de Mercedes, 259–260; American Coordinating Council of Political Education (ACCPE), 255; American G. I. Forum, 253; Arizona S. B. 1070, 377–379; Border Protection, Antiterrorism and Illegal Immigration Control Act of 2005 (H.R. 4437), 348–353; Brown Berets, 268–269; in California, 66; Chicana movements and organizations, 271–275; Chicanas in government, 273–275; Chicano student movement, 265–268; Colorado La Raza Unida party, 258; community service clubs, 226–227; Community Service Organizations and, 252–253; Deferred Action for Childhood Arrivals (DACA), 380–387; demographics and, xxi–xxii, 316; don system, 92–95; DREAM Act, youth movement for, 369–372; dual nationality and, 345–347; future trends, 326; in late 1970s–1980s, 263–271; Mexican-American Political Association (MAPA), 254–255; Mexican-American Youth Organization (MAYO), 260–262; movimiento, radical or reformist, 282–285; national organizations, 270–271; offices holders, listing of (1967), xxii; Political Association of SpanishSpeaking Organizations (PASSO), 255; presidential race (2008), 372–377; presidential race (2012 and beyond), 379–387; Proposition 187 (1994) and, 340–341 Pooler, James S., 212 Poor People’s March to Washington, D. C., 258, 259 Portillo, Estela, 276 Portillo, José López, 333 Portillo, Lourdes, 359 poverty, demographics (2000 and beyond), 314–318, 325–326 presidential race: (2008), 372–377; (2012 and beyond), 379–387

President’s Fair Employment Practice Commission, 156 presidio, 10 Price, Colonel, 90 print media, Spanish language use, 279–280 property rights: Alianza Federal de Mercedes, 259–260; in California, 65–66; in New Mexico, 51–53, 91; system for, 125–126; in Texas, 62 Proposition 187 (1994), California, 336–342 protest movements, 255–262 Pryce, Rhys, 161 Public Law 78 (PL78), 292 publishing companies, 278 Pueblo Indians, 4–6, 9–11, 30–32, 44–48, 124 pueblo rights, water use, 124–125 Puerto Rican immigrants, 23 quartz mines, 106–107 Queer Undocumented Immigrant Project, 371 Quevedo, Eduardo, 254 quicksilver, 108–109 Quiñones, Juan Gómez, 277 Quinto Sol organization, xxiii Quinto Sol Publications, 278 Quivira, 4 race, Mexican as census category, 32. See also discrimination; segregation racial profiling, Arizona S. B. 1070, 377–379 radio stations, 279–280 railroads: Cart War (Texas), 77–78; court ordered juvenile labor for, 202–203; development of, 129–133; effect on culture, 242; impact of, 29; in New Mexico, 90–91; World War II, braceros, 216–217 Ramírez, Belinda Cárdenas, 274 Ramirez, Catarino, 145

444  Index Rangel, Antonio, 219 Raza Unida party, 261–262, 286 Reagan, Ronald, 265, 274, 281, 306, 308, 332–333, 334 Rechy, John, 276 Reclamation Act (1902), 52 Redfield, Robert, 168–169 Red Robber of the Rio Grande, 78–79 Reed, Joseph, 188 refugee legislation, 297–299 Refugee Relief Act (1953), 298 Registry Act (1929), 307 Reid, Hugo, 105 religion: culture conflict and, 169; discrimination and, 219, 220; folk life, San Luis Valley, 70; Martínez, José Antonio, 89–90; in New Mexico, 43–44, 94–95; in the New South, 320–323; Saint of Cabora, 157–158; Spanish influence, 11–14. See also Catholic Church; missions, Spanish repatriations, forced, 152, 153, 155–156 Republic of the Rio Grande, 74 Revolt of the Yaquis, 158 Reyes, Manuel, 183 Richardson, Bill, 373 ricos, 45–47; birth of stereotype, 100–101; don system, 92–95; in New Mexico, 88; Tejanos, 74; in Texas, 61, 63 Rio Grande River, 37–38, 60–63 Rippy, J. Fred, 38 Rivera, Diego, 279 Rivera, Tomás, 276 Robert, Edward, 67 Robinson, Will H., 111 Rockefeller, Nelson, 223 rodeos, 122 Rodino, Pete, 304–309 Romano-V, Octavio, 277 Romero, Casimero, 70 Romney, Mitt, 380, 383–385 Ronstadt, Linda, 278 Roosevelt, Eleanor, 205 Roosevelt, Franklin D., 205–206, 223 Ross, Fred Sr., 256

Ross, Fred W., 174, 175, 227, 230, 252–253 Rosskelly, R. W., 153 Roybal, Edward R., ix, 252–253, 254 Royce, Josiah, 98, 235 Rubio, Marco, 384 Ruiz, Chepe, 183 Russell, John, 42, 117 Ruxton, George F., 90 Saint of Cabora, 157–158 Salazar, Rubén, 269 Salinas, Omar, 277 Salinas, Porfirio, 279 Salinas de Gortari, Carlos, 334 Salt War, 81–82 Sánchez, David, 269 Sanchez, George, ix, 32, 41, 192, 220 Sánchez, Rosaura, 276 San Fernando Valley Council on Race Relations, 204 Santa Rita Mine, 111–112 Sante Fe Ring, 93 Saposs, David, 224 Schecter, Hope Mendoza, 274 Schreiter, Oskar Hellmuth, 213 Schwartz, Harry, 139 Schwarzenegger, Arnold, 381 Scott, Winfield, 75 second-generation immigrants, 361–366, 380, 390–395 segregation, 153–154; Captain Ayres report, 186–187; colonias, 172–176; in Los Angeles, 178–180; movimiento, success of, 283–285; northern settlements, 176–177; patterns of, 171; of schools, 265–268; U.S. history of, 331; Westminster, California schools, 227–230; World War II, braceros and, 217–221; World War II, effects of, 250 Seibold, Doris K., 241 Select Commission on Immigration and Refugee Policy, 305 Serra, Fray Junípero, 5 service clubs, community, 226–227

Index   445 seven cities of gold, 4–6 Shafer, Emmy, 281 sheep, raising of, 112–118 Shell Oil Company, 156 Shinn, Charles Howard, 122 Shumer, Charles, 306 Silex, Humberto, 157 silver mining, 103–112 Simpson, Alan, 299, 304–309 Simpson, J. H., 116 Simpson-Mazzoli Immigration Reform and Control Act (1982), 305–306 Simpson-Rodino Immigration Reform and Control Act (1986), 270, 304–309 Sinarquista movement, 213–214 Siqueiros, David, 279 slavery: in California, 64; in New Mexico, 45, 46; in Texas, 77–78 Slayden, James L., 142 Sleepy Lagoon, 182–186 Smith, Sarah Bixby, 117 smugglers, 359–360 socioeconomic status: data for 1980s, 264; demographics (2000 and beyond), 314–318; future trends, 325–326 Solís, Manuela, 271 Sotelo, Eduardo, 351 Soto, Gary, 277 South Carolina, 318–320, 320–323 South Dakota, 318–320 Southwest: cattle industry, 118–122; community service clubs, 226–227; cultural fusion in, 235–248; demographic changes, post World War II, 250–251; discrimination, World War II, 217–221; feminist perspective of, 330; immigration data (1900–1930), 128; irrigated farming, 122–126, 138–140; migratory labor, impact of, 146–148; mining, development of, 103–112; railroads, building of, 129–133; sheep, raising of, 112–118; Sinarquista movement, 213–214; study of history and culture, 232–234; three waves of immigration, overview of, 287–288;

women and border violence, 355–360. See also borderlands; demographic profile of immigrants; individual state names Southwest Council of La Raza, 270–271 Southwest Voter Registration Education Project (SVREP), 270–271 Spain and Spanish influence: architecture, 49, 201, 240, 245–246; Arizona colony, 57–59; California colony, 63–67; cattle, introduction to U.S., 118–119; Colorado colony, 67–70; cowboy culture, influence on, 120–121; cultural fusion of Southwest and, 245–247; exploration and settlements, 27–29, 50–51, 57, 59–60, 127–128; historical influence on Southwest, 1–14; horse saddles, 120; immigration from, 22–25; irrigated farming practices, 122–126, 140; New Mexico colony, 44–45, 57; sheep, raising of, 113–114; Texas colony, 59–60, 67–70. See also fantasy heritage Spanish-American, use of term, 54 Spanish-Colonial, use of term, 54 Spanish International Network, 280 Spanish language. See language Spanish-speaking, use of term, 54 Spanish-speaking peoples: Californios, 63–67; geography of, 27, 32–36; language patterns, Southwest, 236–245; in New Mexico, 88; Pimeria Alta (Arizona), 57–59; Tejanos (Texas), 59–63 Steele, General, 80 stereotypes, patterns of employment and, 171–172 Stevens, Thaddeus, 92 Stevenson, Coke, 218, 219 Stilwell, Hart, 220–221 strikes. See trade-union movement student movement, 265–268 sugar-beet production, 115, 142–148 Sunol, Antonio, 108 Suro, Roberto, 349 Swing, Joseph, 293

446  Index Tamayo, Rufino, 279 tanateros, 109 Tanton, John, 282 Taos Project, 231 Taylor, Paul S., 77, 80, 138–139, 145, 151 Tejanos, 59–63, 74, 218 television, 279–280 Telles, Raymond, xx, 263 Telles, Robert, 183 Tenayca, Emma, 271 TENAZ (Teatros Nacionales de Aztlán), 276 ten gallon hats, 120–121 Tennessee, 318–323 tepestras, 116 terrorism, effect of, 324–325, 347–353 Texas: border disputes, 36–39; bracero program, discrimination and, 218–221; Cart War, 77–78; cattle industry, development of, 118–122; Chicano leadership in, 263; contractlabor law, 140–142; Cortino War, 78–79; cotton industry, 133–138; the dead-line of sheriffs, 79–81; demographics (2000 and beyond), 318–320; early settlements and conflict, 71–74; economic systems and classes, 60–63; education, improvements in, 231–232; El Paso incident, 290–292; exploration and settlement of, 5, 6, 10, 27–29; farmworker strikes, xxi; fruit and vegetable farming, 138–140; immigrant population data (2011), 314; immigration data (1900–1930), 128; irrigated farming, 123; labor strikes, xxi, 153; lost provinces of, 67–70; Mexican-American War, 74–77; migratory labor, impact of, 146–148; Plyler v. Doe (1982), 367–368; Political Association of Spanish-Speaking Organizations (PASSO), 255; property rights, 125; Raza Unida party, 261–262; Salt War, 81–82; Sinarquista movement, 213–214; slaves and peons, 77–78; Spanish-speaking population, 27,

32–36; sugar-beet production, 143–145; Tejanos, 59–63, 74; Treaty of Guadalupe Hidalgo (1848), 29–32; Trivoli incident, 291–292; violence (1908–1925), 82–85. See also borderlands; demographic profile of immigrants; trade-union movement Texas Emigrant Agent Law (1929), 143–144, 146 Texas Good Neighbor Commission, 136 Texas Rangers, violence of (1909–1925), 82–85 Texas Republic (1836–1846), 73–74 Texas Revolution, 73–74 Thompson, Victor Rodman, 183–184 Tierra Amarilla, 260–261 Tierra y Cultura, 260 Tiguex, 4 Tijerina, Reies López, 252, 259–260, 285 Tijuana, 161–162 Tlascalan Indians, 44–45 Toba, Tirza de la, 162 Tolan Committee, 144–145 Tomachic War, 158 Tonatiuh International, 278 Torres, Francisco, 100 trade-union movement: Battle of Cananea, 159–161; César Chávez and, 252, 256–257, 283, 285, 294; Chicana leadership in, 271; farm worker strikes, xx–xxi, 256, 269, 278; GallupAmerican Company, 154; in Mexico, 159–162; mining strikes, 154–157; myth of docility, 149–154 Trambly, Estela Portillo, 276 transnationalism, women and, 343–355 Treaty of Guadalupe Hidalgo (1848), 29–32, 36–39, 74, 76, 85, 89, 92, 94, 125, 224, 259–260, 287 Trivoli incident, 291–292 Trucha Publications, 278 Trujillo, Teofilo, 96 Truman, Harry, 291, 298 Tuck, Ruth, 227 Twitchell, R. E., 44, 93

Index   447 Ugarte, Manuel, 84 Ulibarrí, Sabine, 276 unaccompanied children, 327 undocumented immigrants: amnesty for, 307–308; Arizona S. B. 1070, 377–379; Border Protection, Antiterrorism and Illegal Immigration Control Act of 2005 (H.R. 4437), 347–353; children of, 380; Deferred Action for Childhood Arrivals (DACA), 380–387; demographics (2000 and beyond), 324–325; DREAM Act (2001), 368–372; economic costs and benefits of, 303–304; education and, 365–369; history of, 289, 302–304; human rights and, 286; recent and future trends, 326–327, 342–345. See also demographic profile of immigrants; immigration unions. See trade-union movement United Farm Workers (UFW), 256–257 United Mexican American Students (UMAS), 266–268 United States border with Mexico, 36–39 United States Steel Corporation, 176–177 United We Dream (UWD), 371 Unity Leagues, 227 University of New Mexico, 55 Upland, 173 Urquiza, José Antonio, 213 U.S. Census: (1900) data, 288; (1930) data, 32–35; (1987) data, 285; (2010) data, 315; Mexican as census category, 32; poverty data, 317; undocumented individuals, data on, 342–343 U.S. English, 282 U.S. immigration policy: amnesty, 307–308; Arizona S. B. 1070, 377–379; Border Protection, Antiterrorism and Illegal Immigration Control Act of 2005 (H.R. 4437), 347–353; commuters, 301–302; Deferred Action for Childhood Arrivals (DACA), 380–387; DREAM Act (2001), 368–372; future trends, 325–328, 326–327; green carders,

300–301; Hart-Cellars Immigration and Nationality Act (1965), 298–299; Illegal Immigration Reform and Immigrant Responsibility Act (1996), 367; Immigration Acts (1921, 1924, 1965, 1986, 1990), 141, 142–143, 163, 297, 326–327; Immigration Reform and Control Act (IRCA, 1986), 331, 332–336; overview, 296; permanent visas, 300; quota system, 297; refugee legislation, 297–299; Simpson-Rodino Immigration Reform and Control Act (1986), 304–309 Utah, 133 Utes, 31–32, 44 Vaca, Cabeza de, 3–4, 8 Valdez, Daniel, 231 Valdez, Luis, 276, 278 Valens, Richie, 278 Valenzuela, Richardo, 278 vaqueros, 114, 120–121 Vargas, Diego de, 6 Vasconcellos, José, 73 Vásquez, General, 74 Vásquez, Richard, 276 Vásquez, Tiburcio, 99 Vásquez de Ayllón, Lucas, 2 vegetable farms, 138–140, 146–148 Vengeance Squad, 195–196 Ventura Press, 278 Ventura School for Girls, 204 vermilion cave, 108–109 Vermont, 319–320 Vigil, Domiciano, 90 Villa, Francisco, 82–83 Villalobos, Gregorio de, 118 Villalpando, Cathi, 274 Villarreal, José Antonio, 275 Villasenor, Pedro, 214 violence, women and the border, 355–360 Virginia, 320–323 visas, 300, 335–336 Viva Kennedy Clubs, 254, 255

448  Index Voces sin Eco (Voices without Echo), 359 Voorhis, Jerry, 210 voting: American Coordinating Council of Political Education (ACCPE), 255; American G. I. Forum, 253; Colorado La Raza Unida party (LRU), 258; Community Service Organizations and, 226–227, 252–253; demographics, xxi– xxii, 316; dual nationality and, 345–347; future trends, 316, 326; late 1970s–1980s, 263–271; Mexican-American Political Association (MAPA), 254–255; Mexican-American Youth Organization (MAYO), 260–262; movimiento, success of, 284, 285; national organization efforts, 270–271; patterns of, 175–176; Political Association of SpanishSpeaking Organizations (PASSO), 255; presidential race (2008), 372–377; presidential race (2012 and beyond), 379–387; Proposition 187 (1994) and, 340–341 Voting Rights Act, 270–271, 284

Wellman, Paul I., 95–96 Wells, Jim, 61 Western Federation of Labor, 155 Western Hemisphere Act (1976), 299 Westminster, California school segregation, 227–230 White, Charles, 211 Widney, J. P., 66, 236 Willard, Charles Dwight, 63 Wilson, Pete, 337, 338 Wilson, Woodrow, 297 Winchell, Walter, 209–211 women: border violence and, 355–360; perspective on borderlands, 330; women and transnationalism, 343–355; women’s movements and organizations, 271–275 Worcester, Donald E., 9 World Refugee Year, 298 Wyoming, 133, 318–320

Walter, Paul, 43 War Brides Act (1946), 297–298 War Food Administration, 215 War on Poverty, 316–317 Warren, Earl, 204 Warren, Nina Otero, 274 Washington state, 133, 153. See also tradeunion movement water rights, 124–126 Watts, community of, 133 Waxman, Al, 199, 206–207 Webb, Walter Prescott, 12, 38, 79, 81, 82, 124 Welles, Sumner, 221

Yaqui Indians, 157–158 Ynostrosa, Henry, 183 Young Citizens for Community Action, 268

xacal, 108 Ximenes, Vincente, xxii, 264

Zacatecas, 5–6 Zamora, Bernice, 277 Zamora, Gus, 183 zanjero, 124 Zapata, Carmen, 274 Zarate, Elías, 83 Zeleny, Carolyn, 43, 44, 55, 95 zoot suit, use of term, 190, 193–194 Zoot Suit riots, 187–191, 195–207, 214 Zuni Indians, 30–32

About the Authors

CAREY McWILLIAMS was editor of The Nation, 1951–1975, and author of Ambrose Bierce, Factories in the Field, Ill Fares the Land, Brothers Under the Skin, Prejudice, Southern California Country, California: The Great Exception, A Mask for Privilege, Witch Hunt, and The Education of Carey McWilliams. MATT S. MEIER was Patrick A. Donohoe Emeritus Professor in the College of Arts and Sciences at Santa Clara University, California. His previous works include Mexican American Biographies: A Historical Dictionary, 1836–1987 (Greenwood, 1988), Bibliography of Mexican American History (Greenwood, 1984), Dictionary of Mexican American History (Greenwood, 1981), and The Chicanos: A History of Mexican Americans (1972). ALMA M. GARCÍA is Professor of Sociology and Director of the Latin American Studies Program at Santa Clara University, California. She received her PhD in sociology at Harvard University. Her previous works include Chicana Feminist Thought: The Basic Historical Writings (coeditor) (1998); The Mexican Americans (Greenwood Press, 2002); Narratives of Mexican American Women: Emergent Identities of the Second Generation (2004); Ethnic Community Builders: Mexican Americans’ Search for Justice, Power and Citizenship Rights (coeditor) (2007), awarded the Oral History Association’s Award for Best Use of Oral History; and Contested Images: Women of Color in Popular Culture (editor) (2012), which garnered the Susan Koppleman Award for the Best Anthology in Feminist Studies in Popular and American Culture. Her article “The Development of Chicana Feminist Discourse, 1970–1980,” Gender & Society (1989): 217–228 was selected by the University of Memphis Center for Research on Women as one of the fifty Classic Articles on Race and Gender in 1997.