After the Map: Cartography, Navigation, and the Transformation of Territory in the Twentieth Century 9780226339535

Citation preview

After the Map

AFTER THE MAP Cartography, Navigation, and the Transformation of Territory in the Twentieth Century

William Rankin

University of Chicago Press Chicago and London

William Rankin is assistant professor of the history of science at Yale University. The University of Chicago Press, Chicago 60637 The University of Chicago Press, Ltd., London © 2016 by The University of Chicago All rights reserved. Published 2016. Printed in the United States of America 25 24 23 22 21 20 19 18 17 16 1 2 3 4 5 ISBN-13: 978-0-226-33936-8 (cloth) ISBN-13: 978-0-226-33953-5 (e-book) DOI: 10.7208/chicago/9780226339535.001.0001 Publication of this book has been aided by a grant from the Neil Harris Endowment Fund, which honors the innovative scholarship of Neil Harris, the Preston and Sterling Morton Professor Emeritus of History at the University of Chicago. The Fund is supported by contributions from the students, colleagues, and friends of Neil Harris. Library of Congress Cataloging-in-Publication Data Names: Rankin, William, 1978– author. Title: After the map : cartography, navigation, and the transformation of territory in the twentieth century / William Rankin. Description: Chicago ; London : University of Chicago Press, 2016. | Includes bibliographical references and index. Identifiers: LCCN 2015037815 | ISBN 9780226339368 (cloth : alk. paper) | ISBN 9780226339535 (e-book) Subjects: LCSH: Cartography—History. | Navigation—History. | Global Positioning System—History. | Electronics in navigation—History. | Maps—Political aspects. | Cartography—Methodology. | Grids (Cartography) | Universal transverse Mercator projection (Cartography). Classification: LCC GA102.3 .R36 2016 | DDC 526.09/04—dc23 LC record available at https://lccn.loc.gov/2015037815 ♾ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

CO NT E NTS

Possibly Ambiguous Terms Introduction I.

vii

Territory and the Mapping Sciences

1

The International Map of the World and the Logic of Representation

1 2

The Authority of Representation A Single Map for All Countries, 1891– 1939

23

Maps as Tools Globalism, Regionalism, and the Erosion of Universal Cartography, 1940– 1965

65

II. Cartographic Grids and New Territories of Calculation

3 4

Aiming Guns, Recording Land, and Stitching Map to Territory The Invention of Cartographic Grid Systems, 1914– 1939

119

Territoriality without Borders Global Grids and the Universal Transverse Mercator, 1940– 1965

163

III. Electronic Navigation and Territorial Pointillism

5 6

Inhabiting the Grid Radionavigation and Electronic Coordinates, 1920– 1965

205

The Politics of Global Coverage The Navy, NASA, and GPS, 1960– 2010

253

Conclusion

The Politics in My Pocket

Acknowledgments

301

Acronyms and Codenames Notes

309

Index

377

295

305

Color gallery follows page 280 For high-resolution images, raw data, and a sortable bibliography, visit www.afterthemap.info.

v

P O SSI B LY A MB I G UO US T E RMS

Small and Large in Cartography. The terms small scale and large scale might mean two things when describing a map: either the size of the land area shown, or the size of the ratio between real-world lengths and lengths on the paper. I follow the standard vocabulary of cartographers and geographers and use these terms in the second sense only. That is, a small-scale map is one that shows a lot of land at once, while a large-scale map shows a smaller area in greater detail. World maps are very small-scale maps, since the ratio between map length and real-world length can be 1:100,000,000 or smaller (1 centimeter on the map equals 100 million centimeters— 1,000 kilometers— on the earth). City maps and property surveys use a very large scale, typically 1:10,000 (1 centimeter = 100 meters) or larger. Notice that the number 1/100,000,000 is ten thousand times smaller than 1/10,000. Regional. This term likewise has several possible meanings. In national contexts, regional usually refers to a relatively well-defined subdivision of a territorial state, such as New England in the United States, the Mezzogiorno in southern Italy, or Inner Mongolia in China. In an international context, regional instead refers to a grouping of multiple countries: the regional NATO alliance, for example, currently consists of twenty-eight countries on both sides of the North Atlantic. But additionally, regional can also refer to areas that do not align with political boundaries at all. Often these regions are climatic or biological— rain forests, deserts, plains— but they can also be cultural, linguistic, or historical. Examples might include everything from megaregions like the Silk Road or the African Sahel to the smaller areas of the Mississippi Delta or the South-American Pampas. In this book, I use regional in all of these ways, but my focus is always on geographies that challenge the primacy of national space. This means that I am usually referring to areas that include multiple countries or span international borders— something more like Central Asia than central Texas. vii

I NT RO D UC T I O N

Territory and the Mapping Sciences

During the two world wars, newspapers gushed at the unprecedented number of paper maps produced for use on the battlefield. Together the Allies printed roughly 65 million maps in the 1910s, and the US and UK alone printed over a billion in the 1940s— almost fifty maps for every soldier. Cartographers and military planners also worried openly about the need to educate both soldiers and civilians on the basics of cartographic literacy, since most people had little familiarity with the abstract symbols and accurately rendered topography of a state-of-the-art map. Between the wars, observers predicted the dawn of a “map-minded age,” where maps would be a “basic need” and make all geographic tasks— everything from military planning to civilian recreation— easier and more efficient.1 But a few decades later, press coverage of the Vietnam and Gulf wars barely mentioned maps at all. Instead, American headlines reveled in the new geographic precision of “smart bombs” and the almost magical promise of GPS— the Global Positioning System. Maps were certainly still important, but perhaps as only one component of a larger system, and pundits talked of a coming “revolution” (again, both military and civilian) in fields that had formerly been quite map centric— especially surveying, navigation, and environmental research. As it turned out, even the most optimistic of these predictions ended up being far too conservative. By 2010, there were roughly one billion GPS receivers in use around the world— one for every Allied map printed during World War II— and only a tiny fraction of these were used by the military.2 The goal of this book is to understand the larger stakes of this shift. With the creation of new forms of geographic knowledge, what is gained and what is lost? Who wins and who loses? More broadly, how do changes in the tools and methods of mapping provoke new ways of understanding and experiencing the world? Politically, all these questions operate at two scales at once. 1

At the micropolitical scale, they are about the creation of a new geographic subjectivity— a new way of seeing and interacting with the earth. But at the macropolitical scale, they are also about the reach of state rationality, the permeability of territorial boundaries, and the transition from European imperialism to the bipolar logic of the Cold War and the American dominance of the post–Cold War world. This book can thus be read at two scales as well. Narrowly, it is a history of the mapping sciences in the twentieth century that situates technologies like GPS within a longer trajectory of spatial knowledge. But more expansively, it is also a cultural and political history of geographic space itself. In most contexts— for specialists and nonspecialists alike— the obvious way to evaluate changes in geographic knowledge is in terms of accuracy, and it is usually fair to assume that more accurate knowledge translates directly into a better user experience and more political power.3 My approach, however, is different. Rather than focusing on the relentless rise of precision or ever more impressive feats of measurement and targeting, I am more interested in changes in the kind of knowledge produced. The term I use is geo-epistemology; what matters to me is not just what is known about the earth, but how it is known— and how it is used. Geo-epistemology is the difference between knowing your neighborhood through detailed stories, a pictorial guidebook, a map, aerial photographs, the coordinates of a GPS receiver, or simply walking around. It is about trustworthy knowledge (how can I know that my world really is what I think it is?), but it is also about our everyday existence in space (how do I understand my surroundings, my mobility, my relationship to others?). Above all it is about the importance— and the unavoidability— of tools: the goggles of geo-epistemology come in many styles, but they can never be removed. And although for most purposes GPS does indeed offer much more precision than a map, it also constructs a radically different relationship between user, landscape, and authority. The comparison is again both experiential and political. Maps operate through representation. They create a miniature version of the world and give us a detached view from above, with the messy complexities of reality simplified and reduced to a legible system of lines and colors. Think of the maps in war rooms or on negotiating tables, where knowledge of distant lands is centralized and assembled for the sake of large-scale strategy or the carving up of continents. The power of these maps lies in their ability to act as a stand-in for the original landscape, so that decisions can be made from afar and any new lines drawn with the diplomat’s pen can be scaled up and projected back into the world.4 But this is no small feat, and early twentieth-century surveyors and cartographers saw the task of representation as nothing less than a problem of scientific truth. Making a “truthful map” meant establishing rigorous rules that would govern the correspondence between map and world, and the virtues of objectivity, neutrality, and comprehensiveness were seen as the foundation of trustworthy cartography.5 Taken to an extreme, this faith 2

Introduction

in representation is what transforms maps (in the plural) into the map—a singular, universal record of geographic fact that includes everything worthy of attention, and nothing more. Armed with such a map, it is no longer even necessary to leave your desk: the world has come to you. Electronic systems like GPS work in a very different way. Rather than creating a miniature substitute for the world, the radio signals sent from GPS satellites instead create a full-scale system of coordinates that overlays and coexists with the physical terrain. The experience of using GPS— or its many predecessors, not all of which were electronic— is therefore much more geographically embedded than the experience of using a map. Rather than contemplating an overhead view of a large expanse of the earth, navigating by coordinates means inhabiting a virtual landscape of reference points, with your position always at the center.6 These coordinates have nothing to do with representation; instead they are simply about presentation— being present, reliable, and ready for use. And instead of any concerns with truth or objectivity, the designers of GPS described their system as a “utility,” one that transformed spatial location into a commodity available in much the same way as electricity or water— on demand, at the place of consumption.7 The power of GPS derives precisely from these qualities. Coordinates shift attention from the area to the point: a stable electronic grid makes it possible to aim missiles, drill for offshore oil, or conduct field research without any overarching awareness of a larger geographic region. The overall ambition is quite different as well. Being glib, one could say that with representation the goal is to know about a place without having to visit. With technologies like GPS, the goal is instead to visit a place without having to know much about it. Historically, the transition was not nearly this stark, and representational maps have certainly not disappeared. After all, some of the most important coordinate systems of the twentieth century were first developed within cartography itself, and for most people a GPS signal is only helpful if it is combined with a digital map or road database. If anything, the creation of new coordinate technologies has probably made the world more “map minded” than ever before. But it is also clear that technologies like GPS have significantly shifted both the way that maps are made and the way they are used, and representational maps do not enjoy the authority that they once did— epistemologically, culturally, or politically. We no longer live in a world where the map (in the singular) goes unquestioned. Instead it is increasingly the coordinates that take priority. My main argument is that this change in the logic of mapping— this shift in geo-epistemology— should be understood quite broadly as a shift in the nature of territory. In the early twentieth century, there was a very tight link between representational maps and a certain ideal of the territorial state. Maps gave real-world traction to abstract concepts like jurisdiction and sovereignty, and they reinforced a strong split between a space of domestic affairs and a space of foreign relations. The logic is relatively simple: spatially intensive Territory and the Mapping Sciences

3

activities like levying property taxes, managing forests, building railways, and defending boundaries all relied on detailed, large-scale maps, and this kind of mapping required sustained access to geographic space, both to produce the maps and to keep them up to date. The production of these maps was seen as the right (and responsibility) of official survey agencies, and each country, colony, or empire used graphics and survey methods that stopped at its borders. Maps therefore tended to reinforce an all-or-nothing relationship between territory and sovereignty, both as an ideal and as a practical reality, since the control of geographic space required control over the production of geographic knowledge, which in turn required control over geographic space. Conceptually, it was difficult to distinguish territory and sovereignty at all. But with the full-scale, pointillist logic of coordinates, there is no longer this tight relationship between geographic legibility and political authority. Electronic coordinates are explicitly designed to exceed the boundaries of individual states, and maintaining a virtual grid does not require any long-term geographic commitment or sustained control over a contiguous expanse of land. Not only does this make things like ultra-long-range bombardment possible as never before, but many of the activities that previously relied on detailed maps can now be pursued with a much lighter presence on the ground. Political responsibility for maintaining these boundary-crossing systems has also been divided in complex ways, and even systems developed and funded by one country alone have often come to be regulated by international agreement and subject to international laws. Both in design and in practice, electronic coordinates cut across the usual categories of the state. What this means is that it becomes possible to imagine territory as something separate from sovereignty. Territory need not simply be a cleanly bounded area of space— an inert “container” for the power of states or empires8—but can be understood as itself a distinct form of power created through geographic knowledge. The kind of consolidated and explicitly jurisdictional territory reinforced by mapping is about making claims to exclusive political authority within a well-bounded area. The unbounded, dispersed, and politically ambiguous territory of electronic coordinates, while not incompatible with these goals, is instead primarily about enabling quick intervention and creating a robust spatial framework— a way of thinking, acting, and governing— independent from any such claims. In the second half of the twentieth century, much of this framework was designed and installed by the United States, sometimes quite forcefully, and in many respects the geographic space we know today is both a result and a cause of American global power. The Americanization of geography was at once specific and general; it took place both at the level of individual technologies and at the more diffuse level of new methods, goals, techniques, and vocabulary. In most cases, the military was the primary sponsor. However, I should make it clear that I am not suggesting any straightforward linear relationship between new technologies and new forms of territory, and it would be a mistake to see 4

Introduction

any particular technology as decisive— not even GPS— or to see the US military as the ultimate origin of our new spatial infrastructure. Indeed, the major historical inflection point occurred during the decades surrounding World War II, at a time when new mapping and navigation systems multiplied not just in the US but also in the UK, France, and Germany, and when military goals often overlapped with civilian, commercial, and academic interests. Even in the decades after the war, American plans were commonly derailed or redirected by unexpected factors, both political and technological. And although the Cold War was a constant motivator, competition between the US and the USSR was often less determinant than competition between various projects within the non-Communist world. In other words, this is a story where American power is absolutely central, but it is not a story where the United States is the central protagonist. Indeed, the United States cannot be at the center: the history of mapping and the history of American globalism do not perfectly align, and especially in the early twentieth century the US simply was not the dominant player that it would become after World War II. But more to the point, the goal of this book is to understand a new kind of infrastructure— one that is successful exactly because it transcends the military or diplomatic strategy of any individual state. American programs of dollar diplomacy, cultural universalism, and Cold War containment must share space with the ambitions of multinational oil companies, collaborations among European scientists, and regional surveying projects in Latin America, East Asia, and elsewhere. After all, what made the technologies of gridded space so ubiquitous and transformative is that they were widely adopted— and sometimes co-opted— for new and unforeseen uses in dozens of nonmilitary fields all around the world. The transformation of territory is thus not a story of large-scale strategy, sweeping political-economic forces, or an intentional program of global domination. It is instead a story about the kind of power that was created when mapmakers and engineers, and eventually millions of everyday users, shifted their concern away from the lofty realm of truth and toward the practicalities of aiming, measuring, and navigating in new ways.

TERRITORY IN THE TWENTIETH CENTURY The central concern of this book is the history of territory. But what is territory? And how does it differ from sovereignty, jurisdiction, or property? At its broadest, territory can refer to almost anything spatial, and territoriality, by extension, can be defined even more broadly as essentially any strategy of spatial control.9 But my interest is in the territory of states: the kind of geographic structure that Max Weber had in mind when in 1919 he famously defined the state in terms of a “monopoly of the legitimate use of physical force within a given territory.”10 I understand territory differently from Weber. Territory and the Mapping Sciences

5

For him, and for countless others in the twentieth century, territory was a relatively unproblematic category: it was simply a geographic area defined by clear boundaries, and it was impossible to imagine a state— or at least a “modern state”—as anything other than a bounded spatial entity. Sovereignty and jurisdiction were folded into this same ideal, and they were both seen as inherently geographic (and perfectly coextensive). More recently, however, scholars in several fields have argued that territory is not nearly so simple and that the textbook definition in fact has a long and contested history that stretches back at least to the medieval era. But despite this emerging consensus about its early history, there is still no coherent story about the last hundred years, and territory in the late twentieth and early twenty-first centuries has seemed to be flourishing and disintegrating simultaneously. My argument is an attempt to address this apparent paradox. The early history of territory has mainly been told as a European story; the most important theme is that territory does not have any necessary relationship with sovereignty or jurisdiction, both of which have their own separate histories. For many centuries before sovereignty and jurisdiction were geographic, they were personal, and authority was exercised over people, not land. In late Roman and medieval Europe in particular, the jurisdiction of the emperor and the pope was universal, in line with the universalism of the monotheistic God, and sovereignty simply meant supremacy. In theory, neither was spatially bounded. Space was definitely still important— and villages and manors were often delimited with perfectly well-defined boundaries— but land was something owned by the ruler, not the thing to be ruled. The spatial boundaries of justice, religion, trade, and taxation were also rather fluid and generally did not align, either with each other or with the shifting boundaries of kingdoms and principalities.11 Jurists began to challenge these social arrangements as early as the thirteenth century, but the change toward a more thoroughly geographic basis of authority (along with the separation of spiritual and earthy power) was remarkably gradual and lasted well into the nineteenth century. In particular, it is now quite clear that the political ideal of “the Westphalian state” has relatively little to do with the actual Treaty of Westphalia of 1648, and even in the modern era the territorial ideal is better understood as an ever-changing bundle of practices, political claims, and convenient fictions than as a monolithic legal fact.12 The historical development of cartography has played a major role in this narrative. Maps have not simply been passive observers to the geographic history of the state; instead, they are often what enabled governance to become geographic in the first place. Although the medieval world was not entirely devoid of maps (or rather, documents classified today as maps), they were not common and they were generally not used for practical tasks. For example, the famous Domesday Book, which was a tax record of most land and property in England and Wales in the late eleventh century, contained only text and was organized not geographically but according to the hierarchy of lords. Land 6

Introduction

was also measured primarily in terms of productivity rather than strictly by geometric area.13 Large-scale maps only began being produced in the sixteenth century, and they tended to appear as a result of active power struggles between elites and commoners over new projects of taxation and agricultural management. Systematic national borders likewise only began appearing on maps in the seventeenth century, simultaneous with the first attempts to systematize boundaries on the ground.14 The history of cartography has been especially helpful for spotlighting parallels between early modern state formation in Europe and other areas, especially East Asia, where large-scale maps began flourishing at nearly the same time and for similar purposes. It has also provided a powerful way of unpacking the persistent fiction of the “nation-state”—in Europe and elsewhere— since maps have been one of the most visible anchors of nationalism and have supported countless claims to imagined homelands and projects of cultural consolidation.15 The implication, in other words, is that the history of territory— both as a widespread administrative strategy and as a cultural imaginary— cannot be separated from the strategies used to make geographic space legible and manipulable. Geo-epistemology and territory go hand in hand. Turning to the twentieth century, however, the relationship between territory and sovereignty becomes much less clear, and the role of cartography is ambiguous at best.16 In particular, we are confronted with two competing narratives, both of which are confident and triumphant, but which point in sharply opposite directions. The first is about the remarkable strengthening and expansion of state territoriality and the almost totalizing dominance of the nation-state ideal. Figure 1 shows the basic story: compared to the massive imperial consolidation and dramatic exchanges of territory that took place in the century or two before 1920 (and to a lesser extent in the 1930s as well), the overwhelming trajectory since 1945 has been the crumbling of empires and the creation of dozens of newly independent states— but without great changes to existing borders. Indeed, with only a few important exceptions, both decolonization and the breakup of the Soviet Union were mostly an exercise in transforming internal administrative or jurisdictional boundaries into external international borders, all in the service of finding ever-closer spatial alignment between ethnicity, law, and political order. To be sure, there have definitely been ongoing disputes over borders— often devastatingly bloody and protracted— but it is remarkable how rarely these conflicts have in fact resulted in a gain or loss of territory, and on the whole it is much more common for territorial disputes to be frozen than resolved.17 This same trajectory can also be seen in changes to international law. After both World War I and World War II, the major peace treaties explicitly embraced national self-determination and renounced the use of military force as a legitimate basis of territorial expansion, and enforcing these norms was one of the main goals of both the League of Nations and the UN. The same principles were likewise written into a series of other important treaties, including Territory and the Mapping Sciences

7

Figure 1: The hardening of territory since 1945. The main theme here is the triumph of the nation-state ideal, especially through territorial partition. (Unification has been much less prevalent.) Most of the shaded countries were created through the breakup of European empires or the Soviet Union, but there are also many examples of countries splitting for internal reasons. In almost all cases, these breakups have taken place along preexisting internal boundaries, and although disputes over international boundaries have been common, they have not led to any great changes in the boundaries themselves. Even major disputes— Kashmir, Western Sahara, the Crimea— tend to preserve boundaries rather than erase them. For high-resolution versions of all images, see www.afterthemap.info. Unless otherwise noted, all graphics are my own.

the 1933 Montevideo Convention (which defined sovereignty in terms of internal territorial capacity rather than external recognition), the 1959 Antarctic treaty (which repudiated the patchwork of territorial claims below 60° south), the 1975 Helsinki Accords (which finally recognized the legitimacy of Soviet territory), and many of the treaties delimiting the borders of newly formed countries.18 Overall, this was a remarkable inversion. At the beginning of the twentieth century, capturing territory had been one of the main goals of war; by the end of the century, perhaps the only internationally legitimate use of force was to prevent such annexations. Alongside this dramatic political shift was a subtler but no less important transformation in the spatial limits of territory. As shown in figure 2, national territory moved from being understood as a two-dimensional area of land to being explicitly defined as a three-dimensional volume of earth, air, and water, with different legal-political rights being maintained in different zones. This was likewise a gradual transformation. Airspace sovereignty was first recognized just after World War I (especially in response to the advent of 8

Introduction

Figure 2: An idealized cross section perpendicular to an ocean coastline, showing the feathered edge of state sovereignty in the late twentieth century. Instead of a political claim over a two-dimensional area of land, international law has come to define national territory as a three-dimensional volume of earth, air, and water with multiple boundaries. Most of these limits are defined by the UN Convention on the Law of the Sea, which has been under negotiation and refinement since the late 1950s.

aerial bombing and reconnaissance) and then given an upper bound with the Outer Space Treaty of 1967. In the oceans, the traditional idea of “territorial waters” from the late eighteenth century— generally limited to three nautical miles, or the typical range of cannon fire from shore— began to be challenged in the 1940s, first by the United States (for oil and gas rights) and then by smaller countries like Chile and Iceland seeking greater control over fishing resources. These unilateral claims eventually led to the negotiation of two UN Conventions on the Law of the Sea: the first in the late 1950s, the second in the 1970s and early 1980s. The second such treaty established a uniform twohundred-nautical-mile “Exclusive Economic Zone” for all countries and gave precise geometric rules for claims to the continental shelf, which were again anchored by new forms of mapping.19 Figure 3 gives a startling illustration of how these new rules have been applied to the Arctic Ocean, which has now been almost entirely carved up and claimed as territory.20 Territory and the Mapping Sciences

9

Figure 3: As of 1971, territorial claims in the Arctic were largely unilateral and ill defined. By the early twenty-first century, national rights had expanded well beyond shore, and all Arctic countries were preparing or revising formal documentation of claims to the continental shelf based on extensive new mapping. Only a few small areas are likely to remain unclaimed. Maps based on official submissions to the UN Commission on the Limits of the Continental Shelf, http://www.un.org/Depts/los/clcs_new/commission_submissions.htm.

Taken together, the dissolution of empires and the expansion of national territorial rights represent a resounding affirmation of state territoriality. This is not to say that imperial power asymmetries have been eliminated or that decolonization was a clean, benevolent process. Indeed, it was often ruthlessly strategic, not just for the metropole and the colonial elite, but also for the United States. But the change is nevertheless striking. Not only has the international political system increasingly come to approximate the ideal of national self-determination and inviolable boundaries (at least in theory), but territorial states remain foundational for stabilizing property rights and managing natural resources. And this is not just a political issue— it is also straightforwardly geographic. The other narrative of the twentieth century is the equally dramatic story of globalization. At its most exuberant, the trajectory here is almost a complete reversal of the first. Instead of focusing on periods of war and the importance of international diplomacy, the story usually begins with the economic changes of the 1970s— in particular the growth of transnational corporations, the expansion of global telecommunications, and the rise of neoliberalism— and tracks the inexorable erosion of the territorial state by the forces of global capitalism. Instead of the familiar jigsaw-puzzle world of stable nation-states, this world looks more like the one in figure 4: a network of global megacities presiding over a nonterritorial flow of finance capital, supply chains, mobile labor, and the immateriality of cyberspace. Or, in short: networks corroding territory.21 Although this narrative can easily be exaggerated to eliminate any role for territorial states, or even geography altogether, it is more properly seen as a story of political economy that is driven as much by policy as by the profit motive. This is especially true at the international scale, where entire economies have been retooled— either from within or through external pressure— in line with a new “market fundamentalism” of deregulation, free trade, and currency fluctuation. There is even an important role here for international law, especially the unique framework of the European Union (notably its elimination of internal passport controls), the regulatory and “development” agencies of the United Nations, and the return of a doctrine of “universal jurisdiction” as a justification both for new forms of international justice (the International Criminal Court) and for humanitarian military intervention outside the UN system (in Kosovo, Sierra Leone, and elsewhere).22 At stake here is not the existence of territory, but its relevance: the globalized world is one where states have lost the ability to fully govern their economies, their borders, and even the legitimate use of force. Historians and geographers have been keenly interested in both of these trajectories, but the main interpretive responses have tended to be divided along disciplinary lines. The most common response, especially from historians, has been to rein in some of the excesses of the globalization narrative, but without challenging its basic premise. Charles Maier, for example, describes territoTerritory and the Mapping Sciences

11

Figure 4: This map of telecommunications cables from 2009 nicely encapsulates the geographic narrative of globalization: it shows connections between cities in terms of bandwidth, uses a different scale for each major region (Africa is shown smaller than Europe!), and ignores most of Asia. Geographically, the main concern here is with networks rather than territory; politically, the relevant categories are megaregions and cities— not states. Image excerpted from Global Internet Map (Telegeography, 2009).

riality as itself a fully global political-economic regime— one that emerged in the mid-nineteenth century and was always heavily technological— but he nevertheless sees a clear shift to a “post-territorial” regime in the 1970s.23 Charles Bright and Michael Geyer likewise stress the interaction of the political and the economic in their well-known analysis of modern “regimes of world order,” but when analyzing the spatial changes of the late twentieth century they focus squarely on “the geography of global capitalism” and argue unequivocally that “it is markets that integrate the world, not states and their international organizations.”24 This response also overlaps with some of the strongest arguments about the emergence of an American empire, especially one that is modified by adjectives like “soft,” “informal,” and “economic.” Neil Smith, for example, has analyzed twentieth-century globalism as a forceful national project of an ascendant United States consolidating its power through “the more abstract geography of the world market rather than through direct 12

Introduction

political control of territory.”25 In this view, there is nothing inevitable or apolitical about global capitalism, but the dominant narrative of the twentieth century is still seen as a transition from explicitly territorial power to new American strategies of nonterritorial hegemony. The other main response, which has come mostly from geographers, is not so quick to discard territory. Indeed, the starting point is that the alleged dichotomy between territory and network only reproduces the dichotomy between state and market at the center of the neoliberal worldview. John Agnew, for example, argues that there is no zero-sum relationship between sovereignty and globalization (or territory and network) and instead calls for a more “pluralist” approach that sees a variety of intersections between “regimes of sovereignty” and “modes of spatiality.” The waning of the national territorial state need not imply that either territory or sovereignty are otherwise obsolete.26 Saskia Sassen’s work has pointed in a similar direction: not only are “state-centered border regimes” not as cleanly territorial as we might expect, but the persistence of state territoriality is perfectly compatible with an ongoing process of “debordering,” “rebordering,” and the creation of new competing territorialities at both the global and the local scale.27 Others have proposed similar strategies for avoiding a simple rise-and-fall narrative of territory, the most prominent of which either deny any conflict between territory and network altogether or argue for a much more active understanding of territory as an ever-present historical process— one where every act of deterritorialization is accompanied by new forms of reterritorialization.28 This work provides a sophisticated analysis of globalization; it also resonates nicely with historical accounts of territory before the modern era. Scholarship in geography also offers an important framework for thinking about twentieth-century cartography and the fate of representation. Since the early 1990s, geographers have drawn from a wide assortment of social and literary theorists— including David Harvey, Henri Lefebvre, Jacques Derrida, Roland Barthes, and Michel Foucault— to criticize the alleged objectivity and epistemological transparency of representational maps. Not only do maps inevitably codify the interests and worldviews of their makers, but even the conceptual distinction between maps and any preexisting “real world” obscures the degree to which cartography constructs, rather than simply reflects, reality.29 In the last ten years, this critique (and its borrowing from critical theory) has expanded into a fully postrepresentational approach that frames mapping as a central part of the ongoing process of territory, one where mapmaking, map interpretation, and map use all become part of what Rob Kitchin and Martin Dodge call “a process of constant reterritorialization”—that is, a constant reaffirmation of certain arguments about land, law, and power.30 This is a compelling framework, and it is deeply resonant with my own research. Yet as a historian it strikes me that the post in postrepresentational is usually meant only in a theoretical sense— that is, as more sophisticated than the usual ideas about representation— rather than in any historical sense relating to new forms of mapmaking or the advent of technologies like GPS. Territory and the Mapping Sciences

13

There is, however, strong continuity between these scholars’ ideas and postwar changes in the workaday world of mapmaking. American military cartographers, for example, went through their own crisis of representation in the 1950s, prompted not by concerns with objectivity or epistemology but by the changing needs of jet pilots and global troop deployment. They responded by redesigning maps and supplementing them with new positioning and navigation technologies that were, in their own way, fully postrepresentational.31 Likewise, although the scholarly critique of mapping has highlighted the importance of the shift away from representation (and “the map” as an ideal of universal scientific truth), it has tended to promote nonrepresentational ideas as inherently— even morally— better. But, historically, it is not at all clear that this transition from representational mapping to GPS was an unambiguous improvement. We thus find ourselves with historical arguments that maintain a largely Weberian understanding of territory and geographic arguments that offer analytic clarity about territories, networks, and representation without much historical engagement. My argument attempts to bridge this divide, and it intervenes in these debates in two ways. First, my work provides historical support for the geographers’ denial of any clean dichotomy between the hardening of territory and the debordering of globalization. The very same technologies that were developed to make borders more permeable have also been used to make them more stable and enforceable— especially in the oceans. Electronic coordinates in particular are not just used for aiming missiles or navigating from A to B; they are also used for stabilizing international boundary surveys, enforcing fishing treaties, and bringing offshore oil and gas deposits within national jurisdiction. There is likewise no zero-sum competition between territorial and global space or between the state and capitalism. For one, nearly all the new mapping systems discussed in this book were either initiated or heavily supported by states themselves. (And it is worth pointing out that the United States applied these systems just as aggressively to its own territory as it did abroad.) Just as important, these state technologies were often developed in tandem with private corporations and enthusiastically embraced by a wide range of “nonstate” users, domestic and foreign alike.32 My second entry into this discussion is a more thorough modification of existing research. The history of mapping makes it clear that characterizing the twentieth century (or the 1970s) as a simple shift from an era of national territory to an era of global flows is problematic both chronologically and conceptually. First off, it is important to distinguish worldwide knowledge from the specific history of the word global and its geopolitics. Starting as early as the 1860s, mapping presents a nearly unbroken history of international collaboration, and geographic interest at a planetary scale stretches back many centuries earlier still.33 From the point of view of geographic knowledge, the major shift of the twentieth century was thus not a transition from national to planetary, but from one worldwide political-geographic framework to an14

Introduction

other. Before the 1940s, there was widespread interest in geographically extensive projects, but these were always framed as international in scope— that is, as the result of collaboration between states. Only during and after World War  II did the word global gain traction as a geographic description— and the counterpart to global was not national, but regional.34 This is the central political-geographic shift of this book, and it happened several decades before the economic changes of the 1970s. The crucial point, however, is that both the national/international space of the early twentieth century and the global/regional space of the late twentieth century were equally territorial— that is, equally concerned with making space legible and governable. As the chapters ahead make clear, the earliest systems of full-scale coordinates evolved directly out of representational mapping, and although coordinates ended up structuring space in a new way, GPS is clearly not a network— it is a system of coordinates, not pathways.35 It might be possible to invent some new term besides territory to describe the pointillist space of GPS, but I see no reason to treat mapping and navigation systems as territorializing only when they reinforce familiar ideas about territoriality. Again, geo-epistemology and territory go hand in hand. The main difference between national/international and global/regional space is instead in the status of political boundaries. With the representational mapping of the early twentieth century, political boundaries were also boundaries of knowledge and thus action— not just conceptually, but at a straightforwardly practical level, too. This same boundedness was also a conspicuous feature of the prewar predecessors to GPS. With the new mapping and navigation systems created during and after World War II, actionable geographic knowledge was instead purposefully made to span international borders. Although this was most clear in the case of coordinate-based technologies, it was also true for some paper maps. (Both maps and coordinates, in other words, were malleable— but only to a point.) This change from bounded to unbounded territory is what I have in mind when I describe territory as something separate from sovereignty: they may well align at certain times for certain purposes, but they can also diverge. My argument about territory is thus both historical and theoretical. Historically, I am affirming that something important did happen in the twentieth century and that the kind of all-or-nothing territory that Weber took for granted no longer exists (if it ever did). But rather than locating the turning point in the 1970s, I place it several decades earlier. And rather than trying to fit GPS into a dichotomy between the “traditional” space of national territory and the new, nonterritorial space of global networks, I see it instead as signaling a modification of territory itself.36 Theoretically, this means that we should not see territory as a category defined by legal, political, or economic geography alone. It is also a category defined by practices of knowledge, and its practical reality can often diverge noticeably from its more familiar meanings. By the end of the twentieth century, territory was not (solely) a bounded block Territory and the Mapping Sciences

15

of space perfectly coextensive with a particular sovereignty or jurisdiction. Instead there were new kinds of territories: territories defined as frameworks of points— neither a block of space nor a network of flows— that organized knowledge in new ways and facilitated new kinds of intervention and new kinds of governance.

STRUCTURE OF THE EVIDENCE How should one go about writing a global history of geographic knowledge from the late nineteenth to the early twenty-first century? Focusing on a series of specific case studies might miss the larger picture, while trying to write about all mapping would be as pointless as it is impossible. In order to negotiate between these extremes, I have written this book as a series of three studies of major international projects, one in each of the three principal branches of the mapping sciences: cartography, geodesy, and navigation.37 These projects are not case studies, at least not in the usual sense, but they are still specific enough to allow for focused research on the actual practice— rather than just the theory— of mapping. In cartography I analyze the International Map of the World, or IMW, which was a hugely ambitious scheme for all countries of the world to collaborate on a uniform atlas of unprecedented detail. It was first proposed in 1891, and its standards were given the force of international treaty in 1909; although its goals and organization gradually shifted in subsequent decades, it remained a going concern until the 1980s. Over the course of its life, nearly every country in the world participated in some way, and thousands of maps were produced. In geodesy— which in a narrow sense refers to the study of the size and shape of the earth but also includes high-precision surveying more generally38—I trace the history of the Universal Transverse Mercator (UTM) system, a grid-based alternative to latitude and longitude that was created by the US Army in the late 1940s and was in widespread use for both military and nonmilitary purposes by the late 1960s. The mathematics of UTM gave every point on the earth a coordinate that was much better suited to calculation and coordination— of everything from missile trajectories to highway construction— than traditional spherical coordinates. It was a global expansion of earlier systems known generically as grids (as opposed to the graticule of latitude and longitude) that had been invented during World War I. Finally, in navigation I analyze the development of radionavigation systems, from the first efforts in the 1910s through the great proliferation of systems during and after World War II to the sponsorship of GPS by the US Department of Defense in the early 1970s. Not all of these systems relied on the point-based logic of coordinates, and only a very few were global or used satellites. But they were still important precedents for GPS, which only be-

16

Introduction

came the singular dominant system we know today at the turn of the twentyfirst century. These three projects— the IMW, UTM, and GPS— were all singular endeavors that, each in their own historical moment, attempted to organize all geographic knowledge. All three spanned several decades of development and debate, engaged scores of scientists from dozens of countries, received wide notice in nonspecialist circles, and were ambitious even to the point of hubris. Each project also intersected with a wide variety of less prominent initiatives. The International Map of the World, for example, was used as a template for several other international collaborations as well as hundreds of national and unofficial maps. The UTM grid was only the most ambitious of literally thousands of similar schemes installed around the world. And in the last forty years GPS has competed both with earlier ground-based alternatives and with several other satellite systems, of both US and foreign design. Tracing the life (and afterlife) of these megaprojects is thus an easy way to navigate the personal, institutional, and intellectual structure of the mapping sciences as a whole, both at the level of ideas and the level of practice, and it is impossible to separate them from any background “context” that was not itself shaped by their influence.39 More important, however, the IMW, UTM, and GPS together form a remarkably unified historical narrative. Spatially, this narrative is about the emerging logic of the grid and its significance as a new way of structuring knowledge. The abstract coordinates of latitude and longitude, of course, have their roots in ancient Greece and have been important politically since at least the early modern era. But in the twentieth century, the gridding of space took a much more geographically intensive form and came to organize a much wider range of everyday activities. In short, the figure of the grid is what connects representational mapping with the virtual space of electronic navigation, and the three projects I analyze in this book intersect as part of this larger trajectory. Many of these intersections are quite specific. For example, the primary designer of the UTM grid— an American astronomer and mathematician named John O’Keefe— drew explicitly from the technical standards of the IMW when creating his global coordinate system. Grids were likewise an organizing metaphor for the new radionavigation systems developed during World War II, and several key figures in the mapping and geodetic projects of the 1930s went on to do work with radionavigation and satellites in the 1950s.40 All three projects also intersected with many of the same institutions. The US Army Map Service was an especially important hub not just for cartography but also for advanced geodesy, radiosurveying, and early satellite work. Other organizations— notably the International Civil Aviation Organization and the International Geographical Union— were also important sites of ongoing debate. The IMW, UTM, and GPS are thus three separate episodes, but they are episodes in the same story.

Territory and the Mapping Sciences

17

At the same time, however, I make no claim that these three projects give anything like a comprehensive history of mapping in the twentieth century. For instance, I make relatively little mention of aerial photography or remote sensing, and the rise of electronic mapping and GIS (Geographic Information Systems) is barely part of my story at all. As other historians have shown, these are hugely important developments with serious political (and territorial) implications, and I do not mean to minimize their importance. But for the most part these other systems can all be seen as part of the history of representational mapping, and they have never seriously challenged either the view-from-above experience of cartography or its political focus on coherent blocks of space. (And when they have, it is in tandem with new coordinate systems or GPS).41 Just as important, a list of the major organizing projects of twentieth-century geographic knowledge only has three entries, and there is simply nothing comparable to the IMW, UTM, or GPS in other subfields of mapping.42 My choice of the project as a unit of historical analysis is closely related to my methodological focus on tools as hybrid objects that exist equally in the realms of thinking and doing. After all, the primary goal of the IMW, UTM, and GPS was not just to discuss or debate, but to make. By taking this approach, I am participating in a growing body of work in the history of science and technology that focuses on materiality, instrumentation, and the infrastructure of scientific practice. The great advantage of this kind of research is that it breaks down any division between theory and practice or between the dryly technical and the supposedly richer categories of cultural, political, and intellectual. As many other scholars have shown, instruments and tools are not simply the means by which theories are discovered, and epistemologically they are neither neutral nor transparent. Instead, tools do real conceptual and intellectual work; they are a way of thinking, a way of opening up new questions, and a way of making decisions. And this is just as true for everyday tools like a map or a GPS receiver as it is for the specialized instruments of laboratory scientists. In other words, maps, coordinate systems, and navigation devices are forms of cognition. They are the way that territory is both understood and performed.43 Writing a history of tools thus makes it possible to avoid some of the blind spots of more familiar forms of legal, political, and economic history, which have a tendency to compartmentalize rather than integrate— separating, for example, technology from politics or concepts from practices. And while I do not want to suggest that the history of territory is only about mapping, I do contend that territory cannot be understood separate from its tools. At a broader level, however, this interest in practical tools is also a working theory of historical agency— that is, of who and what matters. I see two steps to writing a history of tools: first is understanding why the tools were designed a certain way; second is understanding how they work and how they have been used. This means that, in the first instance, I focus on relatively 18

Introduction

undersung mapmakers and engineers (and the institutions that guided and supported their work) rather than on famous geographers, diplomats, or public intellectuals. These relatively anonymous individuals and agencies were perhaps not terribly glamorous, but they were certainly prolific and often remarkably creative. They are what the British in World War II called “boffins,” or what the Americans would later call “wonks”—trained specialists, usually from top-tier universities, with a serious commitment to their careers and their field. And while their personal and intellectual biographies can often be compelling— especially since a surprising number came to the mapping sciences from neighboring fields like geology, astronomy, mathematics, or even chemistry— these are not heroic individualists. They are professionals, and they focus on the task at hand. Yet as much as I want to argue for the importance of these experts, I also want to qualify their agency in two crucial respects. First, it is important to distinguish agency from intentionality. Near the end of his life, Michel Foucault described this distinction very precisely, arguing that “people know what they do; they frequently know why they do what they do; but what they don’t know is what what they do does.”44 In other words, I focus on cartographers, geodesists, and engineers because they are the practitioners of territory, but I make no claim that they saw— or even cared about— the larger trajectory presented in this book. Words like sovereignty and territory rarely appear in my sources, and I imagine that many of my historical actors would be reluctant to see their work as political at all. Second, my interest in the agency of designers is primarily a convenient strategy for understanding the agency of the objects themselves and the people who use them. Ultimately, the tools of mapping are not just used by specialists. They are also used by thousands of related experts without geographic training— development planners, civil engineers, military strategists— along with tens of millions of soldiers, hikers, and homeowners. This activity is often difficult to document, but it is central to my story, especially since many of the eventual uses (and users) of the tools I analyze were entirely unexpected at the time of design.45 At the heart of this methodological approach is thus an empirical claim about the history of mapping. In my story there are certainly important individuals whose decisions changed the course of history, and there are institutions— from government agencies to private corporations— that wielded significant influence. But there are no central figures that guide the action, and there is no single prime mover that can explain the overall trajectory— not the grand forces of American hegemony, the Cold War, decolonization, or global capitalism, and not the local forces of professional ambition, scientific idealism, or the preferences of users. And how could it be otherwise? All history is multicausal, and all agency is constrained. There is, however, one clear unifying pattern: the main site of action is the state, whether in peace or in war. Individuals work for and through state agencies (and vice versa), and commercial and academic projects advance when they enroll or reinforce state interests. Territory and the Mapping Sciences

19

My three studies comprise two chapters each. For the International Map of the World and the Universal Transverse Mercator grid, the dividing line between chapters is the beginning of World War II. In both cases the war provoked a crucial reorganization of earlier practices, especially those relating to national borders and national responsibility for geographic knowledge. For radionavigation systems, the historical transition is less clear cut, and instead my chapters divide local and regional systems from those with “worldwide” coverage (however defined). Note in particular that this transition, which occurred roughly in the 1960s, does not align with any simple technological difference between terrestrial and satellite systems; it is instead primarily a political one. In each chapter, I give attention to both the micropolitics of experience and the macropolitics of treaties, technological systems, and military strategy. This juxtaposition of the familiar world of everyday use and the worldhistorical realm of military goals and international organizations is quite intentional. One of the persistent themes in this book is that military technologies build on civilian and scientific precedents and are quickly civilianized upon completion— often to the chagrin of the military itself. The politics of geo-epistemology are thus complex and ambiguous, and everyday users are implicated just as much as scientists and military engineers. In the conclusion, I return to my opening questions. What are we to make of GPS? What is gained, and what is lost? Ultimately, this is not a story about them— the military, the powerful— transforming the world from on high. It is a story about us inhabiting and participating in new geographies and new forms territorial power, often unaware that our tools and assumptions matter as much as they do.

20

Introduction

PA R T  I

The International Map of the World and the Logic of Representation

CH A PT E R O NE

The Authority of Representation: A Single Map for All Countries, 1891– 1939

Consider the map shown in figure 1.1. It presents a large area of the Appalachian Mountains, including, as its title suggests, all of the Hudson River valley. It also shows Philadelphia, New York City, and a small slice of Canada. Its graphics should be familiar to anyone with an atlas on their living-room table or maps on the walls of their classroom. Water is shown in blue, low elevations are green, and mountain areas are pale orange. (Higher elevations would be bright red.) The lines on the map show railways and major roads, and the names of various towns, counties, states, and topographic features are distinguished with a simple hierarchy of symbols and typefaces. The borders of the map are cut by lines of latitude and longitude, and one can reasonably assume that the neighboring map— say, of Ottawa and Montreal to the north— would use these same graphics. In fact, the graphics here are so straightforward and intuitive that they hardly seem notable at all. They don’t seem to emphasize anything in particular, and they give equal balance to the physical, artificial, and political landscape. The immediate effect is that of looking at the world itself: instead of lines and colors on a piece of paper, we simply see rivers, mountains, cities, and boundaries. This map was published in 1927 by the principal civilian mapping office of the US government, and it was part of the American contribution to a massive international effort to produce similarly intuitive maps of the entire world. The idea for this project had first been presented almost forty years earlier, in 1891, when a young professor of geomorphology from the University of Vienna— Albrecht Penck— had stood before his colleagues at the International Geographical Congress in Berne and suggested that the time had finally come to consolidate all existing geographic knowledge. The best way to do this would be for all mapmakers to agree to a few simple standards to govern their work, so that maps published anywhere in the world could all contrib23

Figure 1.1 (see gallery for color version): Sheet of the International Map of the World, Hudson River, published in 1927 by the US Geological Survey. Detail of the Albany area is shown at actual size.

Figure 1.2: Countries adhering to the International Map of the World framework as of December 1913. After World War I, nine countries dropped out, including the USSR, Mexico, and Brazil. However, fourteen new countries joined, including ten that had not existed before the war. (See note 2 for sources.)

ute to a single universal atlas. Support from Penck’s fellow geographers was all but unanimous, and his project— soon known as the International Map of the World, or IMW for short1—became a long-lasting feature of twentiethcentury cartography. By the beginning of World War I, specifications for the map had been given the force of international treaty, and, as shown in figure 1.2, nearly every country in the world had officially agreed to participate.2 By the time the project was dissolved in the 1980s, thousands of sheets had been issued, nearly all of which followed the official standards quite closely. These maps— of Europe, Africa, and Asia just as much as New York and Canada— were all bounded by lines of latitude and longitude, projected at a scale of 1:1,000,000 (approximately sixteen miles to the inch), and drawn with the same school-atlas graphics. Together they promised to unify cartography, make the world legible to all, and create a lasting monument to the progress of human civilization. Not surprisingly, the grand vision of the IMW occupies an important symbolic place in the history of geographic knowledge. For Penck, the project marked a transition between the age of exploration and a new age of scientific synthesis. As he put it during his talk in Berne, “The era of breakthrough discoveries is over”; the task of the twentieth century would be “filling in the holes.” The new maps, however, would not just be more complete; they would also be more trustworthy— a “faithful picture in every sense,” a total and fiThe Authority of Representation

25

nal repository of stable fact.3 Later cartographers echoed this same rhetoric of universalism and finality. In 1913 the prominent British geographer Arthur Hinks effused that the IMW had inaugurated a “new era in cartography,” since “every sheet [is] written in the same language, without difference even of local idiom, so that who[ever] learns to read one sheet may read them all.”4 As late as 1972 the geographer-historian Norman Thrower made similar claims, using the project to introduce the “modern period of cartography” as a whole. Not only did the IMW signal the end of cartographic secrecy and “nationalistic parochialism,” but it marked the beginning of the kind of systematic mapping efforts that could finally be considered comprehensive, accurate, and technologically progressive.5 As the flagship project of twentieth-century cartography, the IMW is therefore a window into the geo-epistemology of representation as a whole. In particular, this chapter and the next use the IMW to trace the history of a certain ideal of what maps are for and how they work— one that was taken for granted by Penck, Hinks, and Thrower but which was progressively problematized during the second half of the twentieth century and is now openly criticized by scholars and practitioners alike. This is an ideal that I call authoritative representation. It is the assumption that the fundamental task of cartography is to create an objective, comprehensive, and politically neutral record of the world. Conceptually, this kind of map is nothing less than a paper replacement for the physical landscape. These foundational maps are authoritative in two senses at once. As records of geographic knowledge, they are the starting point for all subsequent mapmaking. Known simply as a “base map,” this kind of map is a “general” map, usually topographic, that can be used for any number of purposes, in contrast to “special” or “thematic” maps that show things like geology, population statistics, or weather patterns. Technologically, the idea is that these specialist maps can be made simply by overprinting new information onto the base map, or at least by using as many preexisting printing plates as possible. This shop-floor relationship is simultaneously a model for the institutional organization of cartography and the social organization of geographic research: it implies a hierarchy of mapmaking agencies and a linear progression from surveying to the base map to higher-order analysis. As a result, it is also a powerful argument that certain kinds of information— railroads, mountains, coastlines, administrative borders— are “basic” and universal. But since base maps are primarily produced by national mapping agencies, their scientific authority is matched by a strong political authority as well. They are the way that countries make their terrain legible and available for centralized administration; they also show the geographic limits of national control. Base maps are therefore a powerful political imaginary, transforming physical terrain into political territory. By the late nineteenth century, it was generally assumed that every country was responsible for producing its own set of basic topographic maps, and these were the maps that would provide 26

Chapter One

the source material for the IMW.6 For many decades, making these maps— and making the IMW above all— was seen as the principal way that governments could represent themselves geographically to other states, and this privilege was closely guarded. The history of the IMW is thus not just a record of a certain way of thinking about maps, but also of a certain way of understanding and managing territory. Visual representation and political representation went hand in hand. My two chapters on the IMW use World War II as a historical dividing line. Before the war, the international importance of authoritative representation went largely unquestioned, and the IMW was pursued with confidence and relatively little debate about its fundamental purpose. Despite its strong internationalist ethos, I argue that during this time it implicitly reinforced the assumptions of national territoriality, both through its graphics and through an ongoing debate about its political legitimacy. By the end of the 1950s, however, the project had become increasingly untenable. The once-strong relationship between the IMW base map and its thematic offspring unraveled, and prominent cartographers came to publicly challenge the worth of the entire endeavor. Although the project was not entirely abandoned— and even today there are those who lament its passing and pursue alternatives— by the 1970s the original goals of the IMW had become essentially unrecognizable, both scientifically and politically. Maps were still understood as representations of the world, and national survey agencies and national boundaries were still important parts of mapmaking, but representation no longer implied the same kind of authority as before. Instead of embodying comprehensive geographic truth, maps were increasingly seen as tools for specific functional tasks. And instead of national self-representation and worldwide uniformity, cartographers pushed for a new regional coherence that constructed a global realm quite different from the earlier world of internationalism. The visual logic of paper maps no longer aligned with the political logic of territorial control, and cartography ceased to provide a unifying framework for organizing geographic knowledge. I should stress at the outset, however, that my argument here has relatively little to do with the overall success or failure of the IMW itself. Although the official IMW standards were quite influential within cartography— by the late 1950s, for example, dozens of national and military map series had been explicitly modeled on IMW sheet layouts and graphics7—there is little evidence that the IMW was ever extensively used by politicians, scientists, or even geographers. Satisfactory international distribution of finished sheets was a constant problem, and practical applications of the map are surprisingly hard to find. (The one notable exception— the boundary negotiations at the Paris Peace Conference of 1919— is exceptional precisely because international distribution was unnecessary.)8 Other scholars have reached similar conclusions, and most discussion of the IMW today sees it primarily as a cautionary tale about the limits of international collaboration.9 For my purposes, however, The Authority of Representation

27

these assessments are somewhat irrelevant. If anything, they only underscore that maps today are evaluated more for their usefulness than their authority. Instead, what matters to me is that the project engaged the world’s most prominent cartographers, national survey agencies, and international organizations for the better part of a century. Indeed, it is perhaps precisely because the project did not always proceed smoothly that it was so widely discussed and thus provides such a useful guide to changes in cartography as a whole.10 This first chapter tracks the IMW during the fifty years after Penck’s initial proposal. My primary goal is to show how authoritative representation operated simultaneously in both a visual and a political sense— and how each reinforced the other. The chapter is divided into four sections. I begin by analyzing how the project evolved from Penck’s rough proposal of 1891 to capture the interest of states and eventually gain official international recognition— first in 1909, then definitively in 1913. In the second section I then offer my own analysis of the visual and political logic of the project; here is where I argue that the internationalism of the IMW was remarkably national in character, despite the explicit goal of forcing visual continuity across international borders. The second half of the chapter then tracks the reality of the project’s visual and political authority after 1914. During World War I, a large number of IMW-style maps were produced by British geographers (and eventually used at the Peace Conference), but in general the scale of 1:1,000,000 was not nearly detailed enough for trench warfare, and national mapping agencies directed their attention elsewhere.11 My focus is therefore on the period after international communication was reestablished in the early 1920s. I first analyze the usefulness of the IMW as a base map in the 1920s and 1930s before turning to evaluate the extent to which countries did (or did not) in fact publish maps of their own territory; this is where I discuss the British maps from World War I alongside later projects in Latin America and Asia. In both the visual and the political realm, the ideals of the project did not play out as planned, but the mismatches only reinforced the larger interest in geographic truth and national territoriality.

MAKING REPRESENTATION AUTHORITATIVE: FROM ACADEMIC PROJECT TO INTERNATIONAL TREATY More than twenty years elapsed between Penck’s original proposal in 1891 and the codification of the International Map of the World as an officially recognized project in 1909 and then, in final form, in 1913. During this time, Penck’s project transformed from an ambitious but somewhat vague plan for academic geographers to an exacting set of standards adopted by national mapping agencies. During this time, in other words, Penck’s plan became authoritative. The shift toward authority was a gradual one, and it relied on a close mutual relationship between increasing visual authority and increasing 28

Chapter One

political authority, where each required and reinforced the other. On the side of the visual, this mostly meant putting international cartography— and often cartography as a whole— on a more “scientific” footing. On the side of the political, this meant enrolling state sponsorship— both from the outside and from within— to actually produce the maps. Seemingly mundane discussions about the most trustworthy way to use colors to show elevation thus ended up being no less important than the division of mapping responsibility between countries, since both were necessary to ensure the objectivity and political legitimacy needed to justify the cost of the project. The IMW was hardly the first mapping project to link scientific virtue with political patronage in this way, but the specific historical path that the project followed in the decades before World War I promised a consolidation of geographic knowledge and representational practice that was unprecedented in its global reach. Penck’s initial goals, and much of the early discussion of his ideas, focused on scientific results alone. Even in his very first presentation, Penck made it clear that his highest hope was not just to make maps, but to perfect the study of geography. He justified the project primarily as a way to pursue comparative research into landforms and landscapes around the world, a task for which geographers would need a “unified representation”—a single, trustworthy map that could be perfected over time. For well-surveyed areas, this “faithful picture” would finally provide a “secure foundation” for geographical research; for less-well-surveyed areas, he called on his peers to finally collect “our entire topographical and orographical knowledge” into one place.12 In later presentations, Penck and his supporters likewise framed the project as a way for geography to hold its own among academic disciplines— especially astronomy, geodesy, and geology. Penck’s plan was presented as a direct riposte both to the geologists’ ongoing work on a unified geological map of Europe and to the astronomers’ massive Carte du ciel, a collaborative sky-mapping project that combined results from dozens of observatories around the world. Like these other projects, Penck’s map would rally geography around a common cause and provide an organizing framework for long-term work. It would be pursued mostly by academics, not territorial states, and it was predicted to take “decades” of work, coming to completion “in 50, perhaps in 100 years.”13 For Penck, international collaboration was likewise primarily a scientific issue, and it served his disciplinary goals in two important ways, with the details of his proposal following directly in turn. First was simply that making a new map would require collecting and comparing all existing geographic knowledge, and he saw this as the best way to discover and eliminate errors. At the same time, however, international collaboration would also force geographers to agree to some basic standards that could make this knowledge universally legible. In 1891 Penck hoped that making an international map might finally provoke agreement on the metric system, the Greenwich meridian, and the Latin alphabet, but even more pressing were standards of a particularly cartographic order. By far the most important was the scale of The Authority of Representation

29

the map. Penck described 1:1,000,000 as the perfect intermediate scale for unifying all knowledge, a kind of missing link between the generalized maps in atlases and the detailed work of national surveys. (For twentieth-century examples, see figure 1.3.) The scale was large enough that high-quality surveys in places like Europe could be summarized without losing too much detail, but it was small enough that the rougher surveys of Asia or South America would not be stretched too thin. One-to-a-million was also the scale that best balanced the portability and affordability of the resulting maps with the need to avoid mathematical distortion within the limits of any single sheet.14 The other important point of coordination was a scheme for subdividing the world using a rigid gridiron of latitude and longitude that would ignore both national and physical boundaries. His original proposal called for a grid of maps each five degrees on a side; a few years later this was revised to the 4° × 6° grid shown in figure 1.4. Instead of individual cartographers or countries deciding the limits of their maps according to their particular needs, this grid would eliminate both wasteful overlap and annoying gaps in map coverage.15 Armed with these simple standards, geographers would be assured that their work would fit into a larger, cumulative whole. The epistemological goal was thus to create what Peter Galison and Lorraine Daston have described as a scientific “super-observer,” a Leviathan of hundreds of individuals that could act as a unified author and record phenomena that no isolated scientist could ever observe on their own.16 Among Penck’s fellow geographers, interest in the plan was immediate and widespread, but the few dissenters highlighted the difficulty of relying on scientific coordination alone. The problem was certainly not a lack of scholarly interest: Penck’s project was enthusiastically endorsed not just by the Geographical Congress (on four separate occasions), but also by geographical societies throughout Europe and Russia and by individuals as far away as Japan, Venezuela, and the Congo.17 The most forceful argument against the map was instead that the scale 1:1,000,000 was simply too ambitious given the current state of geography. The German geographer Hermann Wagner, for example, described the map as a “phantom”—nothing more than a grid of latitude and longitude for which, in places like South America, there simply was no content.18 The famous Russian military geographer Aleksey Tillo similarly called the map “reckless,” arguing that “to every epoch there corresponds a certain average scale of our geographic knowledge,” and he estimated that 1:1,000,000 was at least four times too large.19 The problem, in other words, was that academic agreement might never produce the sustained institutional support needed to advance the project. Wagner cautioned that the Geographical Congress, despite its scholarly worth, was little more than a traveling conference series; it had “no power” and was entirely unable to address the “financial side of the question.”20 Tillo likewise suggested that the first step should be to create a permanent and well-funded international organization that could collect maps and enforce standards.21 And indeed, when Penck took 30

Chapter One

Scale 1:2,000,000— a page from The National Atlas of the United States of America (USGS, 1970).

Scale 1:1,000,000— a sheet of the International Map of the World (Lake Superior, USGS, 1966).

Scale 1:250,000— a sheet from USGS Series V501 (Escanaba, USGS, 1967). Figure 1.3: Maps showing Washington Island, Wisconsin, at three different scales, all produced by the US Geological Survey in the late 1960s. Details on the left are at actual size; thumbnails on the right are one-tenth size. Notice the generalization of coastlines, topography, and road detail at smaller scales.

stock of the project in 1899, his great disappointment was that near-universal support for its two headline features— the scale and the gridiron— was still not enough to spur any new mapmaking. Support for these two issues also did not translate into agreement about details like the projection, the meter, the prime meridian, or even the proper name of the project. Eight years after his first presentation, almost nothing had been done.22 The next ten years were a turning point that largely proved Penck’s detractors correct: coordination and patronage could only advance together. But instead of an internationally funded mapping organization, patronage instead came directly from individual states engaged in an elaborate game of cartographic rivalry. This was largely spurred by high-ranking government mapmakers in Europe and the United States— both military and civilian— taking an interest in Penck’s project and using their state resources to make new maps. These maps provoked not only greater scientific standardization, but also greater attention from diplomats and legislators, which in turn led to more maps, more standardization, and still more political support. Overall, this was an intentional strategy that blurred scientific and political goals in service of an almost conspiratorial interest in increasing state mapping budgets. And although these nonacademic cartographers shared many of the same goals as Penck— truth, finality, and comprehensive knowledge— they essentially transformed the project from an academic pursuit to a problem of international relations. This decade-long process advanced in two main steps. The first step consisted of five years of simple one-upmanship. It was first initiated by Henri Berthaut, the chief of cartography (and later director) of the French army mapping service and one of Penck’s most enthusiastic supporters. In 1900, Berthaut presented a series of Penck-inspired maps that he had produced in order to follow the recent wars and uprisings in the Caribbean, eastern Mediterranean, Central Asia, and the Far East. This mapping gave French cartography— and the Paris meridian— a dual scientific and military publicity that was immediately interpreted as an international provocation.23 Figure 1.5 shows the response from other countries through 1905. Germany quickly published a series of 1:1,000,000 maps showing war-torn areas in eastern China and Korea, and the British likewise responded with a series to track various wars and diplomatic crises in Africa. Penck then used the 1904 Geographical Congress in Washington DC as a platform to urge the United States to make its own domestic series, and his request was greeted with gusto by Teddy Roosevelt. A few years later Russia likewise announced plans for a domestic series of its own empire.24 With all this new mapping in progress— most of which used different units, sheet boundaries, and graphic conventions— the second step consisted of a series of agreements to eliminate incompatibilities. These agreements began as private compromises between prominent cartographers, but they became increasingly official and inclusive. From the beginning, the goal was to use scientific standardization as a way to elicit greater levels of state support, and 32

Chapter One

Figure 1.4: Regulating grid for sheets of the International Map. Each sheet is 4° × 6°; there are 2,642 sheets total, but only about 900 that show major land areas. This grid is also used for reference: columns are numbered west to east from 1 to 60, and rows are lettered away from the equator from A to V. The Hudson River map in figure 1.1 is therefore sheet “North K-18.” Tasmania is “South K-55.” Note, however, that the final collection of maps could not all be fit together as shown here, since each map was drawn using its own individual projection (notice the nonrectangular border in figure 1.1). The only way to assemble multiple sheets would be as part of an enormous globe, more than forty feet in diameter.

cartographers used diplomatic channels only as part of their larger strategy. The first move was made at the 1908 Geographical Congress in Geneva, when the renowned American geographer Henry Gannett— recently retired from his position as chief geographer for the US Geological Survey and the US Census— quietly suggested to his British counterparts that the US Congress would be more likely to fund its domestic 1:1,000,000 mapping if other countries were committed to an identical program. In a series of informal dinners and private meetings, a handful of mapmakers from the US, UK, France, and Germany quickly found agreement about basic colors, graphic conventions, and the final trade-off between Greenwich and the meter.25 After these recommendations were endorsed by the full conference, the British military surveyor Charles Close— soon to be appointed director of the (civilian) Ordnance Survey, the British national mapping agency— suggested that if they were given to the Swiss government to be disseminated through official channels, he would be able to convince his government to convene an intergovernmental conference that would elevate the geographers’ nonbinding standards to the level of international treaty, to be known formally as the International Map of the World.26 The first official IMW conference was held in 1909 in London, with twelve countries in attendance and much success, but it was only an intermediate The Authority of Representation

33

Figure 1.5: Various 1:1,000,000 mapping projects as of 1905, most of which were initiated for military purposes and did not align with Penck’s original grid. French maps are shown with light gray overlay; other shaded areas show US plans for North America, British maps for Africa and Asia, and German maps of China and Korea (inset). Note that only sheets with a diagonal mark had actually been published by 1905. From Albrecht Penck, “Fortschritte in der Herstellung einer Erdkarte in Maßtabe 1:1 000 000,” Zeitschrift der Gesellschaft für Erdkunde zu Berlin (1905); shading added.

step. Immediately after the conference, the newly approved standards were again sent out through diplomatic channels for adoption or comment— this time to every government in the world. National mapping agencies were also invited to produce sample sheets following the standards, the best of which would be used to further refine the specifications. Four years later, a small group of geographers again met at the Geographical Congress, after which a second international conference was organized in late 1913— this time by France, with delegates from thirty-four countries.27 The maps sent to Paris for evaluation were found to differ so noticeably that a French geographer argued, with some satisfaction, that “nothing could better justify the need for a new conference than these divergences.”28 At the conclusion of the conference, the organizers announced that this process would be repeated once again, with another conference scheduled for Berlin. World War I, however, cut this process short, and the 1913 treaty remained the final word on the map.29 It was this process of codification and increasing international participation that finally gave the map its authority. This was not just a question of official endorsement, however; much more important was a subtle shift in the underlying motivation of the project. Penck’s original idea was that an international map would be reliable because its content would inevitably be vetted, cross-checked, and stabilized. For Penck, neither the cartographic details of the map nor its sources of sponsorship were all that important. As leadership of the project shifted from academic geography to practical mapmaking, however, these issues came to be seen as central to the project’s success, and by 1913 34

Chapter One

reliable content was hardly discussed at all. Instead, the main concern shifted squarely to reliable representation— again, in both a visual and political sense. On the side of the visual, the most important change between 1908 and 1913 was that the rules of graphic representation became subject to self-consciously scientific scrutiny. This was not necessarily obvious from the final published standards, which mostly just got longer and more rigid. As shown in figures 1.6 and 1.7, for example, the graphic guidelines doubled in size between 1909 and 1913, and for the most part they regulated seemingly minor details like standardized symbols, colors, and typefaces. But these rules were taken very seriously. They were subject to lengthy debate that often spilled over into the published record, and one of the main results of the 1913 conference was the creation of a permanent Central Bureau (to be located in the UK) that was tasked with encouraging full compliance.30At stake here was not just graphic uniformity, but an ongoing discussion about how to put cartography as a whole on a firm scientific foundation. The relationship between graphic standards and scientific authority is perhaps best seen in the debates about the use of color to represent elevation. At the time, this was a major issue. British and American cartographers in the early 1910s described the task of depicting relief as variously “the most interesting,” “the most difficult,” or “the most urgent” problem in cartography. And the problem of color was “practically a new one,” since color printing was a relatively untested technology and before 1909 no fully colored map at 1:1,000,000 had ever been published.31 Using color to distinguish mountains from low-lying areas was seen as a clear improvement over black-and-white relief, but there was no consensus about which colors to use. Different authors argued that high elevations should be perhaps darker, perhaps lighter, or perhaps “brighter” or “richer”; others urged the use of “regional” schemes that colored mountains brown and plains green.32 The pervasive worry in these debates was that decisions about relief and coloration might be determined by nothing more than popular appeal or individual fancy— criteria which were quite at odds with the ideal of perfectible knowledge. In 1908, for example, the German geographer Max Eckert framed the problem quite expansively as a struggle between art and science. He argued that cartography should be “an art governed and determined by scientific laws,” where the cartographer’s imaginative impulses must be constantly kept in check. “In the choice of colours . . . logic and not personal taste should be the decisive factor.” His great hope was that “the dictates of science will prevent any erratic flight of the imagination and impart to the map a fundamentally objective character in spite of all subjective impulses.”33 Other cartographers espoused a similar view. In 1904, for instance, Charles Close had argued that subjecting cartography to the logic of scientific experiment would eventually lead to universal consensus about “the ideal topographical map”—a single prototype for all mapmaking.34 The process of refining the IMW was exactly of this order, and hopes for its widespread influence were common. In 1912, for example, the well-known The Authority of Representation

35

Figure 1.6: Conventional signs and marginalia approved at the 1909 meeting in London. These original specifications were much more detailed than anything that Penck had proposed; they included symbols and typefaces for various kinds of railroads, roads, rivers, towns, and other features, with explanations in English, French, and German. Published in Resolutions and Proceedings of the International Map Committee Assembled in London, November, 1909 (London: HMSO, 1910).

American geographer Cyrus Adams underscored the importance of reducing “the use of color . . . to a logical system” and hoped that the IMW standards would have a “potent” influence on mapmaking of all kinds.35 The next year, the head of the Survey of Egypt likewise hoped that rigorous coloration would soon displace schemes that pandered to public taste, while the head of the Survey of India was sure that the “influential support” of the IMW committee would “put cartographical relief upon a scientific basis.” The epistemological stakes were clear: “If cartography is to progress, its aim must be truth; it must cease to depend upon license.”36 Figure 1.8 (in the color gallery) shows the color schemes debated in 1909 and 1913. Although the differences may seem minor— the earlier version used 36

Chapter One

Figure 1.7: Conventional signs and marginalia approved at the 1913 meeting in Paris. Besides doubling the number of symbols and typefaces, the new specifications also added a new overview map, information about pronunciation, a scale in nautical miles, and multiple ways of classifying the importance of cities and towns. This multiplication of standards was prompted by the unexpected variations seen in sheets produced using the 1909 specifications. Uniformity required legislating even the smallest details. Published in Carte du Monde au Millionième: Comptes rendus des séances de la deuxième conférence internationale, Paris, décembre 1913 (Paris: Service Géographique, 1914).

purple at high elevations— the 1913 version was understood to be derived directly from the laws of optics, and it was forcefully endorsed by Eckert, Close, and several of their scientifically minded colleagues.37 It originated in the research of an Austrian geomorphologist named Karl Peucker, who in 1898 had published a monograph arguing that the interaction of red and blue light with the lens of the eye actually creates a three-dimensional effect, with red literally The Authority of Representation

37

appearing closer than blue. Although evidence for this theory was rather thin (it was based on a misreading of an early treatise by the famous Dutch physician Willem Einthoven), it was not discredited until the 1920s, and for several decades it remained a powerful argument that low elevations should be drawn in cool colors, starting with a bluish green, and mountains in bright red. Purple mountains, in contrast, would produce jarring effects that could mislead even the trained map user.38 The adoption of Peucker’s scheme in 1913 ensured that sheets of the IMW would not just be visually uniform throughout the world, but entirely trustworthy. And the predictions about the influence of the IMW turned out to be largely correct, with the IMW color system indeed being seen as the default choice for several decades.39 This process was most pronounced for colors, but the same procedure of constructing scientific authority through international debate was followed with other aspects of the map’s design as well, including the map projection, conventional symbols, the spelling of place-names, and even printing techniques. Throughout the debates, scientific cartography was seen as essentially synonymous with an international view: almost by definition, the most objective graphic standards were those that were the most widely endorsed. Although this overall approach was similar to what Penck had pursued, in the end improving trust in the map was not framed as a question of increasing the reliability of survey data, but in designing reliable graphics that would truthfully reveal the underlying geography. Standards of political representation followed a similar trajectory toward rigidity and authority; here the decisive trend was the map’s incremental alignment with the territoriality of national states. The various maps published before 1905 had been produced voluntarily by various military offices, but the overlaps and gaps between different sheets quickly revealed the difficulty of relying on an unregulated division of labor. Penck himself called them “not sheets of one map, but sheets of different maps.”40 In response, the 1909 and 1913 conferences advanced more rigorous plans to divide the work. The first approach simply divided the world between the great powers in large contiguous chunks. The UK, for example, was seen to have a preexisting claim to all maps of Africa, and the hope was that all maps of Europe would likewise be published by one office.41 But this scheme quickly raised problems of both an organizational and a political flavor, since there was no way to ensure that smaller countries would share information about their territory and no easy way to equalize the overall mapping burden.42 The 1913 resolutions thus took a much more strictly state-based approach, with each country expected to produce maps of its own territory. Countries that did not have “a suitable cartographical establishment” would be mapped by those that did, but otherwise any maps that upstaged another country’s work would not be considered part of the project.43 Despite this simple logic of national self-representation, however, actually allocating sheets could still be quite contentious, and negotiations over re38

Chapter One

Figure 1.9: The cartographic partition of Africa in 1913, with mapping responsibility roughly following colonial control. Note that although Egypt would not gain political independence from the UK until 1922, it had its own mapping office and budget separate from the metropole. After these assignments had been made, Italy volunteered to publish the two unclaimed maps of Ethiopia as well— an offer that was received as a “gracious proposition” by the French hosts. This map derived from Carte du Monde au Millionième: Comptes rendus des séances de la deuxième conférence internationale, Paris, décembre 1913 (Paris: Service Géographique, 1914), plate 3; quote on 73.

sponsibility for mapping came to replicate real-world struggles for territorial control and political autonomy. This was most obvious in the case of Africa, which was cleanly divided between the colonial powers in 1913, as shown in figure 1.9.44 But the territoriality of the map tapped into more complex issues of national sovereignty as well. The Chinese government, for example, had initially regarded the IMW with some indifference, telling the British Foreign Office in 1911 that a final decision to adhere to the convention could not be made, as “an office for that purpose was not yet established.” But when both Russia and Japan jumped to claim responsibility for the abandoned Chinese maps, the Chinese military quickly decided to create new cartographic offices in each of its provinces, and in 1913 the Chinese delegate in Paris announced that these offices were explicitly charged with making international map sheets.45 The Authority of Representation

39

A similar logic played out even in Europe, and internal negotiations in France give a good sense that the politics here were driven largely by mapmakers rather than diplomats, with the scientific status of the map presented as key to its political importance. Similar to China, French government officials were initially quite uninterested in international mapping and felt that little was at stake. After the 1909 conference, for example, French cartographers approached the minister of foreign affairs with a British proposal that the IMW sheet covering both London and Paris might be produced by France. The minister’s office responded that it had no interest in the project, and the question of who mapped Paris was of no consequence.46 In response, the most prominent of the French delegates— the famous geographer Paul Vidal de la Blache, founding editor of the Annales de Géographie— wrote to the minister of public instruction with dire warnings about the consequences of this indifference. He warned that failure to participate in the IMW would affect not just France’s scientific reputation but its diplomatic position as well, arguing that the map, “homogeneous in conception and execution,” would inevitably carry some “official authority.” He worried in particular that France might find itself without control over maps of its own colonies. The minister— Gaston Doumergue, who would go on to be both prime minister and president of France— immediately wrote to his fellow cabinet members urging them to embrace the project, since it would “without doubt, be consulted in diplomatic negotiations.” His colleagues’ responses echoed this same language, with the minister of war agreeing that participation would “safeguard our dignity and our interests.”47 The territoriality of the map, in other words, did not follow automatically from its international status. Instead it was the rigor of its representational logic— both scientific and political— that convinced governments of its authority. By the eve of World War I, scientific and political authority were thus thoroughly entwined. Not only did the political process of formal conferences and treaties validate the scientific standards, but it was increasingly clear that a trustworthy scientific map would be just as important for statecraft as it was for academic geography. This relationship was perhaps circular, but it ended up anchoring the map epistemologically in ways that the geographers of the Geographical Congress never could. This is especially apparent in the way that words like reliable and accurate were used to describe the fledgling IMW in the popular press. In 1913, the Times of London explained that the latest conference, “strictly official” in status, had “undoubtedly improved” the map’s graphic standards, since it was “considerably larger and more representative”—politically— than the meeting of 1909. The result was a map “designed not primarily to please the eye . . . but as a thoroughly reliable record of the earth’s surface.”48 The New York Independent likewise hailed the first American sheet— of Boston— as “the beginning of a more accurate map of the United States than any that now exists.”49 Cartographically, these statements were, at best, meaningless, since any map at 1:1,000,000 would almost inevi40

Chapter One

tably be less geometrically accurate than the larger-scale maps produced by national agencies. But as statements about authority, they made a strong claim that official international agreement would inevitably lead to trustworthy cartography, despite the fact that individual sheets would still be produced by each country on its own. The international system of states, in other words, was folded securely into the map’s apparatus of trust. The logical conclusion was that the IMW would in turn be folded into international politics as the privileged manifestation of national territory.

WHAT FLAVOR OF INTERNATIONAL? THE POLITICS OF THE VIEW FROM NOWHERE During negotiations about the map in 1909 and 1913, there was relatively little open debate about exactly how the IMW should depict the world. Besides agreement that it should be a “geographical” map of a “general” flavor, the only explicit discussion of its goals amounted to lists of uncontroversial superlatives: the map should be “scientific,” “authoritative,” “definite, precise, and homogeneous”—and it should be produced by “suitable” mapmaking agencies.50 The overall sense was that the IMW should somehow be neutral, presenting no particular point of view except the universal, objective gaze of the disinterested scientist. But there is no such thing as a neutral map. Every cartographer must make choices about what to show and how to show it, and all maps— whether “scientific” or “biased,” beautiful or ugly— make some features of the world visible while keeping others invisible. Likewise, all maps make assumptions about how the map will be used— and by whom. This is not necessarily sinister; it is simply an unavoidable consequence of representation.51 And indeed, the final IMW standards did in fact codify a particular cartographic point of view with a clear political bent. Graphically, the map tended to reinforce the political split between an international realm and a national realm, with a noticeable disregard for the national. Organizationally, it likewise advanced the political project of civilization on the Euro-American model. The “general” map, in other words, was a map of late colonial foreign relations. The geo-epistemology of representation was thus strongly aligned with a particular territoriality and a particular subjectivity. The territorial stakes are relatively clear: a view of the world that was objective and scientific enough to appeal to national governments was one that reinforced the distinction between domestic and foreign. If this again seems circular, that is exactly the point: not only were assumptions about national territoriality so widely shared that they appeared as neutral features of the world, but the power of representational cartography was that it could make this territoriality seem obvious. The subjectivity of the IMW is somewhat more elusive, since the goal of the map was essentially to construct a “view from nowhere”—a god’s-eye The Authority of Representation

41

view that could be shared by anyone, anywhere.52 But this ideal of neutral detachment was again strongly political, since the implied reader of the IMW would always be looking at the world from elsewhere. Perhaps a German geographer contemplating the landforms of Asia, perhaps a French diplomat drawing lines through the Sahara, or perhaps an American rubber magnate tracing the tributaries of the Amazon— a map of foreign relations was a map for foreigners. At stake here is also the precise meaning of international that animated the project. At the time, this was a relatively unexamined term; it mostly meant that the same graphics should be used everywhere in the world and that responsibility for the map should be shared among all countries. But this is still somewhat vague, since uniformity and universal participation come in many flavors. What matters is therefore how the IMW’s international was implicitly defined against other possibilities. What would an international map show that a national map did not? What did an international audience need to know? Who would decide which countries were capable of making international maps? Graphically, there were three main ways that the design of the IMW defined what it meant to be international. The most important of these was the collection of standardized symbols used to label features on the map. None of these symbols were particularly radical in themselves: railroads were shown in black, roads were red, and icons were available to show towns, lighthouses, glaciers, and so on. What was distinctive, however, was the array of symbols available. Compared to the conventional signs commonly used at the time on national maps, the IMW showed a pronounced bias in favor of systems of international transportation, communication, and boundary delimitation. It was also noticeably not a map of natural resources, economic activity, land tenure, cultural sites, or other subjects of domestic administration. Figure 1.10 shows this comparison in some detail. Especially notable is the lack of IMW patterns for showing land cover, land use, and mineral deposits. (Of all the symbolic patterns used by early twentieth-century cartographers— examples are shown in figure 1.11— the IMW initially only included options for forests and swamps.) To be sure, many of these choices were probably influenced by the scale of the map, since it would be almost impossible to show individual factories, windmills, or schools at a scale of 1:1,000,000. But scale was not itself determinant. For example, the 1884 national atlas of Chile— also at 1:1,000,000— included symbols for salt, saltpeter, and borax deposits, while The Times Atlas of 1895 showed castles, abbeys, and parks on its 1:1,000,000 maps of Great Britain. And in 1911, a Swedish cartographer specifically criticized the lack of IMW patterns for sand, deserts, and prairies. What is perhaps most remarkable, however, is that this critique was almost unique, and there seems to have been no sustained discussion about these decisions. The cartographers of the IMW simply had a shared, tacit understanding about what an international map should show.53 42

Chapter One

Figure 1.10: Comparison of the conventional signs of the IMW with those of other topographic maps of the same era. Items in the left column (symbols peculiar to the IMW) are oriented toward shipping, long-distance communication, and boundaries, while items in the right column (symbols not on the IMW) generally refer to economic activity and cultural sites. IMW symbols from figure 1.7; information on non-IMW symbols from textbooks describing common practice in England, France, Germany, and the United States: Edward Charles Frome, Outline of the Method of Conducting a Trigonometrical Survey, 4th ed. by Charles Warren (London: Lockwood, 1873); John Clayton Tracy, Plane Surveying: A TextBook and Pocket Manual (New York: John Wiley and Sons, 1906); Clarence Osborne Sherrill, Military Topography for the Mobile Forces (Menasha, WI: George Banta, 1910); James Kip Finch, Topographic Maps and Sketch Mapping (New York: John Wiley and Sons, 1920).

Figure 1.11: These kinds of land-cover symbols were common on national topographic maps; they tended to highlight vegetation types that were important for resource planning and off-road travel. The IMW only had such symbols for forests and swamps. From John Clayton Tracy, Plane Surveying: A Text-Book and Pocket Manual (New York: John Wiley and Sons, 1906), 521.

The second way that international was constructed was through the treatment of place-names. In both 1909 and 1913, special committees were formed to determine systems for spelling and transliteration, and here the debate was quite vigorous. At both meetings, a significant minority pushed for a universal phonetic system using the Latin alphabet, so that every place-name in the world could be read in its original pronunciation. Not surprisingly, this quickly proved unworkable; British delegates, for example, delighted at the inability of the committee to even agree on the pronunciation of “Southampton.”54 The only viable solution would be to have naming conventions directly follow sovereignty, with all names on the map being those in legal use in each 44

Chapter One

country’s official language. Countries that did not use the Latin alphabet were urged to publish an official transliteration system, or else one would be created for them. Participating in the IMW thus meant that a country would have to meet a minimum standard of national centralization and standardization of its own place-names, but this was consistently framed as a solution to problems of foreign legibility. This was especially true for countries that did not use the Latin alphabet, where finished IMW sheets would be almost useless for domestic purposes. Maps of China, for example, would follow the same transliteration system that the Chinese post office used for routing mail sent from abroad.55 The final graphic meditation on internationalism again focused on the treatment of elevation. There was unanimous agreement that using Peucker’s greento-red color scale and rigorous contour lines would be the most “scientific” solution, but this interest in science severely limited the potential audience for the map and the ways it could be used, since neither colors nor contours were widely understood by casual map users. In 1912, for example, a British geographer worried that the IMW’s colors would give “ordinary people . . . erroneous impressions,” especially when low-lying deserts were colored green. And although millions of soldiers would be trained to read contour lines in World War I, before the war they were used mostly by academic geographers, railway engineers, and other specialists needing to make precise calculations.56 For the IMW, scientific elevation thus implied rather specific goals, and it was seen as especially useful for the specialist task of international boundary delimitation. At the 1913 conference, questions about undemarcated borders were raised in particular by countries with mountainous frontiers, especially in Latin America. The Dominican Republic proposed special symbols for indefinite or disputed boundaries, and Argentina sponsored a resolution urging South American countries to improve the surveys of their border zones. It was specifically noted that depicting relief with colors and contours would make it immediately clear where further survey work needed to be done.57 Together, these three aspects of the IMW’s design gave a relatively clear sense of what the map could do and who it was for. The implied user was a trained specialist interested in problems of commerce, landforms, or boundaries in areas outside their own country. Note, however, that this was not the only way that international could have been interpreted. For instance, the graphics of the IMW did not show the entire world as one unified territory that might be governed from one central capital, and they did not show a collection of selfcontained states each with its own unique economic, cultural, linguistic, and climatic peculiarities. Instead it showed the world as a collection of foreign countries, each of which made its foreign relations legible to outsiders while simultaneously withholding information about its domestic affairs.58 The alignment of the IMW with a Eurocentric understanding of civilization was more diffuse, but it intersected squarely with the expectation that every country would map its own territory. At its most ambitious, the project The Authority of Representation

45

of civilization was an antinational, even imperialist one, a crusading effort to advance all of humanity along European lines. But the construction and support of self-sufficient nations— even non-European nations— was also an important part of this vision, since national self-determination was the political form furthest from savagery.59 The organization of the IMW reinforced exactly this tension, where international was both European and universal at once. The legitimating rhetoric of civilization had been widely invoked during the development of the project, and it would remain important through the 1940s. In 1904, for example, the leading French cartographer Franz Schrader saw the creation of an international map as a clear justification for European paternalism. Not only was it a crucial step in transforming “our industrial and fragmented civilization . . . into a civilization which is scientific and general,” but the absorption of all parts of the world into the “moral, intellectual, industrial, or commercial orbit of Europe” gave Europeans the urgent duty “to know this earth in its entirety, and in every manifestation of its being.”60 The French minister of war opened the 1913 conference in similarly expansive terms: the goal of the map was not just “truth in Geography”—or even “truth itself”— but to bring “more material well-being, more moral uplift, more progress, and more civilization” to “all mankind.”61 After the war, this universalizing progressivism would even be installed as the very emblem of the project. As shown in figure 1.12, the frontispiece to interwar IMW annual reports showed four cherubs— representing the European, Asian, African, and American Indian peoples— together holding up a banner reading “Carte du Monde Internationale.” Yet only the European cherub is looking confidently at the viewer, the globe in the background is centered on western Europe, and the banner is in French. The connotation seems clear enough. The International Map is a collaboration between all races, and all share the same basic values of geographic enlightenment. But these are European values, and the white races lead the way. The mechanics of the project, however, told a slightly different story. There was no doubt that only civilized peoples would be able to make acceptable maps, and it was likewise assumed that the civilized would end up mapping the uncivilized. Penck, for example, had originally seen geographic knowledge as the province of the “Kulturvölker und zivilisirten Staaten” alone, and for several decades other cartographers echoed same the sentiment— the map would be pursued by the “civilized countries,” “l’univers civilisé,” the “Kulturstaaten.”62 But separating the civilized from the uncivilized turned out to be politically impossible. The official requirement for a “suitable cartographical establishment” was hopelessly vague, and after China’s public rebuff of Japanese and Russian overreach, it was decided that no master list of cartographic agencies could in fact be compiled. (And even if such a list were created, it would “not lend itself . . . to transmission through diplomatic channels.”)63 The boundary between autonomy and backwardness 46

Chapter One

Figure 1.12: Frontispiece to IMW annual reports from 1922 to 1939. Left to right are Asian, European, African, and American Indian cherubs. The artist, Ellis Martin, was the first design superintendent for the British Ordnance Survey. He was hired in 1919 by Charles Close and was responsible for map covers, type design, and publicity. From Carte du Monde au Millionième: Rapport pour 1921 (Southampton: Ordnance Survey, 1922).

was thus a permeable one, and drawing a hard line would only undermine the larger project of universal cultural uplift. But the inability to clearly define which countries were capable of cartographic self-representation was a feature, not a bug. It was only the open invitation that ensured that internationalism would be a truly universal project rather than a self-interested European one. The ideals of objectivity, neutrality, and universalism were thus quite narrowly targeted. The “view from nowhere” was in fact a view from the libraries, war rooms, and corporate offices of the Euro-American sphere— or of those that aspired to a similar mastery of faraway places. The important point here, however, is not simply that the IMW failed to live up to its universal ideals. After all, this much is true of every map that has ever been drawn. Instead, what is remarkable is just how widely shared the IMW’s partial view was at the time of its creation. See again figure 1.2: the great majority of the countries then in existence— countries that together controlled more than 93 percent of the land area of the earth— promised to redraw themselves according to the assumptions of European foreign relations and European civilization. And this would all be done under the banner of geographic truth. The Authority of Representation

47

VISUAL AUTHORITY IN PRACTICE: THE IMW AS A BASE MAP As originally conceived, the highest scientific ambition for the IMW was that it would be used as a universal base map. By providing a trustworthy and secure representation of the “general” features of the world, the hope was that individual fields would then create “special” series for their own purposes. These derivative maps might use the IMW without modification, but they might also use only some of the base data— that is, only some of the printing plates, one for each color— and then layer their own information on top. Not only would this validate the scientific worth of the IMW, but it would also save a great deal of effort, since the “basic” features of the world would only have to be mapped once. Using a single base map would also allow for much easier comparison of data across fields— perhaps even across all fields. At the 1909 conference, the American delegation had proposed that the map be officially recommended as a model for “meteorology, geology, zoology, botany, and other sciences,” and by the 1930s the IMW was indeed being used as base map in several domains. It was used for international aeronautical and archaeological maps starting in the early 1930s, and by the end of the decade several other multilateral IMW-based projects were contemplated as well.64 The larger scientific vision seemed well on track to be realized. But even though the IMW was used as a base map, it was not used because it was actually seen as perfectly suitable to every task— that is, as scientifically neutral. Complaints about its suitability were in fact quite common, especially among map specialists. Instead, the main reason why the IMW was used as a base map was because it made financial sense. This conclusion is more significant than it might seem, since what is at stake here is not just the history of the IMW alone, but the broader relationship between cartography and the ideals of neutrality and objectivity. Were mapmakers so blinded by their commitment to the scientific ideal that they did not see how biased their maps actually were? Over the last twenty-five years, many historians and theorists of cartography have suggested as much, and objectivity is often seen as the founding mythology— or even ideology— of both mapmaking and map use, at least in the modern era.65 But the relationship between maps and scientific authority is not just about belief; the politics of state sponsorship are also crucial. And in a world where mapping was expensive and there was only one international map, that map had a tendency to be, by default, authoritative. Even when the public rhetoric of truth and objectivity collided with the reality of unsuitable symbols and problematic colors, the state-centric IMW still retained its place as the singular representation of the world. Or put another way, the impossibility of the IMW— or any map— being in fact truly universal did not prevent it from being treated in practice as if it were. The IMW was hardly the first base map, but in the early twentieth century the idea was still a relatively new one, and the IMW emerged at exactly the time when the disciplinary and institutional division implied by “general” 48

Chapter One

versus “special” was beginning to become important. Although isolated uses of base map as a term— or its equivalent in other languages— can be found throughout the nineteenth century, it only entered common use in the 1880s in the United States. It was then picked up in the UK in the 1910s, with the French carte de base and the German Grundkarte becoming common in subsequent decades. Its American provenance is hardly surprising, since unlike in most European countries, where domestic mapping was generally centralized in one agency, maps in the US were produced by dozens of different specialized offices, and the maps of various American surveying agencies came to be used as “base maps” by other agencies that did not do their own fieldwork.66 In the twentieth century, a similar division of labor emerged elsewhere. In 1925, for example, Max Eckert was one of the first cartographers to write extensively about the base/thematic distinction. He saw “the perfect base map” as the responsibility of the “practical cartographer,” while special maps were instead the task of the “scholar”; this divide between practical and academic is still largely in force today.67 The IMW, then, was not just another base map among many. It instead represented a radical scaling-up of a new diagram for how geographic knowledge should be organized, both epistemologically and professionally. The acceptance of the IMW as a base map was rather gradual, and it took two different routes. Immediately after the official launch of the IMW, there was not much interest from specialized scholars. In 1910, for example, the International Geological Conference explicitly rejected the scale of 1:1,000,000 as entirely too ambitious for its work.68 It was only twenty years later, when several dozen sheets of the IMW had already been published, that it began to gain appeal. The route of adoption followed for aeronautical charts was entirely official, and the IMW ended up displacing the existing standards of the only other major international mapping project then in existence.69 The other route, which was followed first by historians and archaeologists of the Roman Empire, instead created new international collaborations for projects that would not have otherwise been viable. The consolidation of the IMW with international aeronautical maps was the more significant shift, both intellectually and institutionally. International standards for aeronautical charts had been developed at nearly the same time as the IMW, in part by the same people. The first calls for standardized aviation maps came in 1906; the first workable proposal— written by one of the French delegates of the IMW conference— was circulated in 1911. Finally, after World War I, specifications for internationally standardized charts were included in the 1919 treaty that established a new International Commission for Air Navigation (ICAN); these guidelines were then progressively refined throughout the 1920s and 1930s.70 Although this treaty never included certain key countries (notably the United States and Germany), during the interwar period ICAN was the only permanent intergovernmental agency attempting to standardize aviation. As with the IMW, countries were The Authority of Representation

49

responsible for ICAN maps of their own territory, and standards were maintained by a secretariat in Paris and debated by official delegates from around the world.71 In the 1920s, the maps of the IMW and ICAN were seen as mutually independent efforts responding to fundamentally different cartographic needs. The most obvious difference was that the ICAN treaty called for two different map series: a small-scale map at roughly 1:2,500,000 that would be used for route planning and compass navigation, and a large-scale map at 1:200,000 that pilots would use to navigate by following landmarks on the ground.72 And although ICAN maps borrowed sheet divisions and several symbols from the IMW, the overall visual goal of the maps was entirely different. Especially on the large-scale map, ICAN maps were specifically designed to make unfamiliar terrain recognizable from a low-flying airplane. Railroads, for example, being straight, dark, and generally easy to spot from the air, were given much more weight than roads. Steeples, towers, and other visually distinctive features were likewise highlighted. And instead of showing elevation with the greento-red gradient of the IMW, the ICAN specifications called for solid colors supplemented by hill shading. Figure 1.13 shows an example of such a chart. In 1921 a British cartographer described this logic as creating a “picture of the ground”: unlike a typical topographic map, which included invisible features like administrative boundaries or detailed place-names, aviation maps were intended to “give to objects on the ground the value which the aviator places upon them,” even if the result seemed odd to those accustomed to traditional maps. Yet these maps did not try to mimic aerial photographs, either, since they could depart quite noticeably from the conventions of naturalism. 73 Thus, instead of one authoritative map, aviators required several. And instead of striving for universal objectivity, aeronautical maps were explicitly tailored to the subjectivity of the pilot.74 The ICAN map standards, however, were beset by several problems, and the ideal of a stand-alone series of international aeronautical maps soon had to be abandoned. For the cartographers of the ICAN map committee, the main problem was that the original specifications called for far too much new mapmaking, and the ICAN’s goals surpassed even rich countries’ budgets. As early as 1923, a British delegate argued that “a universal series on a scale of 1:200,000 is an impossibility”: not only would such a series eventually require more than twenty thousand maps, but many countries already had well-established map series at scales close to 1:200,000 that would never be redrawn simply to comply with the ICAN treaty. Indeed, during the 1920s only France and Siam pursued the large-scale series with any determination.75 In addition, the graphic specifications, which had initially been drafted with heavy influence from pilots, were cartographically somewhat undercooked. The French delegate lamented that they were “opening the door to abuses” and that the series was “abandoning all homogeneity.”76 By 1930, the ICAN committee was ready to start from scratch. 50

Chapter One

Figure 1.13 (see gallery for color version): A French aeronautical chart from 1911, scale 1:200,000; detail is shown at actual size. Unlike the IMW, this was described as a “picture of the ground.” This chart predates the ICAN standards but uses much of the same logic: railroads are shown as prominent heavy lines, potential obstructions are highlighted, color is used to distinguish forest from open land, and relief is shown using naturalistic shading. In charts from the 1920s or 1930s, roads— which were often difficult to see from the air— were usually shown much less prominently, with light gray or pale teal, for example. This map from P. Pollacchi, “La carte aéronautique du service géographique de l’armée,” Annales de Géographie 20, no. 112 (1911): plate 18.

The solution, made official by 1932, was to discard the large-scale map altogether and use preexisting maps for two new series at much smaller scales. The IMW would anchor the basic series, and a new series of route planning charts would use the small-scale maps of the International Hydrographic Bureau (originally sponsored by Prince Albert of Monaco). Besides saving the expense of preparing entirely new maps just for the needs of pilots, this also meant that many fewer maps would have to be produced.77 Outside the domain of aviation, this seemed like a simple change that fit perfectly with the larger IMW vision. In 1931, for example, the director of the IMW Central Bureau, which administered the new ICAN mapping, argued that making the International Map useful for aviation required “little more . . . than the addition of certain accepted signs,” and in 1938 the Central Bureau likewise described the modifications seen on finished sheets as “slight.”78 The idea, again, was that the only difference between a “general” map and its “special” progeny was the addition of a bit more information. But for the cartographers on the ICAN committee— thirteen European and one Japanese— the adoption of the IMW was not so straightforward. IMW sheets could certainly be modified to accommodate the needs of aviation, and ICAN issued several resolutions detailing exactly how the change would be accomplished. Extra symbols could be added, hydrographic and elevation information could be simplified, and minor towns and boundaries could be eliminated. But as shown in figure 1.14, these modifications were not in fact trivial— they even eliminated the use of green for low elevations— and the resulting maps were certainly not the “picture of the ground” that cartographers had hoped for. The new maps even challenged the basic distinction between “map” and “chart” that was becoming common in aviation.79 During the 1930s, compliance with the new IMW-based standards was also an ongoing issue. On several occasions ICAN had to ask individual countries to respect the original IMW sheet divisions and symbols regardless of how well they fit the needs of pilots. And in 1936, after Italy had produced all of its new sheets, Italian cartographers complained that staying too close to the IMW had only given “a partial and insufficient view of what the . . . Aeronautical Map should be,” and suggested that more modifications would be necessary. Near the end of the decade, ICAN again battled with the IMW standards when trying to include new information for radionavigation, with several cartographers suggesting that even the modified elevation colors squandered ink on features that pilots did not need.80 The IMW, in other words, was certainly useful, but it was hardly ideal, and aeronautical cartographers did not treat its graphics as neutral. Its main appeal was simply that it was cheaper and easier to maintain one international series than two. The use of IMW as a base map in other fields involved fewer obvious compromises but nevertheless followed a similar logic. The first new project to use the IMW was launched in 1928, when the British archaeologist Osbert Crawford convinced the International Geographical Congress that his Map 52

Chapter One

Figure 1.14 (see gallery for color version): An ICAN aeronautical chart from 1934 that used the IMW as a base map. Much new information has been added that was not included on the IMW itself, and the green-to-red color scale has been modified so that green could be used for forests. Red was likewise used to show urban areas where pilots should not expect safe landings. Although many features were redrawn or added for the benefit of pilots— especially railroads, forests, and obstructions— the use of layered elevation colors and the lack of naturalistic hill shading meant that this map was not the “picture of the ground” that pilots and aviation cartographers preferred. The rigid IMW gridiron also did not align with standard air routes, and almost all commercial pilots would have found the edges of this map to be rather inconvenient. Published by the British War Office.

of Roman Britain at 1:1,000,000, which he had produced for the UK national mapping office, should act as a model for an International Map of the Roman Empire.81 In the mid-1930s the British also began using the IMW as a base for domestic maps of population, magnetism, and gravity, and Charles Close suggested that these projects might be internationalized as well. Finally, in 1937 a German geographer proposed organizing an international oceanographic map, even sending a sample sheet to the Central Bureau for evaluation.82 Only the map of the Roman Empire was successfully organized before World War II, but it gives a good sense of how all these projects would have worked. The basic idea was similar to the ICAN maps: archaeologists used IMW printing plates for rivers, lakes, and relief and then drew new layers to show ancient settlements, roads, and so on. The project had its own official regulations, but publication was again coordinated by the IMW’s Central Bureau, and the use of the IMW was of clear benefit. Fifty-seven sheets would be needed to cover the relevant parts of Europe, North Africa, and the Middle East, and without national survey agencies’ interest in the IMW, the Roman map could never have been justified economically. The map was also touted as authoritative in much the same language as the IMW itself: it was described as an “immense” and thoroughly “scientific” project, the goal of which was no less than “perfection.” By the end of the 1930s, eleven sheets had been published.83 Yet reliance on the IMW also meant that the map of the Roman Empire ended up bearing traces of the modern world. For example, although Latin was used for all place-names and marginalia (the official name of the project was Tabula Imperii Romani), using the blue printing plate from the IMW meant that modern river names also appeared in the Roman Empire, standardized according to modern territorial boundaries. Sheet names also followed modern standards, so that the map showing most of ancient Scotland was titled Edinburgh, even though Edinburgh did not exist until the medieval era.84 These were relatively minor problems, but cartographers noticed them nevertheless and saw them detracting from the proper goals of ancient scholarship. Again, the main appeal of the IMW was not that it was perfectly neutral, but that it already existed. With both the aeronautical and academic projects, these slippages did not prevent the IMW from being adopted as base map, but they make it clear that the project’s claims to scientific authority need to be seen in a broader context. If mapmakers had embraced the IMW because it was seen as an unproblematic stand-in for the physical world, it would indeed be tempting to interpret their commitment as ideological and leave it at that. But given that cartographers often questioned the universality of the IMW (at least in private), the more reasonable conclusion is that its scientific authority was as much a question of necessity as belief. The IMW was used primarily because it was available, and its scientific status was reinforced by its usefulness— not the other way around. The only international map of the world was also the true map of the world, and neutrality was just as much an effect of a certain politics of map 54

Chapter One

production as it was an ideological cause. This conclusion, however, does not diminish the map’s epistemological claims; if anything, it makes them all the more important. During the 1930s, the impossibility of financing a different map series for each different task meant that cartographers’ allegiance to objectivity and neutrality would never be challenged, and the authority of the IMW was widely accepted despite its actual— and recognized— tilt toward national territorial (and purely terrestrial) concerns.

POLITICAL AUTHORITY IN PRACTICE: THE AMBIGUITY OF “PROVISIONAL” MAPPING The political logic of the IMW followed a trajectory that was quite similar to the ideals-versus-reality story of the map’s scientific status. Although the original idea in 1913 was for each cartographically capable country to produce maps of its own territory, in fact more than half of the roughly five hundred IMW maps published before World War II ended up being produced by private organizations rather than national survey agencies. In addition, most of these unofficial maps spread well beyond the territorial limits of the mapmakers’ home countries. But this obvious lapse should not be seen simply as a failure of collaborative internationalism. In line with the larger project of civilization, this mismatch instead points to an important tension at the heart of the IMW— one that was productive precisely because it was unresolved. In theory, the long-term goal of universal political participation should have aligned perfectly with the long-term goal of universally trustworthy cartography, but in practice there was significant slippage between these two objectives. During the 1920s and 1930s, the term that codified this tension was provisional, with an important double meaning. Although it was invoked quite straightforwardly as a way of signaling that a published map should not be considered final— either politically or visually, depending on the circumstance— there were great ambiguities between its two meanings. As a result, it ended up performing important work, not only by honoring the project’s larger political sensibility while nevertheless ignoring it, but even by further reinforcing the close association between Euro-American civilization and domestic territorial knowledge. The creation of a category for “provisional” maps was originally quite separate from any grand concerns with sovereignty or territory, but it eventually became the primary way that international politics intersected with the actual progress of the project. In the original 1913 specifications, allowance for provisional publication formed a relatively minor part of the standards for depicting elevation. The idea was simply that areas of the world “not completely surveyed” could use alternatives to the standard green-to-red color scheme “according to the degree of accuracy of the data available.” Mountains, for example, could be drawn using only shading or hachures, much as they had The Authority of Representation

55

Figure 1.15: Official IMW sheets published before 1939. Fourteen sheets were published before World War I; the rest appeared steadily during the 1920s and 1930s. Of the more than two hundred maps shown here, only fifteen violated the ideal of national self-representation. (See note 87 for sources.)

been on earlier maps.85 Scientifically, this logic echoed an argument made by Penck nearly fifteen years before that the map would provoke a slow progress from “provisional representations” to “definitive” ones. But provisional mapping was also a matter of simple expediency, since it was agreed that in areas like Africa, Asia, or South America, delaying publication for the sake of perfection would be counterproductive.86 Almost immediately after the adoption of the official regulations, however, “provisional” also became the category of choice for sheets produced by the private organizations that were usurping the responsibility of national governments. Figures 1.15 and 1.16 show the coverage of official and unofficial IMW maps before World War II, during which time there were three major nongovernmental initiatives. The first was the huge effort by the (private) Royal Geographical Society of London to prepare more than a hundred sheets of Europe and the Middle East during World War I. These maps were made available to the British War Office, but they were also released to the public as early as 1915. Two similarly huge efforts followed in the 1920s: one was a series of roughly fifty maps of Brazil published by the Engineering Society of Rio 56

Chapter One

Figure 1.16: Coverage of the three unofficial IMW schemes before World War II, plus the plan announced by the Swedish explorer Sven Hedin. In contrast to the scattershot approach of the official mapping, private organizations tended to publish in large contiguous blocks. (See note 87 for sources.)

de Janeiro in 1922; the other was set of 107 maps produced by the American Geographical Society of New York between 1923 and 1945 that showed all of Latin America south of the Rio Grande. (A fourth major unofficial series— a Swedish-German collaboration for fifty-four maps of Central Asia— was announced in the 1930s, but only three sheets appeared before World War II stopped the project.)87 Visually, most of these maps did not seem provisional at all; they gave the impression of final, authoritative representation, with full elevation information and colored layers. The Brazilian maps were especially assertive, since they used hard-edged contours and relief coloring even in areas where little precise surveying had taken place. And although every series took some liberties with the official IMW standards, most of the maps were no less compliant than nationally produced sheets, and some were seen as quite a bit better.88 Yet in most cases the word provisional was printed prominently on every map. The obvious implication was that the provisionality of these sheets was not graphic, but political, since each series violated both the official rules of IMW production and the spirit of internationalism. The British maps, for example, were closely linked to British war strategy: not only were they used for military planning by the War Office, but when they were subsequently used The Authority of Representation

57

Figure 1.17: The archival imperialism of the American Geographical Society. This photo shows the source material assembled in New York for the compilation of a single sheet of the Map of Hispanic America. Existing maps— many unpublished— were collected from national archives, libraries, and private corporations throughout the Western Hemisphere. These were then ranked by reliability before being combined to create the finished map. A four-volume bibliography of sources accompanied the maps; it listed about fifteen thousand items. The only new surveying sponsored by the AGS was for the headwaters of the Amazon in the late 1920s. Image from “The Map of Hispanic America on the Scale of 1:1,000,000,” Geographical Review 36 (Jan 1946): 8.

as the official maps of the 1919 Peace Conference, they were seen as giving the UK a strategic advantage.89 The Brazilian maps, although not an international affront, nevertheless clearly clashed with the official— and public— plans of the Brazilian government.90 In turn, the heroic American effort flirted with outright cartographic imperialism. Making the maps required assembling a massive cartographic archive in New York City, with material collected from dozens of libraries, national archives, and private corporations. (Figure 1.17 shows the material used to create just one sheet of the map.) Nearly every article written by the project’s own personnel likewise emphasized— perhaps a bit too much— that several South American governments had been officially consulted and thoroughly convinced of the “disinterested” nature of the project. The meaning of the “provisional” label was later made explicit: it signaled 58

Chapter One

that the maps were “intended to serve only until the official definitive edition has been produced by the proper Hispanic American government.”91 The immediate response to these projects by other cartographers was almost entirely political, and largely negative. Max Eckert, for example, clearly saw the British maps as a threat to international cooperation; he described them as having a “national tinge,” one which caused the original source maps from Germany and Austria to “suffer.”92 The leader of the British mapping— the unrepentantly nationalist geographer Arthur Hinks— himself admitted that the project was not one which the UK had a “particular right to do, from an international point of view.”93 Hinks was even more extreme when describing the American effort, which was led by the equally nationalist geographer Isaiah Bowman. Hinks called the project a “pirate series . . . undertaken in cold blood” and wagered that if Bowman had “formally asked authority to produce the sheets . . . it could not have been granted.”94 Although the venom of these comments no doubt stemmed from the well-known personal animus between the two men, their tone only underscored that scientific mapping beyond one’s territorial borders was politically illegitimate. Indeed, even Hinks’s less skeptical colleagues used similarly legalistic language. For instance, Harold Winterbotham— stalwart supporter of the IMW and future head of the Ordnance Survey— described the American series as “meritorious piracy” that violated the accepted principle that every country had “the right to map itself.”95 Almost uniformly, these cartographers saw internationalism and nationalism as competing, zero-sum ideals. But the issues raised by the “provisional” category were more complex than these public reactions might suggest, and there were two ways that the dual meaning of the term— simultaneously visual and political— ended up doing important political work. First, the term was in fact found to be quite useful, and it continued to be used despite many attempts to clarify or eliminate it. This ongoing confusion is most easily seen in the IMW annual reports, where the Central Bureau managed to keep “provisional” maps in productive limbo by constantly changing its policy. The first reports in the early 1920s tried to enforce a purely political understanding of provisionality, but this was reversed in 1924 so that “provisional” could only be interpreted as a category referring to the treatment of elevation. In 1928, “provisional” instead began to lump together all noncompliant maps, both official and unofficial. The categories were redefined yet again in 1930 and 1933, seemingly at random.96 Throughout these changes, the official vocabulary used for classification also remained ambiguous, with phrases like “properly international,” “lacking in some respect,” and “the authorized model” being used instead of any clear statement about visual or political status.97 There were several attempts made to fix these problems, but in practice the ambiguities always persisted. The most forceful move was made in 1928, when a committee at the International Geographical Congress took a purely visual interpretation of the term and formally abolished it from the official The Authority of Representation

59

Figure 1.18: Relative reliability diagram (actual size) from a map of southern Mexico published by the American Geographical Society (Istmo de Tehuantepec, 1938). The darker the line or hatch pattern, the more reliable the data. While this diagram might seem like a retreat from the ideal of cartographic perfection, at the time it was seen as a way to provoke new survey missions and therefore hasten the arrival of the perfect map. (A 2009 study found that these diagrams did in fact give a good sense of the map’s accuracy.)

specifications. Instead of asking mapmakers to characterize an entire map as reliable or unreliable, the committee instead recommended that all sheets should include a marginal diagram that showed the relative reliability of different areas of the map. This kind of visual bibliography had first been used on the American maps of Latin America; an example is shown in figure 1.18.98 Yet after this change, ambiguous use of “provisional” persisted. It continued to be used on French and Italian sheets of Africa, on American sheets in Latin America, in unofficial descriptions of the project, and even in public speeches by officers of the Central Bureau.99 In the 1930s, questions about the ambiguous provisionality of Italian sheets of Libya even provoked the aeronautical cartographers of ICAN to make diplomatic inquiries with the government of Egypt.100 There was thus a clear benefit to keeping “provisional” open to interpretation. Not only did it allow illegitimate map publishers a certain level of political cover, but it also allowed the Central Bureau to include all maps as part of the project, regardless of their origin. The second way that provisional mapping carried political weight was through the pattern that governed the use of the term geographically. The imperial overtones of politically provisional mapping were clear enough. But looking again at figure 1.15, note that visual provisionality was likewise a category for the colonies— especially in Africa. Indeed, there are no examples of sovereign states producing graphically provisional maps of their own territory. Instead, autonomous countries with large areas of poorly surveyed terrain simply published no maps at all— in particular the US, Canada, Australia, China, and the USSR (which had withdrawn from the project entirely).101 This 60

Chapter One

distinction thus further reinforced the split between domestic and foreign affairs, since the implication here was that, for already-sovereign states, incomplete geographic knowledge was a purely domestic issue. Only in the colonies did half-baked surveying seem to be of any legitimate international interest. Although this pattern was never explicitly discussed, the unbalanced interest in the affairs of the less-than-civilized was further highlighted by the fact that the unofficial series ultimately got much more use than any of the official editions. The Americans’ map of Hispanic America, for example, was used as the map of record in several international boundary disputes in the 1920s and 1930s (most notably the League of Nations arbitration between Bolivia and Paraguay), and it was requisitioned by the US State Department during World War II.102 More than a decade after the British maps were used for redrawing boundaries during the 1919 Peace Conference, they were likewise renovated and republished in response to continuing demand.103 In 1939 Arthur Hinks noted this tension between international production and international attention quite sharply, seeing the American series as a clear sign that “if you want a general map covering a continent, consistent in style, and available in quantity, you must make it yourself, and whether you call it International or not is a matter of choice, or expediency, or perhaps of chance.”104 In other words, the most pressing need was for the uniform mapping of continents that would not otherwise be mapped. Producing maps of Latin America, Africa, or colonial Asia— or even central Europe— was perhaps illegitimate, but it was also seen as a gracious service to the international community. Ultimately, then, the striking split between official and unofficial mapping was never as problematic as it might at first appear. Indeed, even though most actual mapping before World War II exceeded the limits of sovereignty, the unofficial maps did not in fact disregard sovereignty at all; they were instead a powerful commentary on its importance, and the ideal of the territorial state was still central to the project. After all, there is an important difference between transnational mapping that answers to no one and transnational mapping that wears its illegitimacy on its sleeve. The ongoing controversies and self-serving rhetoric of the American maps of Latin America were thus no less a part of the politics of the project than the actual centralization of knowledge in New York. And the Americans’ maps of the Western Hemisphere— or similar slippages in Africa, Asia, and Europe— arguably did a better job of advancing the grander project of civilization, with all its tensions and contradictions, than properly international maps could ever have alone.

CONCLUSION: INTERNATIONAL MAPPING AND NATIONAL TERRITORY Looking back on the first fifty years of the International Map, the American Geographical Society’s sheets of Latin America give a remarkably good sense The Authority of Representation

61

Figure 1.19: Assembly of the American Geographical Society’s sheets of South America into a segment of an enormous globe. This is cartography literally installed as a monument to civilization. Photograph for Life by Alfred Eisenstaedt, 1941. The inset map in the lower left is a relative-reliability diagram for the entire series.

of the IMW’s place in prewar cartography as a whole. For example, consider figure 1.19, which shows the assembly of thirty-three sheets of the Amazon basin into a substantial section of a 1:1,000,000 globe. This exhibit was first constructed for the 1938 World’s Fair and then later expanded to forty-eight sheets and permanently installed at the organization’s headquarters in upper Manhattan. Figure 1.20 shows an even more ambitious synthesis: the assembly of nearly the entire 107-sheet series as part of a photoshoot for Life magazine that went to press just days before the Pearl Harbor bombing in 1941. The published caption described the map— “the biggest ever produced by a private institution”—as “already . . . guiding technological progress into the wilderness.”105 After the war, a lavish dinner was held to finally celebrate the completion of the Latin American series, and the speakers echoed this same sentiment. The leader of the project, Isaiah Bowman, described it as a thoroughly “objective undertaking” which contributed to “universal humanity.” It was a “scientific map” that would promote comparative scholarship, settle boundary disputes, and lead to the kind of “intelligent government” that had its roots in the writings of Plato.106 In turn the US assistant secretary of state— Spruille Braden— argued that “accurate and authoritative” maps were a requirement for world peace; they were the leading edge of “the forces of civilization” and “a permanent monument to the spirit of man.”107 62

Chapter One

Figure 1.20: Assembly of all of Hispanic America in the courtyard of the American Geographical Society. Far too big to be useful for any practical purpose, this seems to have been simply a publicity stunt. A similar photo (credited to Alfred Eisenstaedt) appeared in Life, 8 Dec 1941, 104; this version from “The Map of Hispanic America on the Scale of 1:1,000,000,” Geographical Review 36 (Jan 1946): opposite 1.

In other words, the IMW sheets of the Amazon were not presented as a useful tool that might actually be used in the Amazon. Instead they were something much grander: a (literal) monument to the civilizing power of systematic geographic knowledge, a miniature version of the earth ready for consultation— or perhaps just contemplation— in the centers of power. Cartography was thus both a means and an end. It extracted geographic fact from the jungles of South America, domesticated it, and replaced the messy chaos of the real world with a smooth paper stand-in. On the eve of World War II, this was geographic knowledge in its most privileged and powerful form. To be sure, the legitimacy of the American mapping effort was at best ambiguous, but the The Authority of Representation

63

tensions between empire and national self-determination were just as central to the civilizing project as was the universal scientific view from nowhere. After all, even Bowman himself argued that the main goal of his quasi-imperial map was to advance the strength and stability of national territorial states. But the exact route between scientific knowledge and intelligent government— that is, between visual representation and political representation—was never very clear, and the benefits of objectivity and universal truth, just like the benefits of civilization as a whole, were taken as self-evident. The geographer Neil Smith has described Bowman’s map as part of the “completion of nineteenth-century geographical business,” one of the last American projects of “exploratory and imperial” conquest before the advent of softer forms of influence. A similar statement could have been made by Albrecht Penck himself: the IMW was about finalizing knowledge once and for all.108 But it is important to remember that in the early 1940s, the larger project was hardly complete. There were still hundreds of maps to be made— in North America and Australia just as much as in Asia and Africa— and entire continents still lacked the light of civilization. Had World War II not intervened, the progress of the 1920s and 1930s might well have continued along its same trajectory, with several sheets being published each year, ever-wider use of the IMW as a less-than-perfect base map, and new private initiatives politely upstaging the work of less-than-suitable governments. The potential sustainability of the interwar IMW, however, should only underscore just how great the contrast would be with the actual postwar world. After the war, not only would the IMW begin to falter on its own terms, but so too would the larger logic of authoritative representation.

64

Chapter One

CH A PT E R T WO

Maps as Tools: Globalism, Regionalism, and the Erosion of Universal Cartography, 1940– 1965

The life of the International Map of the World after 1939 is simultaneously a tale of resounding success and abject failure. Although it was essentially ignored during World War II, after the war geographers reembraced the project with great enthusiasm, and its administration became one of the headline responsibilities of the United Nations’ newly formed cartographic office. As shown in figures 2.1 and 2.2, maps were continually added to the series, and by the mid1980s the only major land areas remaining to be mapped were Greenland and Antarctica. From the point of view of an early twentieth-century geographer, it might seem like the goals of the IMW were almost completely realized: all maps used a consistent scale and roughly uniform graphic conventions, and the 1895 estimate that mapping at the scale of 1:1,000,000 could be realized “in 50, perhaps in 100 years” had proven remarkably prescient.1 But rather than any public celebration of this great accomplishment, in 1986 the UN instead quietly withdrew its support for the project, having determined that “the concept of the International Map of the World . . . appears to be no longer appropriate or feasible.”2 Criticism of the IMW had been mounting since the 1960s; by the mid-1970s, it had become “fashionable” to disparage its basic premise.3 Almost without exception, cartographers and historians looking back from the early twenty-first century have consistently dubbed the IMW a “failed” project, its history a “sad story” of the pitfalls of naive internationalism.4 How could a project that had been so universally embraced before 1960, and had apparently accomplished exactly what it was meant to do, come to be so blithely dismissed as a cautionary tale? The immediate answer given in the 1980s— that the maps were not being kept sufficiently up-to-date— hardly seems to justify such harsh treatment. Instead, I want to suggest that the IMW did not fail on its own terms at all; the more profound problem was a shift in international cartographic norms that made the IMW’s original program 65

Figure 2.1: Number of maps accepted as part of the International Map of the World project, from the first annual report, for 1921, to the last, for 1986. Although there is a clear decline in activity after 1960 (when the American and British militaries began to issue fewer revisions), new maps continued to be added at roughly the same pace. Note that maps were not always recorded in the annual report immediately upon publication, especially during wartime. For raw data, see www.afterthemap.info.

seem increasingly unintelligible. Before World War II, the IMW was a project closely aligned with the ideals of national territory and self-determination. International mapping was a way for a country to fit itself— both politically and graphically— into the universalizing framework of the singular, objective base map. It was precisely these goals that began to be challenged in the 1960s and were largely obsolete by the 1980s. The eventual “failure” of the IMW should therefore be seen as but a symptom of a broader retreat from the entire political and geo-epistemic project of authoritative representation. Politically, international collaboration was upstaged from two directions at once. First was the newly global military mapping of the United States and its allies, which directly challenged the supremacy of the IMW. The most prominent of these projects was the World Aeronautical Chart, a hugely ambitious American series (again at 1:1,000,000) that was begun in 1941 and largely complete by the end of World War II. After the war, this series was adopted by the newly formed International Civil Aviation Organization— and therefore by 66

Chapter Two

Figure 2.2: Index map showing coverage of the International Map of the World as of 1986. Except for Greenland, Antarctica, and five sheets in southern Africa, maps were available for all major land areas of the world. In addition, much of the coverage of the 800+ maps shown here was duplicated by overlapping and irregular sheets. From “Status of Publication of IMW Sheets as at 31 December 1986,” UN Document ST/ESA/SER.D/17, Supplement 3 (Feb 1987). 

dozens of individual countries— as an official international map series. The second challenge was a UN-led push for recasting the IMW itself as a practical part of economic development aid rather than as a more general measure of civilization. This meant that instead of all countries collaborating on a single universal project, the emphasis shifted more squarely to national needs alone. In both cases, the end result was geographic fragmentation and a growing disconnect between international cartography and global space. As international mapping projects came to focus increasingly on the national or regional scale, only military maps retained their global reach. Epistemologically, the decline of universalism was manifest as a parallel shift away from the logic of neutrality toward a much more instrumental sensibility. Rather than evaluating maps in terms of “geographic truth,” many prominent cartographers— especially American and British cartographers who had come of age during World War II— came to see them instead as tools that should be tailored to particular ends. A successful map was one that fulfilled a specific task, not one that presented a universal view from nowhere. These cartographers continued to see cartography as a scientific pursuit capable of objectivity and cumulative progress, but they rejected the kind of worldwide graphic uniformity and scientific generalism championed by the IMW in favor of regional coherence and functional specificity. During the 1950s and 1960s, for example, many projects that had been launched with worldwide ambitions ended up being redesigned to better represent specific regional or national geographies. Even global military maps were no longer designed to be “general” maps that could accommodate any number of uses. Instead, each task required its own custom-designed map. Maps as Tools

67

These political and cartographic changes were closely linked, and by the mid-1960s the intertwined ideals that had made the IMW so worthy of state support in the early twentieth century had unraveled. Territorial states were no longer the building blocks of worldwide geography; the focus was either on representing regions (with their shifting, overlapping, or otherwise indistinct borders) or on national projects that could never add up to a larger whole. And without robust epistemological claims to scientific universalism, there could no longer be any one map— the map— that could reduce the social, ecological, or economic complexity of the world to a single, stable image. Instead of one worldwide space amenable to clean jurisdictional partition, cartography constructed countless overlapping domains, each calling for its own governmental logic. Representation, in other words, no longer offered an organizing framework for unifying space at a global scale. This chapter approaches the fate of the International Map through several cartographic registers at once, including popular, military, and academic mapping in addition to the IMW itself. My analysis is divided into two major parts. I begin by exploring the advent of self-consciously global cartography in journalism and military strategy during World War II, when the IMW was essentially dormant. In both cases, I argue that globalism always implied a regional consciousness that departed sharply from the prewar dichotomy of national and international. Turning to the postwar, I then confront the slow decline of the IMW in three shorter sections. Just after the war, competition between the IMW and the World Aeronautical Chart was felt quite strongly, and the widespread hope was that the International Map could be recuperated and redesigned to regain its position as a universal base map. Over the course of the 1950s, however, international and global cartography began to move in a rather different direction. Analyzing three types of mapping in particular— academic projects that were meant to use the IMW as a base map, the aeronautical map standards of the International Civil Aviation Organization, and the design philosophy of US military cartographers— I find a retreat from political, geographic, and scientific universalism at once. These various developments finally intersected in 1962, when the UN held a major conference to revise the specifications of the IMW. Although these new standards were meant to rejuvenate the project, the result was instead remarkably aligned with the new interest in regional specificity and narrowly instrumental goals. Although international maps continued to be published, the universalist spirit was all but abandoned. Finally, I end the chapter by turning my attention from the history of mapping to cartographic theory. Nearly simultaneous with the cancellation of the IMW in the 1980s, a new “critical paradigm” of scholarship began to transform the way that maps were understood; this literature directly attacked the idea of perfect, objective representation and problematized the relationship between map and territory. Although this new approach is commonly seen as marking a sharp break from the ideals of professional cartography, I want to 68

Chapter Two

situate it instead as a forceful mainstreaming of the antiuniversalist impulse that underlay the demise of the IMW. Highlighting this intellectual continuity underscores how decisive the decline of authoritative representation was in the late twentieth century, but it also opens up important questions about the politics of geographic knowledge in a postrepresentational era.

WHAT IS THE GEOGRAPHY OF GLOBAL WAR? Geographically, the 1940s lead a dual life. Perhaps the foremost cliché about World War II, both during the war and in the decades since, is the idea that it was a “truly global” war in a way somehow different from past global wars.5 The globalism of World War  II has been seen not just as a geographic fact about where battles were fought, but as geographically transformative. The war made isolationism both a physical and a political impossibility: the world came to be seen as geographically smaller, and all countries were eventually integrated into a single global political narrative with clearly defined sides.6 But even as World War II provokes grand visions of a new global order, it is also credited with ushering in an unprecedented emphasis on supranational regions. Regional military theaters and postwar organizations like NATO, the Warsaw Pact, and the European Coal and Steel Community are the obvious examples, but regionalism was an important administrative problem in almost every international organization founded in the 1940s, and in many agencies regions became a crucial policy lever.7 Except for some limited prewar collaboration in the Pan American system, all these efforts were a stark departure from the universalism of the League of Nations— or the IMW. This tension is usually framed in political terms, but it is just as much a question of geographic space itself. Scholars of international relations have long argued that political regionalism need not pose a challenge to organization at a global level— the two can, and do, coexist, and they play different roles in furthering economic integration and military reconciliation.8 Analyzing 1940s cartography, however, makes it clear that the global and regional realms are not just overlapping political spheres; instead, the relationship between them is fundamentally spatial, even topological. With prewar internationalism, countries were like puzzle pieces, each its own distinct and autonomous entity, and countries could usually be added or subtracted from international treaties without disturbing the overall political logic of internationalism. Spatially, this is the world shown in the sheet-layout diagram of the International Map of the World: geographic contiguity was valued but not especially prioritized, and the poles were either treated as special cases or ignored entirely. The world was flat, or at best cylindrical. During and after the war, however, political geography became fully spherical: a curved space that could only be experienced— or managed— as a series of shifting regional horizons. Both geographically and politically, continuity became essential, and the Arctic in Maps as Tools

69

particular took on a new importance. The “truly global” world created during the war was composed of regions, not states. During the 1940s, the IMW itself was almost entirely defunct, and the wartime construction of global space took place through other kinds of mapping. Two genres stand out in particular: the journalistic cartography of American newsmagazines and popular atlases, and the official cartography of the Allied military. These efforts were largely independent at the level of specific individuals and institutions, but commercial artists and military mapmakers were trying to make sense of similar global relationships, and they ended up constructing similar kinds of space. In both cases, regionalism was important both experientially and politically. New technologies— especially intercontinental aviation and mass production— were foregrounded as the drivers of spatial shrinkage and interconnection, but this new space was never legible all at once and could not be subdivided into neatly bordered entities. At the same time, globalism was interpreted as holding special lessons for the United States in particular, and the creation of global knowledge (and global consciousness) was seen as crucial to American interests.

Popular Cartography: Globalism Experienced as Regionalism World War II provoked a remarkable rupture in American popular cartography. In the early twentieth century— including during the first World War— most world maps found in newspapers, magazines, schools, and atlases used the common Mercator projection. As seen in figure 2.3, these maps showed the Americas separated from Europe and Asia by vast oceans and graphically reinforced ideas of isolation and the hemispheric solidarity of the Monroe Doctrine. Starting in the late 1930s, several young commercial artists began to challenge this entrenched worldview with new kinds of journalistic cartography. These new maps were mostly published in newsmagazines like Fortune, Life, and Time rather than in reference atlases or school material, and over the course of the war they became incredibly popular by any measure. They dispensed not only with the Mercator projection, but also with long-held conventions like north being “up” or place-names taking precedence over topography and linework. Figures 2.4 and 2.5 are typical, with bold graphic arrows, pictorial symbols, dramatic representations of relief, and map projections that gave a much more immediate impression of a spherical earth, often with no natural orientation. These maps showed the war taking place in a new kind of space. Instead of the static geopolitics of the pastel-shaded world map, the new emphasis on dynamic flows recast even raw topographic facts as contingent and unfamiliar.9 Historians have mostly interpreted this shift as a strong turn toward a new global consciousness. As Alan Henrikson argued more than forty years ago, this cartography was intimately connected with a much larger geographic project— “air-age globalism”—that amounted to a thorough rethinking of 70

Chapter Two

Figure 2.3: Mercator’s world, as seen from the US. Not only is North America shown separated from Europe and Asia by vast featureless oceans, but the Arctic Ocean is distorted to seem just as large as the Atlantic or the Pacific. In reality the main Arctic basin is only slightly larger than the contiguous United States. From Rand McNally Ready-Reference Atlas of the World (Chicago, 1938).

the geography of American foreign policy. More recently, Susan Schulten has argued for a more widespread transformation of the American “geographical imagination” away from hemispheric parochialism and toward the assertive global presence of a rising superpower.10 But it is important to see this as a very specific type of globalism, one that was fundamentally regionalist in orientation. Not only were particular regions emphasized more than others— especially the North Atlantic, the North Pacific, and the North Pole— but the new maps explicitly highlighted the regional horizon of the intercontinental convoy and the long-range bomber. It was this regionalist bias that most challenged the national subjectivity of earlier cartography. The new regionalism, however, was only ever implicit, and at the time the most overt message was indeed a political-technological vision of American global ascendancy, with aviation in particular being seen as fundamentally globalizing. Politically, probably the most vocal wartime promoter of this idea was the prominent lawyer-turned-politician Wendell Willkie, a strong Maps as Tools

71

Figure 2.4: Map illustrating the strategic importance of Fairbanks. The heavy lines are air routes to major cities. The Arctic Ocean has shrunk, and suddenly Alaska is in a central position. From “Alaska: U.S. Frontier Waits for War,” Life, 19 Jan 1942, 67.

classical liberal and anti-isolationist who spent the time between his failed bid for the US presidency in 1940 and his death in 1944 writing and traveling in support of Allied solidarity as a special envoy for President Roosevelt. In a major speech in October 1942, and then in a book published a few months later, Willkie articulated his famous “one world” concept: his high-speed air journeys had convinced him that “there are no distant points in the world any longer” and that the common need of all peoples for “health, education, freedom, and democracy” could not be solved politically, as the League of Nations had tried to do, but only through tighter global economic and technological integration.11 72

Chapter Two

Figure 2.5: Map of oil and supply routes in the Middle East. The transition between dark and light shading in the western USSR is the current line of German control; the heavy arrows show lines of possible German advance. The thinner lines from the south show Allied supply lines. Note the curved horizon line in the background: the perspective gives the viewer the point of view of the Allies pushing supplies from the south, facing a dark menace from the northwest. From Life, 13 Oct 1941, 44.

Cartographically, the most obvious counterpart to this idea was a new obsession with North Pole–centered maps— known popularly as “air-age” maps— that did a much better job than the Mercator in showing major flight paths between the Americas, Europe, and Asia as straight lines. Although not the first, certainly the most popular of these, shown in figure 2.6, was the “One World, One War” map published in Fortune a few months before Willkie’s 1942 speech. (Willkie ended up using a similar map— figure 2.7— as the frontispiece Maps as Tools

73

Figure 2.6 (see gallery for color version): Fortune’s “One World, One War”—a polar view showing the centrality of the US to the global conflict, connected by supply lines stretching around the world. The Allies were colored a strong red (here shown as medium gray), which made Africa, Asia, and North America into one large bloc facing attack on two fronts. Map by Richard Edes Harrison, from Fortune, Mar 1942; foldout map originally 21 × 27 inches.

to his 1943 book.) Although the Fortune map, unlike later versions by Rand McNally and others, showed maritime supply routes instead of airways, in both cases text and map worked together to argue that US involvement in the war was a geographic imperative. Not only were global transportation routes shorter and more direct than casual readers might have assumed, but these same routes also had the potential to be turned into lines of attack from Japan or Germany. As Fortune put it, air-age maps gave a clear sense “that we [the United States] are not remote from the world’s centers of power . . . but that we are in the center, as truly threatened by external dangers as any state in Europe.”12 These maps were thus seen as a long-overdue corrective to the “evil Mercator,” one better suited to the “the facts of modern history.” At their most hyped, they could even be seen as the final, objective portrayal of the world’s true geography. In the New York Times, for example, Mayor Laguardia com74

Chapter Two

Figure 2.7: Wendell Willkie’s One World: the frontispiece to his 1943 book shows his travels around the world on a polar map. As with the Fortune map, “one world” here is not really an argument about the world, but about the United States’ connection to it.

mented that polar maps show “the world as it actually is,” while The American Magazine ran an article suggesting (incorrectly) that a North Pole map was the best choice for understanding battles in the Pacific.13 Although polar projections were the most conspicuous sign of “one world” thinking, the themes of world connectedness and the vulnerability of the United States were found in many other kinds of wartime maps as well. For example, Richard Edes Harrison— the most influential of the new cartographers and designer of Fortune’s polar map— was best known for his perspective-view maps showing the connections of the United States to major war zones. Over the course of the war he published more than a dozen maps showing approaches to and from Europe, the US, and Japan, as in figure 2.8. Not only did these maps draw immediate attention to the sphericity of the earth— Harrison called his perspective views the “missing link” between map and globe— but they also showed that the US had more to fear from the land bridges of the north than any direct attacks from the east or west.14 Other cartographers sought to make the oceans themselves seem less vast, using Maps as Tools

75

Figure 2.8: Two of Harrison’s perspective views illustrating global connectivity and the experiential irrelevance of north being “up.” The best route between the US and Asia goes via Alaska, and the major axes of the Middle East run northwest-southeast, along either side of the mountains stretching from Turkey to Pakistan. Both maps from Richard Edes Harrison, Look at the World (New York: Knopf, 1944).

heavy graphics of shipping lanes and air routes to emphasize that water was not a barrier but rather a very strong connector. The Atlantic Ocean in particular was rebranded as the “Inland Sea” or “Midland Ocean” of a single world continent.15 The idea of an isolated and geographically coherent “Western Hemisphere”—complete with scare quotes— was subject to especially harsh treatment. Not only did the popular North Pole maps show the closeness of North America to other continents, but hemispherical maps in general came to be designed around the importance of their center rather than the importance of their edges. As a result, the number of possible hemispheres multiplied without bound, since the real, spherical earth knew no natural divisions and could not be split into two obvious halves.16 In more extreme forms of air-age cartography, political borders and even coastlines were seen as obsolete, with the world unified by a single “air ocean” where distance would be measured not in miles, but in “air-hours.” Near the end of the war, American Airlines issued a series of advertising maps that showed only cities: nothing but hundreds of dots and names on a uniform background. From these maps Rand McNally in turn made an “air-age globe” that could be oriented in any direction.17 This hyperglobalism— what Henrikson calls the globalism of the “smooth, seamless ball”18—was nevertheless still grounded in convictions about the power of new technology to transform political relationships. The implication was that the United States, as the presumptive leader in both world politics and world aviation, would establish a new international order where American air carriers could compete anywhere in the world, unifying the globe both spatially and economically. Indeed, at the same time that American Airlines made its maps, US delegates were pushing exactly this position at the 1944 convention which established a new “freedom of the air” in international civil aviation law.19 Even today a lasting reminder of this wartime globalism is retained in the official emblem of the United Nations. Originally designed by the US Office of Strategic Services as a lapel pin for the founding UN conference in San Francisco in April 1945, the UN emblem was derived from a popular air-age map published by Rand McNally a few years earlier (see figure 2.9). In notes prepared later about the history of the emblem, one of the original designers of the pin made explicit reference to the connection between the air-age map and Willkie’s 1942 “one world” speech.20 The message of the UN emblem, in other words, was not just a generic call for global unity, but a very specific, politically liberal idea of tight spatial interconnection, the importance of air transport, and the centrality of the United States in the new world order. Alongside this strongly global discourse, however, the visual argument made by the maps themselves was overwhelmingly regional. Harrison’s maps are again exemplary. For example, when his wartime work was published separately as the Fortune Atlas for World Strategy in 1944, only one world map was included: his original “One World, One War,” which in any case did not include Antarctica. Every other map in the atlas was either a hemispherical Maps as Tools

77

Figure 2.9: The lapel pin designed for the 1945 UN conference (left), and the final emblem as adopted in late 1946 (right). Note that the map on the left is rotated so that North America is oriented vertically, as in Harrison’s “One World” map; also note that Argentina and Chile are clipped. This is the American view of an air-supported global war elevated to a symbol of international unity. Both images from www.un.int.

view or a more focused perspective view of a region of particular strategic importance: the North Atlantic, Europe, the Middle East, the Caribbean, the western Pacific, and so on.21 Often this visual regionalism was in direct tension with the text accompanying the maps, which repeated endlessly that the goal was to show global relationships. Nowhere was it acknowledged that these relationships were always experienced at a subglobal scale. The emphasis on regional, pictorial views was also central to Harrison’s argument about the technological basis of wartime globalism. His extreme exaggeration of mountainous relief, coupled with directional captions— “Northwest to Asia” on a map of Alaska, for instance— placed the reader in the subject position of a pilot, rather than, say, a military strategist. This visual rhetoric thus presented aviation as the key to understanding the conflict and built a strong chain of association between global war, aviation, and a regional outlook. Note, however, that this is not the only way that global aviation could have been mapped. In the 1940s, a professional cartographer (which Harrison was not) would more likely have shown long-range airways by using the kind of recently developed map projection shown in figure 2.10, where the route between any two cities could be shown as a continuous undistorted band. These maps showed the entire world, but they did not emphasize sphericity.22 Harrison explicitly rejected this flat globalism, and most of his maps in fact began as sketches or photographs from a desk globe. But defining globalism as simply a view of a globe must inevitably privilege a regional consciousness, since these views will always be bounded by a hemispherical horizon.23 In the popular press Harrison was hardly alone in his preference for regional cartography. For countless commercial artists and air-age pundits, the 78

Chapter Two

Figure 2.10: This experimental world map was drawn to show air-line relationships between North America, South America, and Australasia; along the horizontal line in the center of the map there is no distortion. Although this would be a good map for understanding Japanese ambitions in the Pacific, popular cartography did not use these kinds of exotic projections. (It is worth noting that the cartographer here, Arthur Hinks, was one of the leaders of IMW mapping in the UK.) From A. R. Hinks, “More World Maps on Oblique Mercator Projections,” Geographical Journal 97 (Jun 1941): facing 354.

global, visually, meant globe, which in turn meant regional views.24 Eventually, regional cartography became a hallmark of nonjournalistic atlases produced after the war as well. For example, one of the major innovations of Rand McNally’s hugely popular Cosmopolitan World Atlas of 1949 was its reordering of maps on a “broad regional basis” and its inclusion of new maps of regions like the North Pole, the Atlantic and Pacific arenas, and the eastern Mediterranean. These were not naturalistic views of a globe, but the visual argument was the same: the “revolutionary changes in world geography and geographic thinking that resulted from the war” were primarily manifest as a new emphasis on regions, and these regions were not discrete groups of countries, but rather areas of linkage between still other regions.25 Even the canonical airage maps of the sort adopted by the UN were essentially regional maps; they just presented the North Atlantic and the Arctic as the only regions that mattered in the new “global” age. Cartographically, the polar distortions of the Mercator were simply exchanged for air-age distortions of the entire Southern Hemisphere. (Note as well that Antarctica is not shown on the UN emblem.) Much of this regionalism can be understood as a confrontation with the inability of any world map to show a full and unbiased view of the globe. The limitations of the new polar maps were well known at the time, and popular explanations of maps routinely stressed that no flat map could ever offer an Maps as Tools

79

undistorted view of a sphere. Harrison was particularly vocal about the impossibility of a single authoritative projection, and he routinely lamented the overenthusiasm for his polar maps. He felt instead that what was necessary was the kind of “geographic sense” that came only by looking at many kinds of maps, with north oriented in many different directions.26 But instead of offering, say, a collection of various world maps, each drawn with a different projection, Harrison and other popular cartographers offered a collection of regional maps. And since the edges of these maps were often defined by the receding horizon of a spherical globe rather than geographic boundaries, conceptually there was no limit to the number of regional views needed to understand the earth. They could— and did— multiply quite prolifically. Regional maps and perspective views of globes were undoubtedly more easily digested by nonspecialists than exotic world maps. The emphasis on these maps, however, suggests that the overall message of air-age globalism was not a straightforward awareness of the indivisibility and simultaneity of all human endeavor, despite the textual rhetoric of Willkie or Fortune. Instead, the global was posited as the sum total of an infinite series of fragmented, partial, regional experiences. The primary lesson of the 1940s denigration of the Mercator projection is likewise not just about Mercator; it was that any synoptic, global view will be unreliable. Universalism, in other words, was seen as inherently distorting. But rather than confronting these distortions directly, wartime maps replaced the familiar disembodied view of a flattened earth with the embedded view of an active participant.

Military Mapping: Globalism as Administrative Regionalism In the wartime cartography pursued by national mapping agencies, the relationship between globalism and regionalism was quite similar to that found in the popular press. World War II brought a new and unprecedented geographic scope to military mapping— dominated both during and after the war by the United States— which was again pursued and interpreted as a global technological imperative. But at the same time, the United States and its allies organized their global project through a new administrative interest in regions. As with popular cartography, official cartographic regionalism did not contradict the global— and the major mapping result of the war was indeed a newly realized global legibility— but the 1940s signaled that worldwide mapping as it was envisaged in the early twentieth century would come to be increasingly impracticable. Much of the military-cartographic globalism of World War II was simply a question of increased map coverage. Four major powers— the US, UK, Germany, and Japan— embarked on massive campaigns to increase their map coverage and overcome what was universally seen as a lack of foresight in preparing for a global war.27 The main German army map (the ambitiously named Weltkarte) was a continuous series of roughly 250 maps that ultimately 80

Chapter Two

reached as far as India and West Africa; Japan’s military maps stretched from western North America to Manchuria to India and Australia; together British and American mapping covered the entire world.28 Amid this huge expansion of map coverage, however, the scale of 1:1,000,000 was particularly important. In all four countries the most extensive map series were compiled at this scale, and design standards generally followed the International Map of the World. Part of this was due to the prior existence of many IMW maps, but the choice of 1:1,000,000 can also be seen as a serious deliberation about the inherent geography of the war. Before 1940, for example, the US Army’s preferred scale for strategic-planning maps was 1:500,000, but given the wartime importance of what the chief map designer at the Army Map Service summarized as “island-hopping, swift invasions, and distance-consuming armored thrusts,” a single map at 1:500,000 was found to be much less useful than a combination of a tactical map at 1:250,000 and a strategic map at 1:1,000,000.29 In other words, cartographers responded to the military geography of the war by zooming in and zooming out at the same time, and the new appreciation of speed, boundary crossing, and land-air coordination meant that there was nowhere in the world that was immune from potential action. The most ambitious wartime mapping project— by far— was the US Army Air Force’s new World Aeronautical Chart (the WAC), a global series at 1:1,000,000 that by the end of the war provided almost total coverage of all land areas in the world in 950 sheets. Like other maps, this series was a direct response to the speed and geographic reach of aviation. Army cartographers began preparing maps of possible battlezones even before the US had entered the war (especially in the Western Hemisphere), but they were soon shocked to discover that only 10 percent of the earth’s surface was covered by the kind of small-scale topographic maps needed by fast-moving pilots. The bombing of Pearl Harbor was an especially rude awakening, since at that time the US had zero aeronautical maps of Japan.30 But even given this level of urgency, it is difficult to exaggerate the departure that the WAC represented from the pace and scope of prewar cartography. In essence, the creation of the WAC amounted to the realization in just under four years of the entire program of the International Map of the World, which after thirty years of steady progress was still only about 40 percent complete.31 In terms of map design, the WAC was remarkably similar to the IMW: it was based on a globally consistent set of graphic conventions (including elevation color coded from green to brown), and it subdivided the world using somewhat arbitrary lines of latitude and longitude rather than physical features or national boundaries. Compare the maps in figure 2.11: although there were important cartographic differences between the two, to the untrained eye the topographic information shown on a WAC sheet looks nearly the same as what one might see on a sheet of the prewar IMW.32 The WAC, however, was not just a repackaging of existing precedent. Its most important feature was a reliance on new production methods that together suggested a radically different basis for international uniformity than the treaties Maps as Tools

81

Figure 2.11 (see gallery for color version): Two maps of southern Mexico. At top is a 1938 map by the American Geographical Society, following IMW standards; at bottom is a World Aeronautical Chart from 1946. The scale in both cases is 1:1,000,000, and the main map areas are each about 17 × 25 inches. Both are primarily topographic maps; their differences are in matters of detail, not conception. (The WAC, for example, shows distances and elevations in imperial rather than metric units, does not show ocean depths, and includes information about radio beacons, airways, and magnetic declination. The IMW sheet shows more small towns and gives greater graphic prominence to railroads. They are also cropped slightly differently.)

of the IMW. Although the Army Air Force made extensive use of preexisting maps— and immediately enrolled the American Geographical Society to assist with sheets of Latin America— the typical methods of ground surveying and careful archival scholarship were entirely inadequate.33 Funding for the WAC therefore included a global crash program of aerial photography, and between the authorization of the WAC in early 1942 and the end of the war, the US Aeronautical Chart Service photographed roughly fifteen million square miles of land— an area larger than the entire British Empire, or about a quarter of all land in the world— to supplement existing map data.34 The key to this program was a new triple-lens aerial-photographic camera known as the trimetrogon that had recently been developed for mapping Alaska. Instead of taking a single photo looking straight down, the trimetrogon took three photographs at once: a vertical photo, plus two oblique photos looking to either side of the plane. Together these three photos captured a swath of land that stretched more than 180 degrees and included two horizon lines. This expanded angular coverage meant that planes could fly higher— high enough to photograph behind enemy lines in relative safety— which also enabled greater spacing between parallel flight paths. Trimetrogon mapping thus sacrificed some measure of accuracy for increased speed: high-altitude photographs were not detailed enough to allow large-scale mapping, but land could be surveyed five times as fast compared with earlier cameras.35 This medium-accuracy, high-speed survey was matched by a similar focus on overall efficiency at home. Photographing the horizon meant that raw photographs could be aligned using simpler, faster methods that required less specialized training, and combined with new assembly-line methods of compilation and drafting the time needed to produce maps was literally orders of magnitude less than earlier methods. At the height of the war, the Aeronautical Chart Service reported that finished maps of eighty thousand square miles in Africa (an area slightly larger than Nebraska) were coming off their printing presses only one week after receiving the raw photographs.36 It was this reengineering of mapmaking from a slow artisanal craft to a high-throughput assembly-line process that most impressed working cartographers. Much of the shock was simply a question of adjusting to a new work environment. At the Army Map Service,37 for example, staff went from roughly 150 before the war to 2,500 at its end, and production was moved from a relatively elegant building designed by the famous architectural firm McKim, Mead & White to a windowless, camouflaged warehouse not unlike those used for manufacturing tanks and bombers. Similarly, in December 1941— just before Pearl Harbor— the Aeronautical Chart Service employed a staff of 20; a month later its core staff increased ninefold and more than 5,000 cartographers and lithographers were subcontracted to work on the WAC.38 Skilled government cartographers who had previously been responsible for seeing a map through all stages of its production, spending roughly six months on one sheet, suddenly found themselves as supervisors overseeing a process that had been subdivided into its most rudimentary steps. The logic was the same Maps as Tools

83

as for any other manufactured good; as an officer at the Army Map Service argued, “It is easier to train ten persons to draw contour lines, ten others to stick down spot names, and ten more to delineate highway classifications than it is to train five persons to handle all three operations.”39 After the war American cartographers were fond of pointing out that the two main mapping agencies of the US military— both almost laughably small before the war— eventually printed and shipped a total of 650 million maps between 1941 and 1945. It was just as common, however, for mapmakers to measure their output in tons.40 These new assembly-line techniques were not just a shift in the labor structure of mapmaking; they also anchored a new system of global legibility. The mass production of military maps went hand in hand with a relentless graphic uniformity: after the war an officer at the Army Map Service noted (in explicitly technological terms) that the Allied obsession with standardization was often carried “to a point approaching stereotyping”—stereotyping being a common method of mechanical reproduction.41 This was a universalism of authorship and readership that both exceeded and subverted the precedent set by the IMW. The prewar IMW was meant to be completed by a Leviathan of scientists for the sake of civilization as a whole: the project was subdivided geographically, and individual agencies— in turn individual mapmakers— were asked to suppress their personal preferences in order to address a fictional, disembodied reader.42 American mass production was a Leviathan of a rather more corporate variety, with production subdivided by task rather than place, singular authorship ensured by a rigid management structure rather than the moral qualities of the individual, and the universal reader given a concrete identity. This reader was the American soldier— who might be deployed anywhere in the world, had little to no prior experience with anything other than road maps, and would only receive about a week of map instruction during basic training. Instead of the IMW project of constructing a global space that relied on, and rhetorically supported, the international system of territorial states, the global legibility of mass production was inherently centralized and specifically designed to ignore the existing bounds of sovereignty. The mutual reinforcement between assembly-line process, graphic uniformity, and global reach was a particularly American phenomenon. The German Army, for example, tended to simply reprint captured maps and did not seem to be bothered by graphic inconsistencies across sheets, even in the same series. In a review of German wartime mapping, an American map officer noted that variations of color and symbolization on different sheets of the IMW-based Weltkarte were “considerable.”43 The Japanese also preferred reprinting foreign maps to recompiling them, usually adding just a simple overprint with transliterated place-names. Japanese aeronautical charts were often specifically designed for individual operations, with scales and symbols different from one theater to the next, and with different charts produced for the air forces of both the army and the navy.44 This kind of cartographic pluralism was precluded by the new American production methods, where unskilled operators were trained to 84

Chapter Two

apply the same rules of photointerpretation and symbolization to all areas of the world. For production of the WAC, military subcontractors were supplied with a standard specification booklet containing one only sample map.45 Although the totalizing nature of US military globalism would slowly be modified in the decades after the war, mass production and global preparedness became permanent parts of American military mapping. The Army Map Service and Aeronautical Chart Service essentially never demobilized after 1945. Trimetrogon photography continued at a pace of about 225,000 square miles a month— an area slightly larger than France— with an increasing share devoted to revisions rather than new coverage.46 Wartime map series were expanded and made ever more uniform; even series that were not meant to provide global coverage, such as the army’s 1:250,000, were nevertheless produced with uniform specifications that applied the same symbolization to all parts of the world. (Foreign road types, for example, were consistently reclassified after the war using American categories.) Dissatisfied by the wartime practice of printing maps using portable field presses, the United States also showed increasing preference for keeping massive reserve stocks and excess printing capacity. At the outbreak of the Korean War in 1950, the Army Map Service had more than two million maps of South Korea already in stock; four weeks later, ten million additional maps had been printed and shipped overseas.47 Yet despite this overwhelming American globalism, both during and after the war the Allied mapping effort as a whole nevertheless relied on a specific kind of administrative regionalism— in particular, the regional division of mapping responsibility between American and Commonwealth countries. And notwithstanding their somewhat unglamorous character, these administrative arrangements arguably made a more profound argument about territory than the global graphics of the WAC, since over the course of the 1940s these arrangements progressively dismantled the traditional relationship between mapping and the bounded space of national territory— or indeed any bounded geography at all. The most important trend in Allied mapping agreements was that geographic coherence became increasingly less important. The first Allied division of cartographic responsibility, signed in May 1942, split the world into two large areas; within these areas the US and the UK would each assume “complete responsibility for the production and supply of maps and survey data” for Allied troops of any nationality. The British zone included all of Africa and Eurasia, while the United States took the rest, including Japan and the East Indies.48 As the ongoing war strained British mapping resources, however, this agreement was soon amended to give more responsibility to the Americans. These amendments were made in a somewhat piecemeal fashion, and the original goal of having all maps for a given area prepared by the same agency was almost immediately discarded. In March 1943, for example, the Americans assumed responsibility for the Iberian Peninsula and northwest Africa, but also for maps at specific scales for Germany, France, and the Netherlands. In Asia and Africa, responsibility was reapportioned both by scale Maps as Tools

85

Figure 2.12: The regionalist division of Allied mapping in 1947, showing different areas of responsibility for military-topographic maps, aeronautical charts, and storage of reproduction material. (The last map shows Commonwealth countries only; the US and the UK held material for the entire world.) Note that each country’s area of responsibility was quite different in each case, and nowhere do the edges of mapping responsibility align with political boundaries. From PRO, WO 163/487; shading added.

and by straight lines of latitude and longitude— in Africa, for example, the US took responsibility for small-scale maps west of 6° east, while the UK retained responsibility for all other maps.49 Instead of a clean geographic division of the world, Allied mapping came to be one large global effort, with responsibility for certain tasks apportioned based on available resources. These responsibilities were regional, but fragmented. Although this new pragmatism was the result of emergency conditions, after the war the trend away from territorial coherence was only magnified. At a secret Allied mapping conference in 1947, global mapping responsibility was divided once again— this time between the US, the UK, Canada, South Africa, Australia, and New Zealand. Different agreements were drafted for topographic maps, for aeronautical charts, and for holding archives of mapreproduction material; the lines of responsibility were different in each case, and they aligned not with political or natural boundaries but with the edges of map sheets. For example, see the treatment of the Middle East in figure 2.12: topographic maps would be prepared by the UK, and aeronautical charts by 86

Chapter Two

the US, while original reproduction material would be held by the US, the UK, South Africa, and Australia.50 Later agreements— updated roughly annually starting in 1950— continued to apportion responsibility by map sheet rather than territory. By the end of the 1950s, areas of responsibility were no longer necessarily contiguous; in some cases they overlapped.51 The crucial point here is that although these various agreements were clearly the product of the new military globalism of World War II, at the same time their regional basis reveals a rethinking of the relationship between mapping and sovereignty more profound than the “smooth, seamless ball” of a project like the WAC. There was nothing novel about apportioning responsibility for a global mapping project by sheet lines— this, after all, was exactly how the IMW was planned. But with the IMW, sheet-based divisions were treated as new kind of all-or-nothing territory, pixelated but still coherent: if Egypt was responsible for an IMW sheet that overlapped with Libyan territory, it was also responsible for international aeronautical charts and other thematic maps of the same area. In contrast, the regions that organized Allied military-cartographic activity did not perfectly subdivide the world into well-bounded units. There were slippages between scales and functions, and regional boundaries changed fluidly over time. Despite the obvious differences between official topographic mapping and newsmagazine infographics, the implicit argument of this administrative regionalism is rather similar to the regionalism constructed by Harrison’s oblique views in Fortune or air-age maps centered on the pole. In both cases, wartime regions were not defined as contiguous expanses of land with clear borders that together would perfectly cover the world. Likewise, there was no implication that regional space was a symmetric foil for the global, the way that national and international each implied the other and yet were in opposition. Instead, the regionalism of both Harrison and Allied mapmakers was perfectly identical with the global: thinking globally meant thinking regionally, and global collaboration meant regional collaboration. In both cases, this new globalism was a sharp departure from cartography before the war. This thematic overlap between popular and military cartography is important for two reasons. First, the idea that there was a relatively broad, coherent shift in the nature of (American) global geography in the 1940s is useful analytically for understanding globalism and the fate of the IMW after 1950. The long, slow dissolution of the project was not foreseen by cartographers at the end of the 1940s, and the difficulties it faced were quite diffuse. Much as during the war, these difficulties were more a question of a changing sensibility about global legibility— in many registers at once— than the actions of any crucial individuals or agencies. Second, historians of cartography have mostly framed mid-twentieth-century mapping as a self-contained moment of experimentation and enthusiasm that dropped off quickly in 1945 and slowly faded over the next decade.52 But even though enthusiasm for the imagined air age did indeed wane, the issues that popular wartime cartography raised— about 88

Chapter Two

the limits of cartographic universalism, the subject position of the pilot, and differential attention to certain key regions— would become central to the actual air-age mapping pursued by official mapping agencies. Questions of universalism and regionalism would not just change how global projects were conceived and administered, but would also call into question the privileged relationship between cartography and global legibility altogether.

DOUBLING DOWN ON THE UNIVERSAL BASE MAP: THE IMW AND THE WAC Although World War II saw the great expansion of mapping activity of all kinds— including several series at the scale of 1:1,000,000 that roughly followed the precedents of the International Map of the World— during the war the IMW itself received almost no attention. The project’s Central Bureau in Southampton ceased functioning after 1939, and in 1941 all its records were lost in a German bombing raid. Of the forty-one countries that had adhered to the IMW convention before the war, only four— India, Australia, Ireland, and the Sudan— kept their dues up to date throughout the conflict. After the war it was unclear whether the International Map was viable at all.53 When members of the International Geographical Union (the IGU, successor to the International Geographical Congress) first began reconsidering the IMW in late 1948, it was especially clear that the existence of the World Aeronautical Chart presented a profound challenge to the project. Did the world really need two map series at the same scale, using similar graphic conventions? The view that prevailed at the time was a resounding yes. Crucial to this conclusion was the idea that the WAC and IMW were actually more different than they seemed. The WAC was a map for pilots. The IMW, in contrast, was a base map, and finishing it was a scientific imperative. Yet interest in the IMW as a base map— which was stronger than it had ever been before the war— generally did not reflect any ongoing success of the IMW plan; it was instead a response to the increasing difficulty of this ideal. Renewed interest in the IMW as a base map came in response to two unrelated developments of the immediate postwar era. First was the transition of the WAC from an American wartime series to an international civilian project. Second were proposals for a handful of new international statistical and thematic mapping projects. This multisited profusion of new mapping efforts not only threatened to lead to redundancy and wasted resources, but it also presented a serious affront to academic geographers who sought general oversight of all mapping. Epistemological retrenchment around the ideal of a “general” map was the result. The existence of the Americans’ WAC might not alone have called into question the viability of the IMW; the real problem was that, even before the end of the war, the WAC became the template for the primary aeronautical Maps as Tools

89

chart series of the new International Civil Aviation Organization (ICAO). Created as the result of the month-long aviation conference held in Chicago in late 1944, ICAO was the successor to the interwar ICAN and the first regulatory body in aviation to claim global authority; it was also one of the first of the agencies that would later coalesce into the UN system. Its mandate, although initially quite contested, quickly became that of setting the kind of technical standards that would promote a free and open market in international aviation— standards like pilot licensing, search-and-rescue agreements, and aeronautical charts. Because viable standards were needed as quickly as possible, at the Chicago conference the small subcommittee on map design had only one option to consider: an American offer to make the original reproduction material for the WAC available to other countries. Although the US was not proposing to discontinue its own WAC work, the US effectively asked its allies to adopt American-designed maps as their own.54 The chart specifications approved in Chicago were copied almost verbatim from the Army Air Force’s own documents, and no attempts were made to harmonize ICAO standards with any other map series.55 This newly international WAC— officially the “World Aeronautical Chart ICAO 1:1,000,000”—presented two challenges to the IMW. First, the IMW and the International WAC were visually similar without being at all coordinated. For example, the fact that they were drawn using slightly different map projections made no difference for most users, but it did mean that printing plates could not be shared. Each series also used a slightly different set of symbols and prioritized a different visual hierarchy: the WAC showed aeronautical installations and highlighted features that were easy to spot from the air, while the IMW emphasized administrative and economic relationships. As seen in figure 2.13, the two series’ sheet layouts were likewise just different enough to be inconvenient. Second, the International WAC was administered much as the IMW had been, but with greater success. The publication of both series depended on the initiative of national mapping agencies, but ICAO members were officially obliged to produce and maintain ICAO-branded charts— of which the WAC was the “basic” series— as a condition of membership. For countries short of funds or expertise, it was therefore difficult to justify producing the IMW when the WAC might do just as well, despite its inclusion of features that might be distracting for nonpilots.56 The new interest in statistical and thematic mapping was also a legacy of the war, as such maps had proven quite useful both for wartime resource planning and for military strategy. Proposals for international civilian projects were presented at the very first postwar meeting of the International Geographical Union, held in Lisbon in 1949. Senior geographers from the UK and the US floated ideas for globally coordinated maps of population distribution and land use, while a well-known French academic called for new international maps for geology, precipitation, soil, agriculture, and vegetation.57 All of these series would be produced by overprinting new layers on existing 90

Chapter Two

Figure 2.13: Sheets of the International Map of the World (top) and the US World Aeronautical Chart (bottom). Shaded areas indicate maps that cover the same extent in both series. (But note that water areas were not mapped in either case.) The IMW grid was uniform throughout the world, although combining sheets in latitudes greater than 60° was permitted. The WAC grid was designed so that sheets generally aligned with land areas and covered roughly the same amount of area at all latitudes (the columns to the right show the relative sizes of printed sheets at different latitudes). The 1949 ICAO specifications for the WAC differed slightly from the American precedent in order to align better with existing national map series; this made for more IMW/WAC overlaps in Australia, India, and Southeast Asia. WAC lines from November 1943 (original in PRO, BT 249/2).

maps. The obvious choice was between the WAC and the (still-defunct) IMW, and this was a case where the radio and airway information of the WAC could be a serious problem. In response, the IGU immediately formed a committee of six eminent cartographers to reevaluate the IMW, especially in comparison with the International WAC and “with a view to its use as a base map for all geographical purposes.”58 In the discussions that followed over the next few years, it became clear that ideas about base mapping and the competition between the WAC and the IMW tapped into deeply held convictions about how mapping should be organized, both practically and politically. The strongest statement in support of maintaining the IMW came from one Louis Hurault, the director of the French national mapping office and a member of the committee. The graphic conventions of IMW, he suggested, made it a “geographical map of a completely general character.” In contrast, he saw the International WAC as a “special” series that depicted “only what is absolutely necessary for aviation, and in a form particularly adapted to this object.” For Hurault, this was a stance of principle, and these two categories were a priori incompatible; he argued that “a single map would not be able to have simultaneously a general and a special character without losing all its clarity.” These were the only two options: either a map is general, or it is specific. And more to the point, no proliferation of specific maps could ever make up for the lack of a general series.59 For the committee, this cartographic separation of general and specific was at the same time an argument about how a newly reconstituted IMW would be administered. In particular, the IGU committee had to consider whether administration of the map should be lodged at the UN headquarters in New York or with UNESCO in Paris— or even with ICAO or the American Geographical Society. The preference for the UN was framed in direct analogy with the base map. Lodging the IMW in New York made perfect sense, since the relationship between the UN and specialized agencies like ICAO or UNESCO exactly mirrored the potential relationship between the IMW and the various proposals for new thematic-mapping projects. This hierarchy was simultaneously embraced by the UN itself, which in 1949 saw its cartographic mandate as nothing less than facilitating a new dawn of “base maps for world needs.”60 This was an embrace of universalism that rather exceeded prewar precedent. After the First World War, the IMW had remained separate from the League of Nations, and territory that was not controlled by a signatory to the IMW treaty was, administratively, out of bounds. But with the IMW at the UN, it would now have a much more explicit connection to global governance and bureaucratic centralization. Naturally, the vision of Hurault and his fellow committee members— that the IMW should absolutely be continued, and that the WAC did not change its worth one whit— was not without its problems, but these were approached only as technical problems in need of technical solutions. Hurault himself admitted that the map projection used for the IMW would no longer suffice 92

Chapter Two

for aviators using radio services, and when Hurault’s comments were sent to the heads of several dozen national mapping agencies for comment, many officials— especially in the US and Commonwealth countries— argued that trying to maintain two separate global map series was, economically, simply a pipe dream.61 But even these detractors agreed in principle that the IMW and the International WAC did not duplicate each other. In the three years between the 1949 conference and the publication of its final report in 1952, the IGU commission did not try to resolve the basic tension between the IMW and the International WAC. Its final recommendation was simply to maintain the status quo— with the IMW and the WAC as competitors— until the time when an official international conference might be convened to incorporate both series into a single international scheme. Although such a conference would not be organized for another ten years, the implication in 1952 was clear enough: what was needed was a modification of IMW standards so that it could accommodate the new needs of aviation. The IMW would have to live up to its universalist ideals— and then some— so that it could again serve as the master map of the world. It would have to be geographically global in a way that it had not been previously, and it would have to accommodate a wider range of users. By the time of the 1962 conference, however, the IMW would be the only mapping project still tied to these goals.

THE 1950S RETREAT FROM THE GOD’S- EYE VIEW Over the course of the 1950s, interest in the IMW itself was relatively robust. The UN did indeed begin administering the project in 1953, annual reports were resumed, and hundreds of IMW-style maps, mostly prepared by the US and the UK as offshoots of military work, were added to the official roster. But at the same time, all of the projects associated with the IMW—those that were meant to use it as a base map— ending up shifting direction quite noticeably. Many abandoned their global ambitions in favor of regional (or even national) coherence, while the larger commitment to cartographic generalism was replaced by a growing interest in matching a map one to one with a specific purpose. The slow decline of the base-map ideal was thus not a question of waning enthusiasm for one particular project; instead it was the consequence of a broader reconfiguration of international cartography, both politically and visually. The earlier relationship between cartography and sovereignty— where the visual authority of the singular, neutral map went hand in hand with the political authority of each country mapping its own territory— simply no longer made sense. Cartographers (in many fields and many countries) found that having a universal base map did not actually further their specialist goals, and after the military mapping of the war it was difficult for any country to claim the exclusive right to map itself. Analytically, this can be seen as simulMaps as Tools

93

taneously a question of authorship and readership: who should produce the maps, but also who they should be produced for. Civilian initiatives— both academic and ICAO mapping— ceased to operate as Leviathan-like collaborations, while the vanguard of practical map design— by the US military in particular, but also again at ICAO— abandoned the “general” map user as an unhelpful fiction.

The Death of the Universal Author: The Civilian Retreat from Worldwide Mapping Central to the international base-map ideal was the belief that the mapping of the world should be pursued as a large-scale collaborative enterprise, with maps of different areas— each produced by different national agencies or scholarly societies— appearing as if drawn by a single hand. After 1950, this kind of internationalism withered in several registers at once. Worldwide map coverage, universal standards, and multilateral decision making were all discarded in favor of more immediately workable results, and by the early 1960s it could no longer be assumed that a universal, collaborative intelligence could (or would want to) apply the same rules of representation to all parts of the world. This trend was clearest in international projects for thematic mapping, but it was also important for the aeronautical chart standards of ICAO. Of the half-dozen or so proposals raised at the 1949 IGU meeting, four projects were actually pursued: a World Land Use Survey, a World Population Map, a Carte Internationale du Tapis Végétal (“The International Map of the Vegetation and Ecological Conditions”), and the International Map of the Roman Empire. Over the course of the 1950s and 1960s, all four projects faced similar problems of administration and graphic uniformity, and although a large number of maps were published in these series, by dozens of countries, almost none followed the projects’ original designs and only a very few actually used the IMW— or any single map series— as a base map.62 The faltering of these projects was not due primarily to logistical problems like lack of funds or suitable IMW material; instead geographers themselves slowly abandoned their universalist goals. The issue was not just that internationally coordinated thematic mapping was inexpedient (although it certainly was); just as important was the new consensus that worldwide projects were not actually useful. The first problem was that international cooperation simply proved to be unworkable, and by 1960 all four IGU schemes had been reorganized either as entirely uncollaborative or as just a loose coordination between national initiatives. In 1954, for example, the original instigator of the International Map of the Roman Empire, the British archaeologist Osbert Crawford, declared that the project as originally conceived in the late 1920s was “completely and irrevocably dead”: archaeological collaboration across borders tended to provoke 94

Chapter Two

“absurd national jealousies,” and national interest in publishing IMW sheets rarely aligned with interest in the Roman Empire. By the end of the decade, pan-European planning meetings for the project had been discontinued. 63 The IGU committees for the World Land Use Survey and the World Population Map both abandoned the goal of globally consistent mapping after only a few meetings— in 1956 and 1958, respectively— once faced with the impossibility of dictating national policy. In both cases their IGU commissions instead became simple clearinghouses with the more modest goal of issuing nonbinding recommendations and encouraging national land-use and population maps of any kind, at any scale.64 And although twenty-five maps of the Carte Internationale du Tapis Végétal were eventually published to uniform standards, covering areas in ten countries, the scheme as realized was essentially not international at all. Starting in 1961, research and publication of the maps were almost entirely funded by the French Institute at Pondicherry, India, and all but four were personally supervised by the project’s original founder— the French botanical geographer Henri Gaussen— and two of his close colleagues.65 The political and administrative difficulties that led to the deinternationalization of these projects were consistently framed in terms of authorship. Conventional wisdom came to see successful projects as those which had single, identifiable authors, and it was assumed that logistical problems would be solved not through more cooperation and coordination, but less. Crawford, for example, suggested that what was needed for the Roman Empire was “one person” to oversee operations at “a single institution which can employ draughtsmen and has the money to pay for the printing of the maps.”66 The presence or absence of a single individual in charge of a wellfunded staff was used to explain the failure or success of the other projects as well: several reviews of the land-use and population projects lamented the lack of a centralized secretariat with adequate funding to pursue original surveys or publication, and the progress of the Tapis Végétal was directly attributed to the efforts of Gaussen, the “master” of vegetation mapping.67 In the 1920s, negotiation and international debate had been valued as ends in themselves; in the 1950s the goal was simply to make the maps. These logistical changes went hand in hand with a thorough reconsideration of each project’s scientific goals. Over the course of the 1950s, precisely the kind of local graphic variation that had been the scourge of international cartography before the war became instead the clearest mark of responsible mapmaking, and the recognition of “local conditions” became a cartographic shibboleth. The issue here was not just that an international design committee could never anticipate all the special cases encountered in the field; more important, it was argued that no color palette or set of symbols could be simultaneously universal, legible, and helpful. This was a not a problem of coordination— it was instead a basic problem of cartography. Maps as Tools

95

Graphically, all four projects again faced similar questions. With the Tapis Végétal, for example, prototype maps prepared of southern France and Tunisia together exhausted the entire color spectrum, and it was unclear how one color system could both provide worldwide coverage and fine gradations between different types of plants.68 The original specifications for the World Land Use Survey likewise tied each of its nine major land-use categories to specific colors; the inevitable discovery of overlapping and hybrid land uses threatened to render finished maps a visual mess of hatch patterns and dots.69 The problems facing the World Population Map stemmed in turn from the great unevenness of rural density throughout the world: if population was to be shown by small dots each representing a fixed number of people, as geographers agreed it should be, how many people should be symbolized by one dot? Figure 2.14 shows this dilemma: in Europe, any less than two hundred would make individual dots merge into a single solid fill, while in the western United States having more than fifty people per dot would obscure desert settlement patterns.70 Even with the Roman Empire, Crawford argued that the goal of universal symbols was inappropriate, since “social and geographical conditions were entirely different” in different parts of the Mediterranean.71 These were not minor questions: showing the flora of southern India as a single expanse of purple, or northern Canada as entirely uninhabited, would render individual maps completely useless for local needs. Solutions to these problems took two forms. Standards for the World Population Map and the World Land Use Survey both became loose templates that could be modified or “amplified” as needed, meaning that rigid mapping standards for color, symbols, and even scale were downplayed in favor of having a few common categories which would allow for rough statistical comparisons. Of the dozens of countries that undertook land-use mapping in the 1950s and 1960s, only Cyprus, Iraq, and the Sudan used the original international classification scheme; other countries’ schemes varied widely, as shown in figure 2.15. For population mapping, IGU-approved graphics were likewise honored mostly in the breach, and it was rare for any two neighboring countries to use the same standards.72 The second strategy, used for the Tapis Végétal and the Roman Empire, was simply to pursue regional coordination and leave it at that. Although Gaussen did propose a worldwide color scheme for vegetation mapping at a meeting of experts in 1960, his colleagues decided that a plurality of regional schemes would be more useful. Later vegetation maps were all confined to areas in the tropics, and all used a color scheme designed specifically for South Asia.73 Finally, while maps of the Roman Empire retained their uniformity through the 1960s, later sheets became nothing but national projects, with late twentieth-century maps of Israel, for example, looking nothing like those of Spain.74 This fragmentation of thematic mapping certainly shows growing disillusionment with scientific internationalism, but there was also a broader shift here in how mapmakers understood global space. In the past, the fact that 96

Chapter Two

Figure 2.14: World Population Maps of Tuscany (top) and Ventura County, California (bottom), both shown at actual size (scale 1:1,000,000). The map of Tuscany symbolizes rural population with one dot for 200 people; in California, one dot shows 50 people. Urban areas are also treated differently, including very different formulas for scaling the size of urban population circles. At the time these maps were made, for example, Florence was more than seven times as populous as Oxnard, but both cities are shown with circles of roughly the same size. Thus despite the shared use of dots and circles, it is actually quite difficult to compare the two regions. Maps from W. William-Olsson, “The Commission on a World Population Map: History, Activities and Recommendations,” Geografiska Annaler 45 (1963): 247; Norman J. W. Thrower, “Anglo-American Population Map on the Scale of 1 to 1 Million,” Geografiska Annaler 45 (1963): 281.

Figure 2.15 (see gallery for color version): World Land Use Survey maps from central Italy (top) and western Japan (bottom), both reproduced at actual size. The maps use different scales— 1:200,000 and 1:50,000, respectively— and different classification and color schemes. (The Japanese map, for example, uses eight categories for urban land, while the Italian map uses only one.) The Italian map is part of a series covering the entire country; the Japanese map was a prefectural map for “a specific development project.” As with population mapping, comparing land-use patterns between the two areas is tedious at best. From Hans H. Boesch, “The World Land Use Survey,” Internationales Jahrbuch für Kartographie 8 (1968): facing 141.

maps of Finland or Belgium had used a different logic for classifying roads or labeling towns was seen as a result of the political fragmentation of a single geographic whole, and the goal of international coordination was to create universal standards that could apply equally throughout the world and anticipate all exceptions. In the 1950s and early 1960s, however, not only were the differences between land-use maps of Italy or Japan no longer seen as fatal, but these differing standards were seen as the result of the natural, rather than political, subdivisions of the world, and the goal was instead to tailor mapping to local needs.75 In other words, cartographic particularism went from being a logistical problem— one potentially solvable through standardization— to being a reflection of fundamental geographic differences that could not (and should not) be reconciled. Comparing these four projects to the aeronautical-chart activity of ICAO makes it clear that these trends were not confined to academic geography; they were instead general shifts in cartographic sensibility. ICAO itself did not make any maps— it was simply a site for government mapping officials from around the world to reach agreement— and the aeronautical-chart committee did not meet at all between 1951 and 1959. But as a result, debates at ICAO closely tracked a growing generational divide between wartime and postwar mapmakers.76 Cartographers at ICAO in the 1940s had pushed for the maximum standardization possible for high-quality cartography; their successors instead pursued the minimum amount of standardization necessary for safe flying, and regional coherence became more important than global uniformity. At the 1951 meeting, for example, the ICAO committee decided that charts in the Americas should use different elevation colors than the rest of the world, and countries could choose between several colors for radio information.77 In 1959 even greater divergences were sanctioned: colors for elevation could be shown either with the traditional green-to-brown gradient or with new pastel shades of pink and purple, and many nonaeronautical symbols were made optional. The logic was the same at both meetings. Unlike the original American WAC, which was produced for combat pilots who might be deployed anywhere in the world, civilian pilots using the International WAC would stay much closer to home and generally fly only along familiar routes. The goal in 1951 was to eliminate “undesirable rigidity” in standards that were “of no consequence to safety”; it was decided that individual maps need only be in “general conformity” with the specifications to receive the ICAO logo.78 As the head of the ICAO map division argued a few years later, aeronautical charts simply did not need to be perfect, and experience showed that they were “good enough” to avoid accidents.79 This graphic permissiveness was accompanied by similar changes in the logic used to divide mapping responsibility between countries. The original specifications had included a rigid agreement for allocating sheets between thirty-six countries; the goal at that time had been to keep sheet lines for the Maps as Tools

99

International WAC as close as possible to those of the ongoing American series.80 Two things changed in the 1950s. First, many sheets were consolidated or eliminated for the sake of local convenience. Irregularly shaped countries like New Zealand, the United Kingdom, and Portugal merged many of their sheets, and roughly fifty charts were eliminated in ocean areas where having a 1:1,000,000 map was relatively pointless.81 But second, and more important, was the changing process by which these decisions were made. The original sheet divisions and allocations were made directly by the ICAO secretariat in order to ensure that all mapping conformed to a single plan. In the early 1950s, this scheme was regarded as essentially fixed: when Egypt wanted to adjust the borders of its sheets in 1951, it had to submit a formal petition, and permission was granted only with reluctance and caveats.82 At the 1959 meeting, this kind of central planning was radically curtailed in favor of decentralized regional coordination. Neighboring countries were allowed to reach bilateral agreement about sheet edges and responsibility, and individual states could design their own charts for national convenience as long as no land went unmapped.83 Although never stated so baldly at the time, taken together these changes in the 1950s were essentially an admission that the International WAC was not, and would never be, a global series like its American forebear. Eventually this became not just a question of graphic standards and national responsibility, but map coverage as well. Compare, for example, the maps in figure 2.16; these are index sheets from the official ICAO chart catalog from March and October 1962. Maps much like the former had appeared in every catalog since the first, in 1951, and the latter has been the template ever since. In the earlier map, the edges of all possible sheets are shown, including those covering Antarctica, the Soviet countries (none of which were members of ICAO), and isolated islands. This is a map of optimistic universalism. A few months later, these unassigned sheets were no longer indicated on the index, and instead of showing a worldwide system slowly progressing toward completion, the ICAO catalog began to present a fragmented system designed only for the benefit of its members. This graphic change foreshadowed real changes in ICAO mapping: over the next decade, factors as diverse as decolonization and the densification of European air traffic would lead to sheets being dropped from the series in large regional blocks.84 These changes in thematic and aeronautical mapping left the idea of the universal base map on rather shaky ground; after all, it was these five projects that were the ultimate justification for a robust program of IMW mapping. The overall trend was toward ever-greater cartographic regionalism. The slippages in civilian mapping reinforced the idea that the world was not divided into cleanly bordered areas, but into a complex system of overlapping geographies— some natural, some political— each with its own mapping needs. The territorial state was clearly still important, but the prewar vocabulary of national sovereignty, “dignity,” or scientific competition was nowhere 100

Chapter Two

Figure 2.16: Index sheets showing coverage of World Aeronautical Charts produced to the standards of the International Civil Aviation Organization, from March 1962 (top) and October 1962 (bottom). The earlier map shows a global system, one that includes all land areas of the world, even those covering non-member-states like the USSR. In contrast, the later map shows only sheets actually assigned. This was the only time the graphics of the ICAO index map were changed. From ICAO, Aeronautical Chart Catalogue, Doc-7101 (Montreal: ICAO).

in evidence; instead, the goal was simply to match the space of the map with the space being investigated or administered, be it at the level of the state or not. And no civilian agency— not even the UN— actually administered anything at a global scale. The question here was as simple as it was radical: Why produce a globally uniform map when the world itself showed no signs of uniformity?

The Death of the Universal Reader: Pilots, Soldiers, and Functional Specificity The base-map ideal rests on a powerful theory about the sociology of geographic knowledge, with the relationship between the base map and its various thematic offspring corresponding directly to a distinction between general and specialist readers. The basic/thematic dichotomy thus makes two social claims: not only does it suggest that there are in fact general readers who want general knowledge (where general knowledge is a universal core held in common by all specialists), but it also suggests that specialist readers are simply general readers who want more information. During the 1950s, both of these ideas would be subject to profound critique. Crucially, they would be most thoroughly criticized by exactly those cartographers who continued to work on global projects, especially the aeronautical and topographic mapping of the US military. Not only did military mapmakers abandon the idea of the general reader, but accommodating the specialist reader— that is, every reader— was seen as requiring a wide variety of maps that could never share a single source. By the early 1960s, this sensibility began to influence both ICAO map standards and the IMW itself. The clearest meditation on specialist users, map function, and changing functional requirements was in aeronautical chart design. These themes had always been important for aeronautical mapping, but after World War II they were foregrounded, and cartographers in several countries came to worry, as early as the late 1940s, that charts were not keeping up with changes in aircraft technology. Given developments like radar or high-speed jets, traditional measures of cartographic quality could no longer be trusted; at the same time, standardization— the typical way cartographers reached consensus— was seen as much too slow and conservative a process. What was needed was rapid innovation.85 Almost without question, the presumed source of this innovation came to be social-scientific (often quantitative) research on the use of maps by working pilots. Nothing about this research was especially complicated. One of the most prominent early studies— a 1952 project to evaluate two new design alternatives for the American WAC, sponsored by the US Office of Naval Research— involved only simple knowledge-recall tests and questionnaires given to pilots and navigators. Later work was similar, with pilots being treated both as test subjects and as informants.86 But despite its simplicity, this work made rather 102

Chapter Two

ambitious claims that cartography could be a research-driven science, a subfield of psychology and psychophysics capable of cumulative progress. This redefined both the task of the cartographer and the nature of the map. Air force cartographers began to speak of a distinction between what they “believe” about good map design and what they “know,” and one of the first principles of responsible design became distrusting their own expert sense of what made a readable map.87 In turn, maps were treated not as repositories of geographic information, but as “a visual display problem in the man-chart system”—that is, as an arrangement of stimuli designed to produce a certain set of effects.88 There were certainly precedents for this kind of attitude— the 1921 magnum opus of the German economic geographer Max Eckert, Die Kartenwissenschaft (subtitled “Research and Foundation for Cartography as a Science”), is perhaps the most famous— but much of this earlier work sought to find universal principles of perception, especially as related to color theory or the psychology of propaganda.89 What distinguished aeronautical-chart research was that it did not treat pilots as a subset of humans in general. The emphasis was not on applying the rules of human perception to maps, but on creating specific “tools” (scare quotes included) specifically for pilots, whose specialized training and operational constraints did not make them typical at all.90 Nearly all 1950s research on aeronautical chart design supported two basic conclusions. First was a rather harsh indictment of information overload and unnecessary abstraction. Map-design research showed unequivocally that American pilots preferred charts that were uncluttered and that used fewer, more pictorially evocative symbols, as in figure 2.17. As simple as this conclusion was— and it was quickly accepted by cartographers in other countries and contexts as well— it radically reframed the goals of military chart design. It suggested that the problem was not one of giving the pilot enough information to cover any contingency; instead, the goal was to transmit only what was necessary, as quickly and efficiently as possible. As one air force cartographer put it, “If the user cannot readily extract the information he requires or is forced to perform complex mental integration before he can use the information presented, the design is inadequate.”91 Even geometric accuracy could be compromised. The pilot did not need to know his exact location; what he needed instead was “rapid, positive recognition of his approximate position.”92 And for high-altitude flight, the pilot did not need to know precise elevation information; he only needed to recognize topographic features from the air and know the minimum flight altitude necessary to clear all obstructions.93 The pilot, in other words, did not require more information than a general user; he required less. The second trend, a direct corollary of the first, was a gradual but decisive shift away from discussion of a “basic” or “general” navigation map, and toward what American cartographers called a “family of charts.” Criticism of “general” cartography was explicit and unrelenting. The authors of the 1952 study, for example, reported that “the intent in the past has apparently Maps as Tools

103

Figure 2.17: Design alternatives tested for the US Office of Naval Research in 1952. All three maps show roughly the same area in southeast Virginia (note the same square “Caution Area”), and all are shown here at actual size; the two experimental charts (below) used a smaller scale than the 1:1,000,000 WAC. The two experimental charts showed vastly less detail than the original WAC: gone is information about airways and radio frequencies, and features like railroads and urban areas are much less prominent. The goal was to give the pilot only what was necessary, using the smallest amount of paper. From John E. Murray and Rolland H. Waters, “The Design of Aeronautical Charts II,” Navigation (US) 3 (Dec 1952): 193.

been to prepare an all-purpose chart”; they acknowledged that “if this could be made, it would be ideal,” but they ultimately saw this as an “impossible” task. After all, trying to make the WAC be all things for all users was exactly what had led it to a state of cluttered confusion, and the different needs of highspeed military planes and low-speed civilian flying were only growing wider apart.94 The head of the US Aeronautical Chart Service was especially fond of specialization; his attitude was that “each chart must be tailor-made for a 104

Chapter Two

particular purpose— a particular aircraft-mission combination.”95 This meant that the distinction between “general” and “specific” was simply inoperative; as an air force mapmaker put it in 1959, “It is also evident that a classroom wall map is not simply a ‘general’ map, but is a graphic required to transmit certain information to the individuals with whom it is to communicate.”96 The task of the cartographer was to make as many different kinds of charts as there were tasks to perform: not only did flight planning, ground maneuvers, takeoff, flying, targeting, approach, and landing all require different maps, but each type of airplane or weather condition might call for ever more variations. Air force cartographers began to refer to such a collection of aeronautical charts as a “family” as early as 1951; the metaphor was meant to convey that all varieties of charts could form an organic whole— one where different charts would share certain resemblances without being rigidly uniform. The aim of the 1953 “Family of Charts Development Project,” for example, was to make various kinds of planning and approach charts integrate seamlessly with the WAC; the goal was not to have all maps look the same, but to make each type of chart do what the others did not, while still maintaining harmony throughout the series. This task involved much more than the kind of coordination and graphic standardization that had long occupied cartographers. Instead, as shown in figure 2.18, the American WAC was radically reconceived and redesigned for the specific purpose of supporting “on the deck” flying at high speeds and low altitudes.97 This kind of specialization continued through the 1950s, and by 1958 the air force was maintaining no less than eighty different kinds of aeronautical charts, “each designed for a specific purpose.”98 By the end of the decade the WAC was still the “cartographic workhorse,” but this was only because most flying still took place at altitudes and speeds where it was the most useful. For newer planes, the “primary chart” was the Jet Navigation Chart, a global series at 1:2,000,000 that was designed for pilots who relied more on radar than visual flying.99 In the 1960s, the WAC itself was gradually superseded by the new Operational Navigation Chart, another global series at 1:1,000,000 that was designed for midaltitude flying and used an entirely new set of conventions and sheet lines.100 Although  ICAO charts never approached this kind of microspecialization, the international trend in the 1950s followed a similar logic. In 1958, for example, a French cartographer challenged the purpose of the International WAC in terms that directly echoed American language: “Who are the actual users of this chart? Is the chart in great demand?” After statistics were collected showing that the map was not in fact the most popular of ICAO maps— and that most airlines did not actually use ICAO maps in the cockpit at all— British cartographers (and their counterparts elsewhere) saw this only as additional evidence that rigid standardization was counterproductive, since experienced commercial pilots had different needs from the casual users of government-issued maps, and airlines had every right to design maps “specifiMaps as Tools

105

Figure 2.18: American WAC of 1948 (top) and 1957 (bottom); both maps show the same area in northwest Italy, reproduced here at actual size. The major difference is the use of naturalistic shaded relief on the newer WAC instead of the elevation layer tints on the 1948 version. But note as well that the new version includes more detail near large cities, such as in the circle around Turin, and numbers giving maximum elevation in each square of latitude and longitude (for example, 125 means that the maximum elevation is 12,500 feet above sea level). The newer WAC was specifically designed for high-speed, low-altitude flight. Both maps from Richard W. Philbrick, “New Design Features for the World Aeronautical Chart,” Surveying and Mapping 17 (July–Sept 1957): 303, 306.

cally designed to their requirements.”101 A comparison of successive editions of ICAO chart standards also shows a clear shift toward “family” thinking during the same period. Figure 2.19 shows these trends in graphic form. From the late 1940s to the mid-1960s, new types of charts were added at each meeting of the ICAO map committee, sometimes even over the complaints of mapping agencies themselves; all were for special-purpose maps.102 The internal organization of ICAO standards also changed to deemphasize the identity of the International WAC as a “basic series” for general use. Between 1957 and 1961, it moved from being the very first chart discussed— in chapter 2— to being the seventh, in chapter 9. Also in 1961, a new chapter was added for “General Specifications” that included common standards that applied equally to all ICAO-sanctioned charts. In other words, ICAO charts were increasingly tailored to specific purposes and came to be designed and presented as a single unit, ordered not from “general” to “specific,” but as a motley collection with no clear hierarchy.103 Aeronautical charts were the most common target for functional specialization and user-centric design, but a similar logic permeated the more traditional topographic map design of the US Army as well, and concerns with innovation and “user requirements” were commonplace. As early as 1953, a cartographer at the Army Map Service predicted— quite correctly— that the “military topographic maps of the future . . . will become more specialized and will be designed to meet the particular needs of the individual service users.”104 As this sensibility came to permeate army operations, the assembly-line rhetoric of World War II shifted decisively from a Fordist to a post-Fordist idiom. As an officer stationed in the Philippines explained in a 1956 conference paper, army maps are a mass-produced product, but “the product is something like soap,” where “the customer demand for soap results in its being offered in a wide variety of qualities and forms to suit different tastes and uses.” A trafficability map for tanks was like laundry soap, and a geological map for engineers was a baby bath: different products for different markets.105 Like their air force counterparts, army map designers thus came to reject undue abstraction and graphic overload, and they also came to favor designing a multitude of different maps rather than having one map for all tasks. At the same time that the air force was researching experimental jet charts, the Army Map Service initiated a major program to test “entirely new concepts in graphic presentation of terrain” using questionnaires given to active soldiers.106 And although the army still maintained a distinction between “common” and “special” maps and coordinated its map series primarily by scale rather than function, each scale was increasingly associated with a specific user group, and no scale was seen as a one-size-fits-all solution. Larger-scale maps were meant for artillery, tactics, and fieldwork; smallerscale maps were for planning, strategy, and office work. These associations between scales and uses were not themselves new, but they had never before been so explicit.107 Maps as Tools

107

Figure 2.19: The place of the International WAC in ICAO standards, 1949– 2001. Although the WAC was always described as the “basic” chart, it moved from being first in a clear hierarchy (in the early 1950s) to being simply one among many charts, each designed for a specific purpose. All information from ICAO, International Standards and Recommended Practices: Aeronautical Charts: Annex 4 to the Convention on International Civil Aviation (Montreal: ICAO, various dates).

And just as the American WAC slowly faded in importance, military topographic maps at 1:1,000,000 likewise had a rocky history during the 1950s. During World War II, both the British and American armies had made generalpurpose, IMW-style maps to coordinate between land and air, but these series were discontinued in the late 1940s; the director of British military survey later described designing this kind of multipurpose map as “a well nigh impossible task.”108 A program for joint US-UK mapping at 1:1,000,000 was relaunched in 1952, but this time the maps’ designers were quite specific about how it would be used, and no attempt was made to accommodate users outside the army. One engineer expected the new map, known only (and rather dryly) as Series 1301, to be “one of the most popular and useful maps in higher headquarters” but pointed out that it was not suitable for field use and would only really be useful when multiple sheets were stitched together in classic war-room style. This meant that the map had a projected audience of perhaps a few hundred officers.109 Despite initial enthusiasm and the achievement of near-global map coverage in just a few years, the limited and desk-bound audience meant that this series was given increasingly less priority over the course of the 1950s. By the end of the decade the head of the Army Map Service was predicting that in nonthreat areas like South America and Australia, ten or twenty years might pass before maps could be updated. Even in medium-priority areas, three or four years might lapse between starting and publishing a sheet, and many maps were obsolete even before their ink was dry. Although the series continued to plug along until the late 1970s, the larger-scale maps used by actual soldiers were always higher on the list.110 All told, the new functionally specific approach to global mapping pursued by US and Allied militaries— and in some measure by ICAO— had two implications for the International Map of the World: one direct and one indirect. The direct impact came through Series 1301. Many of the maps treated by the UN as official IMW sheets (and included in figures 2.1 and 2.2) were actually Allied military maps that did not strictly adhere to IMW requirements, especially in areas like the USSR. (They were not, however, classified as “provisional.”)111 The impact of these sheets on the total IMW program was huge: of the seven hundred available sheets listed in the 1961 IMW annual report, more than six hundred were from Series 1301. As military interest in 1:1,000,000 mapping faded, so too would the IMW seem to languish.112 The indirect impact, however, was perhaps even more profound. Over the course of the 1950s, the professional cartographers who might most be expected to defend the usefulness of uniform worldwide mapping— the designers of aeronautical charts and topographic maps for the US military— had come to explicitly reject the idea of making a “general” map. This was not so much about rejecting any particular part of the IMW program— the scale, sheet lines, graphics, or allocation of responsibility— as it was about asking a new kind of question: Who would use this map? Once a specific user group could be identified, then a map (or multiple maps) could be made. But without a clear sense of who and for what, nothing could be done at all. Maps as Tools

109

THE QUIET DEATH OF CARTOGRAPHIC UNIVERSALISM: THE 1962 CONVENTION AND ITS AFTERMATH In the ten years after the 1952 IGU recommendation that an international conference be convened to revise the specifications of the IMW and solidify its place as a global base map, the shape that these changes might take was a topic of ongoing debate. Leading up to the conference— finally held in 1962, in Bonn, under UN sponsorship— the IMW was discussed at no less than nine other international meetings, both within and outside the UN system, and more than a dozen articles were published in Europe and the United States detailing the goals of the project and the possible relationship between the IMW and the International WAC.113 The conference itself was also no small affair: it lasted three weeks, included 188 cartographers and diplomats from forty-two countries and four international organizations, and was accompanied by a huge exhibition of several hundred 1:1,000,000 maps published since the late nineteenth century.114 The unquestioned goal of the conference was to shore up the ideal of universalist cartography. The way this goal was pursued, however, ended up leading to exactly the opposite result, and the finalized IMW became an awkward compromise between the base-map ideal and a much more regional, functional sensibility. The immediate prelude to the conference was a 1960 British-sponsored IGU resolution stating that, in contrast to the 1952 decision to maintain two global series, “modern cartographic thought considers it to be possible to devise a common base map from which both IMW and ICAO charts can be derived.” A leading German aviation cartographer predicted that the conference would lead to the “organic renaissance of a uniform world scale,” with a “true world map” achieved by 1980.115 The French delegation likewise submitted a preliminary paper which championed the epistemology of the general/specific dichotomy: the IMW would be “like the trunk of a tree,” supporting all manner of other projects.116 There was some disagreement about exactly which international series should take precedence— the French, German, Brazilian, and Spanish delegations all wanted to derive aeronautical charts from the IMW, while the UK, Norway, and Italy wanted the reverse— but no one challenged the importance of creating a single, global map.117 Yet despite this agreement in principle, the actual debate at the conference stressed regional standards, decentralized decision making, and a need for a more focused understanding of the map’s audience. Regionalism was embraced even by the staunchest supporters of universalism. The French, for example, submitted a large pamphlet detailing all the various map symbols in use around the world, with recommendations for possible compromise. This pamphlet, however, was organized into three climatic zones— temperate, arid, and humid tropical— with different recommendations for each, including variations for such major elements as elevation colors and road classifications.118 The British delegates likewise pushed a scheme that 110

Chapter Two

would have provided different standards for “highly developed” and “less developed” parts of the world.119 The American representative suggested that the symbols for towns should not be standardized at all, since even within the United States variation was necessary for the sake, once again, of “local conditions.”120 In other words, even though all agreed that a single base map should exist for all parts of the world, no one actually pushed for global standardization. This conclusion was only underscored by attempts to better define the function of the map. Before the conference several articles still described the IMW as a “general and scientific” map, but several skeptics had started to question the purpose of such an endeavor. Immediately before the meeting, Gerald Crone (one of Britain’s most prolific cartographers) asked explicitly: “What type of map-user should be kept in mind in re-designing the map?”121 At the conference, the answer was given on the first day: in line with overall UN goals, the IMW would henceforth be a map for planning economic development, and the prospective users would be development experts like those employed by the official UN Technical Assistance program. This seemed to justify the overall map scale (not too big, not too small), but besides a general push for portraying more “economic features,” this imagined audience proved to be rather unhelpful— and no real-life planners were actually in attendance at the conference.122 This emphasis on economic development, however, did reinforce the idea that the IMW was not a worldwide series. Not only could maps of the United States, western Europe, or the USSR hardly be justified as development assistance, but in line with prevailing cartographic sentiment, the geography of development was itself an uneasy mix of natural climatic or hydrographic regions and national political units. The IMW might, for example, help with transborder projects like those on the Mekong River, and helping countries produce IMW sheets might also serve to develop a country’s internal mapping capacity. But even jet-set technical experts did not need a global map. The ultimate (and much-lauded) result of the conference was a rather drastic loosening of IMW specifications, in the name of “modernization,” “flexibility,” and “economy.” The new standards closely followed the changes that had recently been made to ICAO charts; the general sentiment was that setting some minimum amount of standardization was better than pushing for too much. Symbols were made less rigid, certain colors and variations were made optional (the conference approved two signs for rice paddies, for instance), and special conventions were adopted for polar areas. Directly following ICAO precedent, the precise limits of sheet boundaries were also turned into national decisions. The conference did recommend that future IMW maps should use the aviation-friendly projection used for the WAC, but this was nonbinding.123 In other words, far from establishing a tight merger of the IMW and WAC to create a uniform, universal base map “for all geographical purposes,” the conference instead emptied the IMW standards of much of Maps as Tools

111

their former prescriptive force; their new flexibility was an unmistakable nod to decentralization and regional specificity. In line with these changes, work on the IMW in the years after the conference was increasingly fragmented and national. Large countries like Australia and Canada used IMW specifications to publish national maps of their vast and previously unsurveyed hinterlands. Awkwardly shaped countries like Japan, New Zealand, and the UK, along with smaller countries like Hungary, Cuba, and Malawi, all took advantage of the provision for modifying sheet boundaries to publish purely national wall maps branded with the “Carte Internationale du Monde” title. Contributions of Series 1301 from the US and UK, in contrast, dropped off precipitously.124 The UN secretariat admitted quite forthrightly that the new IMW was no longer universal; summarizing the work produced in the decade after the conference, it argued that the manifest “lack of absolute uniformity in cartographical details has little, if any, effect on the usefulness of the sheets”—which is to say, the project was no longer aimed at cross-border cooperation or comparative study between continents.125 And even though the IMW was recast as a map for economic development, the lingering idea that it was a useless (and authorless) “general” map continued to be the major theme in later debates about the project. The harshest critique was also the earliest: in 1964 Arthur Robinson— perhaps the most influential American cartographer of the later twentieth century— branded the map as little more than “cartographic wallpaper.” He argued that the “design by committee” and “attempt to serve too many purposes” left the map “cumbersome” at best. The scale was too small for fieldwork and too large to act as an effective base map for most thematic problems: “How it can be used effectively for ‘pre-investment surveys and economic development planning’ I do not know.”126 More than thirty years later, cartographers in the 1990s and early 2000s continued to echo these sentiments, especially when comparing the (by-then-defunct) IMW to Japanese proposals for international cooperation in electronic mapping data. For the promoters of this “Global Mapping” initiative— which in fact mixed national initiative, regional coordination, and several global data sets provided by the US— the main lesson of the IMW was “the need for clear, consistent and manageable objectives”; collaborative mapping had to have a defined audience, and a strong leader. Although it was similar to the IMW in many respects, geographers predicted that Global Mapping could succeed where the IMW failed because of its “clear focus on the environment” and the enthusiastic sponsorship of the Japanese government.127 These retrospective judgments underscore the degree to which universalism has faded as a cartographic goal, but they also make it clear that only certain parts of the IMW program have been rejected. In particular, the 1962 conference certainly did not signal the end of base-map thinking in general; if anything, layering new information onto existing maps has only become more prevalent since the 1960s. The new specifications did, however, signal a 112

Chapter Two

decisive shift in what it meant to create a base map. First, even if Robinson was correct and the renovated IMW was not actually useful for development planning, the implication in 1962 was that the distinction between “general” and “specific” was not as important as it once was. If an economic-planning map could be a base map, then why not a geological map, or a demographic map?128 Second, the newly flexible, regionalist IMW standards likewise acknowledged that the same area could be mapped in multiple ways, and that each of these maps could be equally authoritative— equally basic. No longer must there be only one neutral, comprehensive map.129 Considered just as arguments about how mapping should be organized, both of these propositions have proven to be quite prescient. Both, however, were also arguments about the political importance of cartographic knowledge. If any map could be a base map, and if the hierarchy between “basic” and “thematic” information no longer held, the implication was that the complexity of the world could no longer be reduced to a single authoritative image.

CONCLUSION: THE TERRITORIALITY OF CARTOGRAPHIC TOOLS Taken simply as the history of an ambitious international project, the rise and fall of the International Map of the World is an important story about the limits of collaboration, the place of cartography in international relations, and the shifting role of the United States and the various agencies of the UN. But it is also story about the fate of authoritative representation as a whole. These larger stakes were never clearly articulated by the cartographers involved in the project, and even the most skeptical among them still held fast to the ideal of cartography as a perfectible science. Beginning roughly in the mid-1980s, however, just as the IMW was officially coming to a close, a small but influential group of scholars began providing a new framework for understanding the waning ideal of universal cartography. The major figures— especially J. B. Harley, Matthew Edney, Denis Wood, and David Turnbull— drew from many sources (social theory and the history of science in particular) and analyzed all manner of maps, especially maps that had been ignored by previous scholarship. Their goal was nothing less than to rewrite the history of cartography using a new vocabulary of power, politics, and cultural bias; the label they used to describe their project was “critical.” The main thrust of this work was an analysis of cartographic representation that explicitly problematized the relationship between map and territory. On the one hand, Harley (and others) cited Alfred Korzybski’s 1941 aphorism that “the map is not the territory” to argue that no map can be an objective, comprehensive record of terrain: representation must not be confused with mimesis.130 At the same time, Turnbull— and Harley and Wood after him— argued the reverse as well, that in fact “maps are territories” because they actively construct geographic space. They constrain thought and action, and Maps as Tools

113

they enable certain practices and not others. Representations, in other words, are insidiously transformative: they are propositions, arguments, and tools.131 This critique of representation was a twofold indictment of the conventional wisdom about maps. First, it challenged the idea that maps— especially official, state-sponsored maps— could ever be epistemologically neutral, objective, “scientific,” or transparent. Rejecting such an interpretation as “empiricist,” the scholars of the 1980s argued that because maps necessarily involve selection, interpretation, and translation, they cannot avoid having a strong point of view. This point of view is constructed through the use of certain graphic conventions, marginalia, or— most important— strategic omissions. Maps are thus laden with the specific political or cultural goals of those who produce them: they are descriptions of how the world is, but they are simultaneously arguments about how the world should be.132 Second, and related, the critical approach also challenged the idea that the “scientific” map— again, especially the state-surveyed topographic map— was singular and comprehensive. Harley criticized the reductive distinction between “basic” and “derived” maps (that is, between the base map and its thematic offspring), while Edney saw the view that “at any one time, there should be only one map of one territory” as an ideological throwback to the totalizing project of the Enlightenment.133 One could easily see this program as a thorough, if implicit, critique of precisely the ideals codified by the International Map of the World. The goal of the 1913 specifications, after all, was to create a single, authoritative representation of the entire earth. It sought to create a seemingly neutral, objective, and comprehensive map— the repository of all geographic knowledge. And it is relatively easy to imagine how Harley, Wood, or Edney might describe it as an attempt to mobilize a discourse of accuracy and science so that the great powers could naturalize their own point of view, reify their interests, and preempt criticism. (Indeed, this approach informed my own analysis in chapter 1.) But even though the critical project is often seen as a radical break from earlier interpretive frameworks, it was not in fact a wholesale rejection of practitioners’ sensibilities. It is no small irony that two of the cartographers most criticized in the late 1980s as unrepentant empiricists— Arthur Robinson and Gerald Crone— were also two of the most articulate critics of the IMW in the early 1960s. Robinson and Crone had critiqued the IMW on pragmatic rather than political grounds (seeing it as useless, rather than pernicious), but their message had much in common with what would come later: trying to create a neutral, comprehensive, mimetic map (designed for a “general” reader) is a fundamentally misguided goal.134 Ultimately, what for Harley or Edney was primarily a historiographic shift— a new way of analyzing maps— is perhaps best understood as a solidification, even a popularization, of a broader historical shift in twentieth-century cartography. This is the shift from representation as a mimetic practice to representation as a tool-making practice— that is, exactly the shift that rendered the IMW obsolete. 114

Chapter Two

And as a historical phenomenon, it is important to keep in mind that much of the new sensibility emerged as a reaction— mostly an American or Allied reaction— to World War II. This was not a straightforward co-opting of cartography by the military; instead it was a realization, military and nonmilitary alike, that universalist mapping was wholly inadequate to meet the needs of global mobilization, both for the soldier abroad and the citizen at home. Even Robinson and Crone, who spent most of their lives as scholars, were deeply influenced by the war— Robinson even began his career as head of the Map Division for the US Office of Strategic Services (forerunner of the CIA)— and both saw clear connections between wartime mapping and a need for postwar innovation.135 After all, when maps became understood as tools, they were first tools for Allied pilots, then for national officials, commercial airlines, and development experts. The ability to critique these maps as tools of power is thus itself a continuation (although also a subversion) of a rather specific politicalgeographic project, one that sees an unbreakable link between new kinds of globalism and the protection or expansion of state interests— especially American interests. Seeing this connection between cartographic practice and cartographic theory is significant historically, but it also offers two broader analytic lessons. First, the “critical paradigm” provides a useful vocabulary for understanding the territorial significance of the IMW’s demise. As I have stressed throughout the chapter, the history of the IMW reveals a major change in the goals of cartography, from uniformity and universalism to regionalism and fitness for purpose. But this change in mapmakers’ priorities was also a change in the relationship between cartography and national space. In the early twentieth century, the IMW was seen as a fundamental project, both politically and scientifically, and the strong national bias of the map was therefore easily naturalized. (In 1941, “the map is not the territory” was an important admonition indeed.) After 1950, however, as authoritative representation was increasingly rejected as unhelpful, or even epistemologically bankrupt, there was no longer any single base map that might organize geographic space. Instead, each separate project— from the International WAC to the various maps of population, soil, and vegetation— used a different graphic language and mostly sidestepped the earlier politics of national responsibility and national graphics. (Instead of the map and the territory, the admonition that “maps are territories” is largely about plurality.) In other words, the reorganization of international cartography in the decades after World War II was a direct challenge to the primacy of national territoriality. With every regional graphic standard and every pilot questionnaire, it became increasingly difficult to see a well-ordered international system governed by a clean separation between domestic and foreign affairs. Territories multiplied, and not all were national. And this is true even though the key players were still firmly lodged within national agencies. The second reason I end with the “critical paradigm” has more to do with its larger politics. Since the 1980s, much of this project has been about unMaps as Tools

115

masking the power of mapping; the hope has been that showing the valueladenness of maps could lead to new kinds of resistance and perhaps even new strategies of countermapping.136 But treating geographic knowledge— and power— as a question of representation alone does not cast the net nearly wide enough. Not only have practitioners themselves largely abandoned the stronger versions of cartographic authority, but geographic knowledge as a whole has ceased to be only a representational problem. This is not to say that dreams of universal legibility were cast aside, or that spatial technologies ceased to be instruments of the state. Instead, the project of legibility shifted, and as authoritative representation waned, the geo-epistemology of the paper map was supplemented by other ways of organizing space.

116

Chapter Two

PA R T  I I

Cartographic Grids and New Territories of Calculation

CH A PT E R T H RE E

Aiming Guns, Recording Land, and Stitching Map to Territory: The Invention of Cartographic Grid Systems, 1914– 1939

It is two hours before dawn on Thursday, August 8, 1918. The fog is dense. A group of French soldiers is concealed in a dark stand of trees, along with their 155-millimeter howitzer and the map shown in figure 3.1. Since arriving at the front a few days before, they have not fired a single shot, and they cannot see the enemy. What they have done, however, is locate themselves precisely on the map: they are just northwest of the intersection of the lines labeled 24 and 51, and they report their coordinates as “3.85/1.06.” These coordinates— and the lines on the map— are measured in kilometers: 3.85 kilometers east and 1.06 kilometers north. (Giving their full coordinates would add another 120 and 350 to these numbers, but only the abbreviated versions are actually used.) They also have the coordinates of their target: 5.11/1.90, which is a German trench about one and a half kilometers away. They have no doubt that they can hit their target, despite having never seen it or practiced their aim. Indeed, calculating the aiming distance and direction are remarkably easy, since their coordinate system— invented only a few years before— has turned the entire western front into a flat, Euclidean gameboard subject only to the simple rules of plane geometry. In an hour, they will open fire and the Battle of Amiens will begin. It will be a finely coordinated surprise attack, with more than three thousand Allied guns using similar maps to launch their shells all at once, and it will finally push the Germans into permanent retreat.1 To cartographers, this Euclidean coordinate system is called, simply enough, a “grid,” as opposed to the more familiar “graticule” of latitude and longitude. The obvious advantage of grids is their simplicity: they make it easy to report locations, measure distances, and perform calculations. But this is a rather startling mathematical trick, since the earth is not in fact a flat Euclidean surface. And while the curvature of the earth can usually be ignored within a small enough area, this is not what is going on here: grids make it 119

Figure 3.1 (see gallery for color version): French map of the western front near Amiens, edition of 5 August 1918. Allied trenches (north and west) are in red; German trenches (in the southeast) are blue. Shown at actual size, scale 1:20,000, with lines spaced every kilometer. (Sheet Moreuil, Service Géographique de l’Armée, 1918.)

possible to use the same flat coordinate system throughout a remarkably large (and noticeably curved) region. In World War I, soldiers could ignore the curvature of the earth all the way from the English Channel to the Alps— a distance of more than five hundred kilometers. Grids are likewise independent from any single map sheet, just like latitude and longitude. The same point will retain its grid coordinates no matter the map’s scale, sheet size, or orientation; indeed, a usable grid can exist even if there is no map at all. As a US Army mathematician explained in the late 1940s, “Grid lines can be considered as a system of coordinates whose position on the earth’s surface is as exactly fixed as meridians and parallels.”2 So even though latitude and longitude will probably always remain useful for sailors navigating by the stars and the sun, for the ordinary map user, grids are a fully workable alternative with many advantages. Despite the fact that most people have, at best, only a vague awareness that these grids exist at all, they are arguably one of the most successful cartographic innovations of the twentieth century. After the first large-scale system was designed in 1915 by survey engineers in the French army, grids quickly became a general-purpose technology that could rival latitude and longitude not just in northeastern France but for the world as a whole. During the 1920s and 1930s, grids were civilianized and adopted in dozens of countries; they were used not just for artillery but also for property surveying, civil engineering, and marking boundaries of all kinds. Another big expansion took place during World War II, when grids began to overlap and cover the entire globe. Immediately after the war, the US Army unveiled a unified global system known as the Universal Transverse Mercator, or UTM, that has remained essentially unchanged ever since. This grid was again designed for military purposes— not just for aiming guns, but also for long-range missiles and airground coordination— but within a few decades it likewise found many additional uses, including international development work, field sciences like archeology and botany, and even recreational backpacking. For all these tasks, UTM made the entire world flat. It made geographic space more calculable, more accessible, and more connected— in a word, more coordinated. Indeed, grids are above all a technology of geographic cohesion and consolidation; as a popular postwar geography textbook put it, they are “probably the most definite and efficacious linkage of locality, region, nation, and world since man first worked out a concept of territory.”3 Today grids appear on nearly every large-scale map— see figure 3.2 for a recent example with UTM lines— and grid coordinates are a common option on GPS devices and Internet mapping services.4 As a new way of organizing geographic knowledge, grids have had widespread impact on maps, mapping, and the experience of space in the twentieth century. They prompted almost all national mapping agencies to redraw their maps using new projections with special mathematical properties; they also spurred extensive new surveys and the recalculation of old work.5 For Aiming Guns, Recording Land, and Stitching Map to Territory

121

Figure 3.2: Map of coastal Maryland showing the Universal Transverse Mercator grid. As on the World War I map, lines are spaced every kilometer, but the UTM coordinate system covers the entire world. The original scale is 1:24,000— shown here reduced by about three-quarters. (Sheet Solomons Island, USGS, 1987.)

the user, inhabiting a grid is also much more geographically immersive than the traditional god’s-eye view of a paper map. Unlike latitude and longitude, which locate all points with respect to the equator and the Greenwich meridian, grid coordinates usually locate points regionally rather than globally. (The abbreviated coordinates of World War I reduced the soldier’s horizon to a ten-kilometer square; UTM coordinates are more sophisticated, but they work in a similar way.) And unlike a representational map, which translates and miniaturizes the world onto paper, a grid is only successful if it leaves the paper behind and becomes installed directly in the landscape at full scale. Mathematically, grids are thus a hybrid technology, existing simultaneously on and off the map. Experientially, however, the geo-epistemology of grids has more in common with the full-scale logic of GPS than with the representational authority of the International Map of the World. The invention and expansion of grids therefore signals an important turning point in the history of territory, when the logic of the map starts to transition to the logic of what French military surveyors called the “canevas de points”—the framework of points.6 The question of scale is crucial here. Many scholars have enjoyed analyzing the idea of a map drawn at a scale of 1:1, as found in stories by Lewis Carroll or Jorge Luis Borges. The absurdity of a full-scale map exposes the hubris of the cartographic ideal and highlights just how selective any map must ultimately be.7 But the history of grids suggests that this is exactly how military engineers in the twentieth century made geographic space legible. Again the US Army mathematician: “A grid may be regarded as a set of perfect squares ruled on a plane map, scale 1:1, and then transferred to the earth’s surface.”8 Instead of unrolling a gigantic paper map (causing, as Carroll would have it, the farmers to complain about spoiled crops), the twentieth-century world was instead only overlaid with the mathematical structure of the map: the points of the grid, transferred to the earth. And this full-scale framework of points is what millions of soldiers, scientists, and hikers have used— and still use— to situate themselves in the world and interact with their surroundings. Together, this chapter and the next trace the long history of this prosaic yet powerful spatial technology. This first chapter analyzes the grids of World War I (and earlier precedents) before turning to the civilian grids of the 1920s and 1930s. The next chapter begins with World War II and then offers a detailed analysis of UTM from the 1940s to the early twenty-first century. My broadest questions are the same in both chapters: how do grids change geographic space, and how have they intersected with the territoriality of national states? In response to these questions, I argue that while the experience of full-scale geographic embeddedness is nearly universal to gridded space, it has served two very different political-geographic ends. Before World War II, grids reinforced existing jurisdictional boundaries and a sharp division between distinct national and international spheres. The grids of World War II, and especially the postwar UTM, instead actively subverted national territoAiming Guns, Recording Land, and Stitching Map to Territory

123

riality at both a symbolic and a straightforwardly practical level. Indeed, the entire point of UTM was (and is) to enable intervention across international boundaries and coordination in unfamiliar locales. This chapter is divided into four sections. The first two sections focus on the invention and the experience of artillery grids during World War I. I am interested in how the first grids were designed, the problems they were meant to solve, how they were used, and how they created a new technological ensemble— one that included not just maps but also printed text, large pieces of metal known as plotting boards, and extensive networks of survey monuments. It is primarily these practical relationships that guide my discussion of how the geo-epistemology of grids differs from that of representational maps and operates at a scale of 1:1. The second half of the chapter focuses on the politics of grids in the interwar period. I look first at domestic politics by analyzing the civilian grid system of the United States. I ask about its relationship to the state, to professional engineers, and the changing rationality of space. Then in the final section I turn to the international politics of grids by looking at debates that took place at meetings of the International Union for Geodesy and Geophysics, the major site for scientific discussion and collaboration in high-precision surveying. Here I track how grids were connected to some of the largest problems in geodesy— including continental survey networks and even the size and shape of the earth— and how international debate helped police the boundary between national, territorial space and an international space that was explicitly nonterritorial. Methodologically, my goal here is to connect the design and use of an eminently practical technology to a conceptual history of mapping and territory. The main actors in this story, however, are mathematicians, surveyors, scientists, and engineers (mostly French and American), and these specialists did not often speak in grand terms about their work. Likewise, although the roles of particular individuals and institutions are often perfectly clear, grids as a whole have no single inventor or systematic theorist, and they do not serve any single set of interests. But this is part of their power. Grids are invisible even to most of their users, and their politics are concealed in difficult debates about map projections and the mathematics of surveying. Appreciating grids does not require a detailed understanding of all of these technical conversations, but it does require understanding how technical decisions can regulate both the experience and the politics of space.

THE INVENTION OF GRIDS IN WORLD WAR I The creation of the first grids during World War  I sits at the intersection of the grubby practicalities of military mapping in trench warfare and the often-abstract history of map projections. On the side of practicalities, survey officials were confronted with an unexpected “war of position,” where 124

Chapter Three

the successful use of long-range guns required detailed knowledge of local terrain rather than a more general sense of regional topography and lines of movement. Drawing on a variety of prewar precedents, the idea of giving gunners an easy-to-use coordinate system was seen as relatively obvious and was embraced by both sides quite early in the war. On the side of mathematics, however, the story is somewhat more complex, and the invention of grids intersects with some of the broadest themes in the history of cartography. The basic story is an intellectual gestalt shift that transformed a well-established (but almost always invisible) mathematical scaffolding— known to mapmakers simply as “rectangular coordinates,” as opposed to a proper “grid”—from a background technique of map production to the most visible and useful feature of a large-scale map. This initial change was made by both the French and German armies (joined later by the British and Americans), but only the French developed the new technique into a wide-ranging system— known as the système Lambert— that unified the entire western front into a single mathematical space. Before World War I, maps were generally not used to aim artillery. For longrange guns (especially those that could fire “over the horizon”), the preferred method was instead to position human observers close to the gun’s target who could watch each shell land and communicate with the gunners to adjust the range and direction after each shot. When communication was done through field telephone, or later radio, this also meant that the gun could be placed in a concealed position, greatly enhancing the safety of the gunners. This technique was first successfully used in wartime during the Russo-Japanese war of 1904– 1905, and on the eve of World War I a French artillerist described it as “the normal” way of engaging the enemy.9 When the Germans invaded Belgium and France in 1914, this same technique was widely used by both sides— and often to great advantage. But when the German advance was suddenly stopped in September at the Battle of the Marne and both sides began measuring progress in meters rather than kilometers, the limits of forward observation became increasingly apparent. One major problem was that the necessary telephone lines were now subject to prolonged bombardment and sabotage, and observers were often unexpectedly cut off from the gunners. At the same time, static trench warfare quickly blurred the traditional distinction between immobile heavy artillery and nimble field artillery, which introduced the daunting prospect of having to provide observational support for every gun.10 Dispensing with observation and having gunners instead calculate range and bearing using a map— a technique known as “map firing”—was well known as a possible alternative, but it was still an unproven method and previous attempts to use it during wartime had been relatively fruitless.11 The biggest problem at the start of World War I was that map firing required a level of cartographic precision that was simply unobtainable from the preexisting maps of northeast France. Figure 3.3 shows a close-up of one of these maps, Aiming Guns, Recording Land, and Stitching Map to Territory

125

Figure 3.3: The primary topographic map of France in the nineteenth and early twentieth centuries, the carte de l’État major. This detail— from an 1885 revision of an 1837 original— is shown at actual size, scale 1:80,000. The thin lines show latitude and longitude, spaced every tenth of a grade (0.09 degrees). Note the use of hachures to show elevation, rather than contour lines. (Sheet 21, Montdidier, Dépôt de la Guerre, 1885.)

known as the carte de l’État major (map of the General Staff). Some of the problems with this map were relatively straightforward— the scale, for example, was too small to show enough detail, and locations were known to be inaccurate on the order of tens of meters— but the fact that references were most easily given in latitude and longitude rather than in meters also made the map more useful for wayfinding than artillery. There were also problems with the map projection itself— the Bonne projection— which deformed lengths and angles far too much for artillery work. (Figure 3.4 explains this projection.)12 Even if map firing had already been a fully mature technique (which it certainly was not), it would have been impossible to deploy in the early months of trench warfare. The response from all four countries on the western front— the French, British, Belgians, and Germans alike— was thus to begin producing maps at a much larger scale and with much greater levels of accuracy, complete with a gridded overlay of reference lines measured in meters. The first grids were printed in late 1914, and the first new surveys were complete a few months later.13 None of these early trench maps solved the mathematical problems with the projection, but by enabling communication in meters— or for the British, yards— rather than in latitude and longitude, even the most slapdash of the first grid systems succeeded in transforming the curved surface of the 126

Chapter Three

Figure 3.4: The world map that would result if the Bonne projection used for the État major were extended. Every map projection— that is, every mathematical translation between curved earth and flat paper— must inevitably introduce distortions. Different projections, however, introduce different kinds of distortions. The Bonne was chosen for the État major because it preserves area, meaning that one square centimeter of paper always corresponds to the same amount of land or water, anywhere on the map. But as a result, lengths and angles are unavoidably distorted— notice that with increasing distance from the Paris meridian, right-angle intersections of latitude and longitude, along with north-south distances, become ever more mangled. On the western front, the Bonne projection caused errors of 0.17 percent in scale and 21 centigrades in angle (about 0.2 degrees); French survey officials found the former “perceptible” and said the latter “is relatively considerable and markedly exceeds the limit allowed by modern artillery”; see Service Géographique de l’Armée, Rapport sur les travaux exécutés du 1er août 1914 au 31 décembre 1919 (Paris: SGA, 1936), 35.

earth into a flat Euclidean plane and allowing gunners to aim from the map, at least locally. These grids also helped integrate a wide array of users into the same geographic system: they were used not just for heavy and light artillery, but also for aerial photography, reconnaissance, and, for Germany, even small weapons like mortars and machine guns.14 While it may seem remarkable that these earth-flattening reference grids began appearing on each country’s maps almost simultaneously, very similar systems had in fact been lurking in the background of mapping in nearly every country for over a hundred years. Even the creators of the État major had used them. And understanding the deep history of grids makes it clear that these systems were not just a clever solution to an artillery problem. Instead they reveal an important and long-standing ambivalence about the basic problems of cartographic projection and scale— and by extension the logic of representation in general. Indeed, the earliest and most important precedent for wartime grids was developed as part of the very first systematic national map series: the original carte de France begun in the 1730s by the renowned astronomer César-François Cassini de Thury (also known as Cassini III, the third in a line of four famous Cassini astronomers). As shown in figure 3.5, one of the notable Aiming Guns, Recording Land, and Stitching Map to Territory

127

Figure 3.5: Detail of a sheet from Cassini’s carte de France. On Cassini’s projection, this corner is 100,000 toises to the right and 37,500 toises up from the Paris observatory (see location in figure 3.6). Shown here enlarged from the original scale of 1:86,400. (Sheet 110, Verdun, 1760.)

features of this map is that in the corners of every sheet Cassini included the x-y distance of that corner from the Paris observatory, which served as the astronomical anchor for the entire survey. (The distance was measured in toises, each slightly less than two meters.) Figure 3.6 shows how the same coordinate system was used as an index for the series as a whole.15 Mathematically, Cassini’s system was all but identical to some of the early grids of World War I. The crucial difference was that his coordinates were only used when making the map— not when using it— and the finished maps were ruled with lines of latitude and longitude rather than an even rectangular grid. Indeed, while Cassini’s coordinates were certainly noticeable, they were hardly prominent, and his system was meant primarily as a way of simplifying the process of laying out sheets and plotting survey data. Instead of using latitude and longitude as the framework for the map— and therefore having to convert all survey measurements and perform all calculations using complex trigonometry— Cassini simply used a Euclidean grid aligned with the Paris meridian as his starting point and projected lines of latitude and longitude, along with key points surveyed using high-precision triangulation, into his two-dimensional system. What this meant was that the detailed filling-in of the map could then be done using intuitive calculations in everyday units, without having to constantly revisit the underlying mathematics. Although Cassini’s map would be described today as using the Cassini projection, historically his method was more often described simply as “plotting by rectangu128

Chapter Three

Figure 3.6: The index sheet for Cassini’s maps of France. The numbers along the top and right sides are Cassini’s rectangular coordinates. The shaded area is the sheet from figure 3.5. (Published 1797; shading added.)

lar coordinates” or “the rectangular system.”16 In other words, Cassini’s technique allowed a mapmaker to plot a map— and, by extension, to record survey data— as if the world were perfectly flat and no projection were necessary, even over a very large area.17 (Of course this could only ever be a convenient fiction, since any translation from curved earth to flat paper is, by definition, a projection; Cassini’s was simply implicit. Consider again the index diagram shown in figure 3.6: his rectangular grid preserved accurate distances horizontally and along the main Paris meridian, but elsewhere vertical distances and angles were not correct. These errors were usually ignored.) As a way to simplify mapmaking, Cassini’s system spread rapidly in the nineteenth century and was dominant around the world by the time of World War I. As early as 1805, the head of the Bavarian survey, Johann Georg von Soldner, called Cassini coordinates the “common method” for transforming raw survey data into a map. In 1810 the main popularizer of rectangular coordinates in France, the mathematician and ingénieur géographe Louis Puissant, wrote that even ordinary surveyors could use them with ease: they could Aiming Guns, Recording Land, and Stitching Map to Territory

129

“reduce the tracing of a map projection to a purely mechanical operation.”18 Cassini-like coordinate systems were created for each county in England and many of the German states; they could also be found in Switzerland, Massachusetts, New York City, and throughout colonial Africa.19 Even when divorced from the particularities of Cassini’s own projection, his basic approach to plotting could still be quite helpful, and rectangular coordinates could be used with any projection. This had been the case with the État major, which continued to show (projected) distances from the Paris observatory in the corners of each sheet. The initial grids of late 1914 and 1915 thus took this time-honored cartographic technique— along with its ability to flatten the world— and brought it to the attention of everyday soldiers, thereby turning “rectangular coordinates” into a “grid” just by printing a few extra lines. From a mathematical perspective, this indeed meant enlarging the map to the scale of 1:1, since it involved taking the rectangular coordinates of the map, measured in meters but previously confined to the miniature world of paper, and using them as full-scale, real-world coordinates to refer to guns and targets. This is how surveyors had always used these systems when doing fieldwork, but now a full-scale reference grid became one of the standard features of the battlefield. This initial breakthrough, however, presented several new problems. The first was that transferring rectangular coordinates from the paper map into the real world meant that the distortions of Cassini’s projection (or the Bonne, or any other) were also transferred at the same time. Trenches might have grid coordinates that differed by 8,000 meters but were in fact only 7,990 meters apart, and yet all map firing would still proceed as if the map distances were correct. The easiest way to confront this problem was to limit the geographic size of a particular coordinate system and therefore keep these errors below the inherent accuracy limits of the guns themselves. The Germans, for example, found that they could apply a Cassini system within sixty kilometers of a particular line of longitude without introducing any noticeable inaccuracies, and the French experimented with similarly sized grids based on the Bonne projection.20 In turn, however, this only introduced the secondary problem of dealing with the junctions between neighboring grids. Figure 3.7, for example, shows a detail from a German trench map that includes a collision between two Cassini coordinate systems. Across this kind of grid mismatch, easy calculation and targeting would have been impossible, not just because of discontinuities in the coordinates themselves but at times because of discontinuities in the underlying survey data as well. (At the junction of French and German surveys, for example, coordinates could be mismatched by as much as 230 meters.)21 And since grid boundaries tended to align with the administrative divisions within an army, they also exacerbated the difficulty of coordinating between units. At the height of the war, Germany used as many as nine separate grids on the western front— one for each field army.22 The British took an especially dim view of these German grid junctions—after 130

Chapter Three

Figure 3.7 (see gallery for color version): Detail from a German trench map showing a junction between two grids: the Netz von Paris and the Netz von Lille. Notice the discontinuity both in the reference numbers on the left margin and in the grid pattern overlaid on the map (the grids are also rotated roughly 0.7 degrees from each other). It would be impossible for artillery to aim across this break; it also led to general confusion when reporting coordinates. Other German maps, instead of showing a sharp discontinuity, would sometimes provide overlap between adjacent grids and show them in different colors— but this also caused confusion. This detail of sheet Bourlon, edition of 18 Sept 1918, is shown at actual size (scale 1:25,000). Facsimile from Oskar Albrecht, Das Kriegsvermessungswesen während des Weltkrieges 1914– 18 (Munich: Bayerische Akademie der Wissenschaften, 1969).

the war, one survey official called them a “crazy patchwork”—but their own early approach was not necessarily better, and often much worse. All armies faced the same problems.23 During the war, only the French solved the problem of grid junctions on the western front. Their solution was simply to make a much larger grid. In late 1914, a handful of French survey officers— most of whom had extensive mathematical training from the prestigious École Polytechnique—undertook a wide-ranging study of map projections to find the best framework for an artillery grid. What was necessary, they realized, was a conformal projection: one with no angular distortion, where lines that intersect at a certain angle in the world intersect at the same angle on the map. This would allow a grid to cover a much wider geographic area; it would also make it easier to correct for length distortion.24 The immediate source for this insight was a long article published in 1912 by André Courtier, a French hydrographic engineer who had been working in Madagascar before returning to France for the war. Courtier’s work had been motivated by the same considerations as Cassini’s, except that his goal was to make rectangular coordinates useful not just for everyday surveying, but also for the angular calculations of the high-precision triangulation being done in Africa.25 Conformal projections had already been known for many decades— conformal coordinate systems had even been developed for surveying by the famous polymath Carl Friedrich Gauss as early as the 1820s— but the mathematical simplicity of other projections (especially Cassini’s) always outweighed the benefits of conformality.26 Even the projection recommended and formally mathematized by Courtier was quite old; it had first been proposed in the late eighteenth century by the Swiss mathematician Johann Lambert. In June 1915, the French Army formally decided to adopt Courtier’s equations for all its artillery maps, and the Bonne projection was mostly abandoned by the end of the year. Figure 3.8 shows how the resulting grid— named the système Lambert, after the projection— put the entire western front on the same coordinate system, with no disruptive junctions. (Figure 3.9 explains the projection.)27 As convincing as Courtier’s mathematics were, however, there were also strategic considerations at play. One of the properties of the Lambert projection is that it can easily be extended east and west without increasing distortion, and hope remained that the static trench warfare would be broken and the front pushed into Germany. (Ironically enough, the same considerations convinced Germany not to create a large conformal grid. In mid-1915 the Germans considered a system based on Gauss’s projection that could instead be extended north or south, but when the leader of German military mapping, Siegfried Boelcke, formally recommended it that autumn, it was rejected by high-ranking generals who thought they would soon advance westward into France.)28 The kind of large regional system created by the French was not strictly required for map firing, and it did not solve all problems once and for all— it 132

Chapter Three

Figure 3.8: The coordinates of the système Lambert. The French army applied a kilometer grid (as in figure 3.1) to maps in the shaded area between late 1915 and the end of the war; the black line shows the location of the front as it stabilized in late 1914. With this grid system, artillery fire could be coordinated all along the western front. The center of the Lambert projection was put in western Germany in hopes that the trench warfare would break and the French armies could advance east using the same grid. This map (with the prewar French-German boundary shown dotted) is based on Service Géographique de l’Armée, Rapport sur les travaux exécutés du 1er août 1914 au 31 décembre 1919 (Paris: SGA, 1936), 39, 309, and plate X.

Figure 3.9: The world map that would result if the Lambert projection used for the système Lambert were extended. Compare to the Bonne projection in figure 3.4, and note that here all intersections of latitude and longitude are shown with no angular distortion— all perfectly perpendicular, just as in the real world— even though length and area distortions can be quite significant. Projections that preserve angles are known as conformal. Only a few such projections exist, but they have been used for nearly every grid created since the système Lambert.

was Germany and Britain, for example, that most developed the technique of “predicted fire” (that is, map firing by surprise), and bringing speed and flexibility to blind targeting remained difficult for everyone29—but by the end of the war a conformal grid was universally seen as the best solution from a mathematical and logistical point of view. Although German survey officials failed to create such a grid on the western front, they did use large conformal grids in the east, and near the end of the war they were pursuing grid unification with Austria as well.30 And in the spring of 1918, when the Allies, including the United States, finally formed a unified command hierarchy in response to the German “Spring Offensive,” survey officials from the UK, Belgium, Italy, and the US all recognized the superiority of French methods and agreed to adopt them as quickly as possible. In their official reports after the war, both American and British survey officials were happy to acknowledge the advantages of the French grid. The American report in particular called the système Lambert a “splendid system” and said in no uncertain terms that “a battle map for modern warfare should be first and foremost an artillery map”—that is, a map that would flatten the earth and allow coordination and easy calculation within a large geographic area.31

THE EXPERIENCE OF THE GRID As important as the printing of kilometer grids and the mathematics of conformal projections were for the success of blind map firing, the creation of artillery grids during World War I was as much about the way that maps were used as the way they were designed. During the war, not only were maps printed by the millions and used for the first time as a day-to-day tool by nonspecialists,32 but the advent of artillery grids put paper maps in a very different relationship to other forms of geographic knowledge. Calling the new artillery technique “map firing” is perhaps even slightly misleading, since it relied not on the map alone, but on several innovations at once, including the gridded map, a dense network of physical survey monuments, published lists of coordinates, and a large piece of metal called a plotting board. The map itself was only the public face of this wide-ranging system. Indeed, for the soldier in the trenches, changes in the paper map were arguably much less important than changes in the experience of geographic space itself. What mattered was not the miniature world of any particular map sheet, but the ability to experience the entire front as a single (two-dimensional) region with no internal boundaries and to use the same set of coordinates to inhabit both the miniature world and the real world at once, with priority given to the full-scale accuracy of the latter. The diminished importance of the paper map followed directly from the kind of geographic knowledge required for effective map firing. To aim a gun, three pieces of information are needed: the coordinates of the gun, the coordinates of the target, and the orientation of the gun with respect to the 134

Chapter Three

grid. Although a well-designed grid system is crucial, the map itself is only of limited help in providing these values. There are two main reasons why. First, paper is liable to deformation over time— by as much as 1 percent— and it is difficult to locate features with a precision greater than about 0.2 millimeters. These may not sound like large numbers, but if a map at a scale of 1:20,000 (a common scale in northeast France) were used to aim a gun at a target eight thousand meters away (the range of most French guns during the war) the shell could miss its target by almost ninety meters. The second issue is that even without these defects, actually determining the coordinates of a point is quite difficult from a map alone. Except for precise features like buildings or road intersections, connecting an unknown point in the world with a point on a map almost always requires some amount of new surveying, even when the map is entirely accurate. And especially when a gun is positioned in a concealed location without a view of faraway landmarks, the survey points used to connect it to the grid must be located with greater accuracy than is ever possible with a paper map— sometimes to within a fraction of a meter.33 To overcome these problems, the artillery map during World War I was supplemented by two other objects. The first object, which solved the problem of paper deformation, was the plotting board: it was simply a piece of zinc, reinforced by wood, that had been ruled with grid lines. (At times the paper map could be carefully cut into grid squares and pasted on the board as well.) The typical plotting board during World War I was about two feet on a side, but longer-range guns used larger boards— in the case of the boards used by the Germans for their forty-kilometer guns, up to four meters long. Plotting boards had been used by Germany before the war, but were a rather new technology; the British first encountered them in 1915.34 The second object, which addressed the problem of precision, was a pamphlet known as a “trig list.” As shown in figure 3.10, this is simply a printed list of the exact grid coordinates, accurate to ten centimeters, of physical beacons and monuments whose locations have been previously surveyed and noted on the map. (Trig is an abbreviation for trigonometric, which is how these locations were calculated.) These monuments can include existing landmarks like church steeples, towers, and windmills, but during the war most were instead newly constructed and placed at strategic points in the field, as was the wooden beacon shown in figure 3.11. With these objects close at hand, the actual aiming of a gun required the paper map only briefly. The gunner begins by using the map to identify nearby survey monuments, but the exact coordinates of a monument (or monuments) are read from the trig list, and the gun is connected to these monuments by a short, but precise, field survey. As shown in figure 3.12, the gunner then plots the location and orientation of the gun on the plotting board, along with the location of nearby trig points. Targets can then be located by several methods: by direct survey, by aerial reconnaissance, or by the new techniques of sound ranging and flash spotting, which could locate enemy guns using precise measurements of the sound or flash they emitted when fired. These different Aiming Guns, Recording Land, and Stitching Map to Territory

135

Figure 3.10: Page from a French trig list, issued April 1917. This table gives the coordinates of survey points on the système Lambert grid, accurate to ten centimeters. Soldiers used these points to locate their guns on the grid with much greater precision than would be possible using a map alone. “G. C. T. A.” is the Groupe de Canevas de Tir de l’Armée— the army group responsible for creating the “firing framework.” Reproduced from Peter Chasseaud, Artillery’s Astrologers: A History of British Survey & Mapping on the Western Front, 1914– 1918 (Lewes: Mapbooks, 1999), 514.

Figure 3.11: British surveyors measuring a captured German trig beacon after an advance, the coordinates of which would eventually be included in a trig list. While prominent landmarks like church steeples and factory chimneys were often preferred, these makeshift beacons were common in the countryside or in areas subject to heavy bombardment. Image from H. Winterbotham, “Geographical Work with the Army in France,” Geographical Journal 54 (July 1919): 13.

Figure 3.12: Before firing, a gunner must know his grid coordinates, but he must also know the angle of his gun with respect to the grid. This is not found using a map, but with a simple survey. Shown here is a somewhat complex example of this process, with all work done on a plotting board. The gun is located at P, and the only visible landmark is a tree along the line PR. In order to draw this line— known as the “aiming line”—the gunner starts with an overlapping sketch of all known points in the area. A and B are survey beacons that cannot be seen but whose coordinates are taken from a trig list; P and R are again the gun and the tree, and c is a temporary beacon set up by the gunner with a view of both the beacon B and the gun at P. By measuring angles at B, c, and P, the angle of PR can be firmly tied to the grid. The finished line PR is then drawn parallel to PR in a convenient location in the center of the board. Image enlarged from Manual for the Artillery Orientation Officer (translated from French original) (Washington DC: USGPO, 1917), 100.

coordinates are either received in full from other units or determined graphically on the board. Finally, the gunner simply measures the range and bearing to his target with a ruler and a protractor, which were often installed directly on the board as well. After making adjustments for elevation and weather conditions, the soldier aligns his gun and fires— often several rounds per minute.35 The actual work of aiming a gun was thus all done using trig lists and plotting board, since these gave much more reliable results than the map itself. As an American artillery general explained after the war, “The elaborate ‘Plan Directeur’ [the French large-scale map series of the western front], while desirable in every respect, is not essential. As a matter of fact, beyond furnishing within a reasonable degree of accuracy the difference of level between gun and targets, it is not necessary at all.” With the inclusion of altitude data in trig lists, even this last requirement could be eliminated.36 The grid, in other words, was crucial— but the map was vestigial. This relative displacement of the map by the plotting board and trig lists suggests two general conclusions about the geo-epistemology of the grid. First, what really mattered to the gunner were survey coordinates, accurate at full scale. Although locations could be found by referring to maps, directly surveyed locations were always preferred.37 What was most important for successful map firing was thus not necessarily accurate maps, but a greater density of precisely located beacons and monuments spread throughout the battlefield. This is why French surveyors described their work in 1914 and 1915 not as improving the maps themselves but as providing the crucial “canevas de points” for the front as a whole.38 And later in the war, this is largely what separated Allied and German surveying. Although German techniques of air survey and long-range bombardment were much better than those of the French or the British, the Germans’ lack of adequate trig data was a serious handicap. In one battle in the spring of 1917, for example, they correctly located only a third of Allied guns. The British, in contrast, located more than 90 percent of German positions. The success of the grid— or even, from the gunner’s point of view, the very existence of the grid— was established as part of the landscape, not as lines on paper.39 The second conclusion is again about the relationship between grids and the traditional representational logic of the map. As seen even with Cassini’s original system, grids are irreducibly hybrid, always oscillating between paper and world. The same coordinates used on the map also exist off the map, independent from any piece of paper, and the deformations of the underlying map projection operate at full scale as well. Even if a soldier never refers to a map at all, his gun may still miss its target due only to the mathematics of the grid’s projection. By extension, since Cassini’s coordinates and the système Lambert really are mathematically equivalent to a map at a scale of 1:1, the way that the grid is actually used on the front suggests that the gridded soldier is in fact inhabiting this full-scale map. Performing surveys, locating targets, coordinating with other units— these activities are all done using grid coordinates, but they do not require an actual map. 138

Chapter Three

This consideration of how grids work is what separates them from a representational project like the International Map of World. Although the grids of World War I were certainly a cartographic technology, they did not actually represent anything at all. Instead, they were a new kind of spatial infrastructure, overlaid and installed as a new way of inhabiting geographic space. (And although representation can certainly be powerful, it cannot hit a target eight kilometers away on the first try.) This is why analyzing grids requires shifting from purely cartographic questions to a wider-ranging concern with the politics of space. How do the calculational properties of grids influence the kinds of activities that take place on the ground? How do grids create boundaries? Who benefits from the creation of a grid, and who suffers? These are not questions about maps; they are questions about the world itself.

THE DOMESTIC POLITICS OF THE GRID: INTERWAR SYSTEMS IN THE UNITED STATES Immediately after World War I, all four major powers created systematic military grids for their homelands: the British and Americans in 1919, the French in 1920 (by extending the logic of the système Lambert throughout the country), the Germans in 1922.40 But over the next two decades, military grids came to be seen as quite useful for nonmilitary purposes as well, and grids were created not just in countries with war experience, but nearly everywhere that state-sponsored surveying took place. By the mid-1930s, conformal grids had been adopted by countries as diverse as the Netherlands, Switzerland, Belgium, Norway, Hungary, Egypt, Siam, and the USSR.41 Why would a mathematical technique developed for aiming guns be so widely embraced in peacetime, even in places far removed from the threat of trench warfare? The simple answer is that grids, wherever they were adopted, were a technology of territorial consolidation. Just as the système Lambert enabled the Allies to enforce seamless coordination all along the western front, civilian grids replaced a scattering of incompatible local surveys with a single national system (or in the US, one system per state) and helped to route all survey information through a centralized mapping agency. For example, unified national grids replaced the thirty-nine county-level systems of the UK and the fifty separate systems of the German Empire; the US state grids likewise replaced dozens of city-level coordinates.42 This spatial unification was matched at the administrative level as well, since each of these new systems required a wellmaintained and regularly cataloged landscape of grid-referenced monuments. And the goal of coordination was often explicit: the hope was that grid coordinates would be used not just by soldiers, but also by highway engineers, land surveyors, forest managers, police, private companies, and even lay citizens.43 Within this overall trend toward geographic and functional consolidation, each domestic grid has its own history. In every country, grids were pushed Aiming Guns, Recording Land, and Stitching Map to Territory

139

by different actors with different goals, and grids intersected differently with divergent traditions of land registration, large-scale public works, and political centralization. In short, there is no such thing as a typical domestic grid. That said, looking more closely at the particularities of one system— for convenience, that of the United States— helps open up an important set of questions about the politics and power relations of gridded space. In particular, how do grid systems operate as projects of state building? What is the relationship between grids and state power, and what kind of governmentality do they construct? The anthropologist James Scott has suggested that statesponsored projects of legibility— of which the map is the paradigm case— are typified by a drive toward the kind of overcentralization, simplification, and will to control that doomed so many of the technocratic, high-modernist projects of the twentieth century.44 One might thus expect grids to follow a similar pattern, with their mathematical precision being wielded by the state as part of new forms of taxation, surveillance, or the militarization of everyday life. But even though the American experience was hardly universal, the history of grids in the US makes it clear that this interpretation is too simple. In particular, the interwar grids of the United States offer two important lessons. The first is that grids do not in fact channel the interests of “the state” as a monolithic, power-hungry force. Even though all grids are maintained by state agencies, the creation of grids has as much to do with the professional ambitions of certain surveyors and engineers as it does with state power in general. State building, in other words, is not just pursued by the government. The second lesson is that even though grids do channel certain high-modernist ideals of rationality, accuracy, and simplification, their abstract Euclidean logic was not what most impressed the people who actually designed and used the new systems. Instead, grids were seen primarily as a new technology of trust, authority, and permanence that differed markedly from traditional surveying or paper mapping. Rather than centralizing information into the hands of the few (as with maps), grids were a way to make geographic knowledge widely accessible and to put everyone— tax man and farmer alike— on a level playing field. So while grids do certainly have a politics, they are more about the stabilization of spatial relationships than the expansion of state power, and grids end up constraining strategies of domination just as much as they do those of resistance. The US civilian grid system— officially known as the State Plane Coordinate System— was designed in 1933 and 1934 by Oscar Adams, an Ohio-born-andeducated mathematician at the US Coast and Geodetic Survey (the federal agency responsible for high-precision survey work) who had designed the main homeland military grid for the US Army just after World War I and had published widely on map projections. The most important feature of the State Plane system is that it gives each state its own separate grid system. Adams worked out the mathematics for each state individually, and the legislation that gave the grids official recognition was also enacted on a state-by-state basis 140

Chapter Three

Figure 3.13: The State Plane Coordinate System subdivided the United States into a patchwork of distinct grids. Some states are large enough to require multiple grids, but note that no grid boundaries cross state (or county) borders. Light gray zones use a Lambert projection similar to the original système Lambert; dark gray zones use the Transverse Mercator. (Texas, for example, is divided into five east-west Lambert grids; New Hampshire is covered by a single north-south Transverse Mercator grid.) The boundaries here are from the late 1960s, but they are quite similar to Oscar Adams’s original design in the 1930s. Map from Alden Colvocoresses, “A Unified Plane Coordinate Reference System,” Surveying and Mapping 27 (Dec 1967): 622; shading added.

over the next few decades.45 When designing the grids, Adams’s main concern was to match the size and shape of each state with a suitable conformal projection. He used two projections in particular: the Lambert, already well known from the war, and the Transverse Mercator, which was another name for the equally-well-known Gauss projection. The Lambert was useful for areas of east-west extension, while the Transverse Mercator was best for north-south areas. (The Transverse Mercator simply takes the classic Mercator projection and rotates it 90 degrees. So just as the Mercator shows the equator without any distortion, the Transverse Mercator does the same for a particular northsouth line of longitude.) Figure 3.13 shows how Adams applied this logic to the country as a whole. For a few small states like Massachusetts or Vermont, only one grid was needed. But in order to keep distortion low— that is, in order to maintain the mathematical fiction of a flat earth for surveyors in the field— most states had to be split up into several different zones.46 The State Plane system was undoubtedly a project of spatial unification and increased legibility. If nothing else, the boundaries of the State Plane grid zones clearly reinforced the political geography of states and counties. EveryAiming Guns, Recording Land, and Stitching Map to Territory

141

day use of the grid coordinates would thus ensure both coordination and rational description throughout the state. Articles explaining the benefits of State Plane called it a “common language,” where local descriptions would be replaced by those legible to outsiders. “The maple tree sixty feet south of Mrs. Smith’s barn” would become “the maple tree at 432342, 12435,” and new highways— especially large, state-funded highways— could be connected to old ones at less cost and with less confusion.47 This spatial logic was reinforced at the legal level. There were a number of reasons why individual states were motivated to pass legislation recognizing the system, the main one being simply to give State Plane coordinates some amount of official codification. But in addition, most articles on the State Plane system in the 1930s coupled a description of the benefits of spatial coordination with a call for the creation of a centralized survey and mapping bureau for each state— an unprecedented move. In 1938, for example, an official committee in New York State argued that “the adoption of a State plane Coordinate system makes it imperative that a central clearing house or authority be available to which matters of local policy and controversial questions can be referred.” As State Plane legislation began to be passed in the late 1930s, legal recognition of the grid generally went hand in hand with recognition of exactly these kinds of agencies.48 The creation of State Plane coordinates was intimately linked with other state-building projects of the 1930s as well. As with all grids, actually making the State Plane system useful for everyday surveyors required not just clever mathematics but also a network of monuments or beacons on the ground. The monuments that were already available from the country’s high-precision triangulation projects were almost useless for local purposes, since these were widely spaced and often located on inaccessible hilltops, with coordinates recorded in latitude and longitude. Establishing a grid thus meant filling in the gaps between these triangulation stations with many more points— a conveniently located monument every five miles was the typical goal— and issuing long lists of their precise grid locations. During the 1930s, much of this work was undertaken as unemployment aid, with funds from the Emergency Relief, Civil Works, and Works Progress Administrations all used to place markers like the one shown in figure 3.14.49 In addition to this state sponsorship, government agencies were also some of the most enthusiastic users of the new grids. One of the earliest adopters was the Tennessee Valley Authority, which used them to describe property purchased by the government, but they were also used by agencies as diverse as the Army Corps of Engineers, the Port of New York Authority, and individual municipalities.50 But as much as the State Plane system was aligned with state building, it would be difficult to describe it as a case of an activist state extending its power/ knowledge systems into the daily lives of individuals. If nothing else, the legislatures that passed State Plane laws often had no idea what they were voting on. In 1946, for example, the Washington Evening Star ran the light-hearted headline 142

Chapter Three

Figure 3.14: A survey monument embedded in the sidewalk just outside the Mt. Vernon courthouse in northeastern Texas. Similar brass markers were placed in convenient locations throughout the country to help tie local surveys to the coordinated grids of the State Plane system. Flickr photo by QuesterMark, 2008.

“Virginia Legislators Don’t Understand Plane Coordinate Bill but Pass It Just the Same” and quoted a delegate joking that the bill contained “a formula . . . by which you can compute the national debt.”51 The people pushing hardest for the adoption of statewide grid systems were in fact not central to the legislative system at all, but were instead individual engineers and surveyors with somewhat tenuous relationships to the state. For example, when Adams presented his grids, he consistently noted that he designed them only “in response to the demand from the engineers of the country.” The first grid he designed was for North Carolina, after a highway engineer contacted him for help coordinating road surveys.52 Most articles on grids were likewise written by professors and engineers in private practice, not government officials. The most vocal proponent of domestic grids— by far— was in fact an assistant professor of civil engineering at Princeton named Philip Kissam. Kissam was primarily an academic, and he wrote endlessly about the virtues of grids in professional journals. But he was also an activist expert who led the unemploymentrelief surveying work for New Jersey, personally wrote the grid legislation for the state, and pestered the federal government to recommend the system to others. Aiming Guns, Recording Land, and Stitching Map to Territory

143

(The New Jersey grid was the second that Adams designed.)53 The deep involvement of someone like Kissam in the State Plane system suggests that the project should not be understood as a top-down project of the state itself, but more as a project of professionalization that made use of the state. For Kissam, the relevant power struggle was not between the central state and individual citizens, but between ambitious engineers like himself and what he tended to describe as the “local” or “average” engineer. For example, in articles for the American Bar Association and the National Association of Real Estate Exchanges, Kissam presented himself as a champion for the values of efficiency and transparency; his goal was to save the taxpayer and property owner the waste and confusion of obsolete survey practices. He noted that local engineers often “resort to private marks to make it impossible for other surveyors to reap the rewards of their expenditures,” a system which inevitably leads to “expensive boundary litigation as well as costly errors in the location of buildings”—all of which could be avoided by defining boundaries using grid coordinates.54 In articles to his peers, he likewise attacked the conservatism and make-work of local surveyors not just as a drag on society, but on the status of the surveying profession as a whole.55 And there is ample evidence that local surveyors did indeed see the new grids systems as a threat to their skills and their income. Just the sheer number of introductory articles on the State Plane system in workaday journals—several per year throughout the 1930s and 1940s— suggests that grids were not quickly embraced by mainstream practitioners. Even well into the 1950s, high-level engineers noted that grids were “often received with suspicion and mistrust by old-time surveyors.”56 In other words, rising experts like Kissam did not see their engagement with the state apparatus as an end in itself; it was also a means to advance their own professional project. The reason why the blurring of state and nonstate interests is important here is because for Kissam and his allies, the crucial benefit of grids was not that they expanded the role of the state or opened up local knowledge to faraway eyes, but that they could eliminate wasteful spending by redefining the idea of geographic permanence. For most American surveyors, the greater part of their time was spent trying to find and repair old property markers, since the physical monuments were what established the legal boundary and a lost or disturbed monument could lead to major legal headaches. But a grid would render much of this work obsolete, since monuments defined in relation to a nearby grid reference could easily be corrected as needed. As a government report on the State Plane system argued, grid coordinates “will hold fixed the position of the boundary point on the ground with greater fidelity and permanence than the most massive monument of rock or metal.” Even if the grid monuments themselves were disturbed, nearby markers would still remain as witnesses to the original location.57 Kissam was particularly effusive, describing such monuments— in strong italics— as “permanent well-known indestructible points of reference,” but adjectives like “permanent,” “reproducible,” and “indestructible” were almost cliché in American writing about grids.58 The ba144

Chapter Three

sic argument was thus that a governmental system— where survey data would be safeguarded and published, and lost stones continually replaced— would outlast any physical artifact and therefore increase efficiency.59 The larger implication, however, was that grids were a direct challenge to the traditional state apparatus of geographic trust. Indeed, the purely geometric logic of grids was a remarkable departure from existing strategies for making the world legible. For centuries, the best way to stabilize a geographic reference point was to provide several overlapping descriptions that together gave an overabundance of evidence. The historian of science Graham Burnett has identified this tactic as the classical rhetorical strategy of consummatio (“a heaping up of arguments to make a single point”)— the idea being that the combined worth of a map, field notes, written description, and sketches was much greater than any individually.60 For example, the surveying of a new international boundary following the 1848 Mexican-American War did not just consist of following a line of latitude through the desert and periodically erecting stone and iron markers, since it was clear that such monuments would continually deteriorate or, more likely, be intentionally destroyed by Indians. The official reports on the boundary were instead filled with many kinds of evidence, including naturalistic drawings of the monuments, wide-ranging topographic maps, and a detailed narrative of travel and exploration. No single type of evidence on its own would be enough to establish the boundary, but slippages in one description would made good by the strengths of others.61 Defining a boundary with grid coordinates is a complete reversal of this strategy. Rather than creating stability through an overabundance of evidence, stability is instead established with a single, authoritative reference that admits no ambiguity and cannot be contradicted by other forms of description. Likewise, grid coordinates are seen as permanent because they are defined in terms of an easily reproducible process rather than in terms of physical artifacts or historical claims. And again, a grid does not operate using the representational— and therefore fallible— logic of a paper map. It is instead a full-scale and fully self-contained system that, when adequately maintained, promises to supersede the messiness of the physical terrain with a purely geometric lattice of points. Is this a case of new forms of legibility making possible new forms of topdown power and state control? Not really. Certainly, much here works to the benefit of the state. It becomes much more difficult for landowners (or Indians) to resist state oversight, and local cadastral or engineering information becomes more accessible to state and federal agencies. But grid coordinates can just as easily be used against the state as well. Kissam, for example, argued that grids were an excellent way to for individuals to resist governmental overreach or incompetence. He explained how grids could help reestablish coastal property boundaries after a major storm and how grid-defined county boundaries could resolve problems of ambiguous taxation.62 The creation of the American grid system, in other words, was not an attempt to tip the existing scales of Aiming Guns, Recording Land, and Stitching Map to Territory

145

power. It was instead a rearrangement of power, an attempt to keep everyone honest. Or rather, it was a strategy of displacing governmental power from individuals and historical documents onto a purely mathematical system. In sum, perhaps the most important governmental effect of the State Plane system— or of similar domestic systems in other countries— was simply the reinforcement of existing forms of territoriality. Indeed, the great dream was to render it permanent. This was true spatially, administratively, and legally: grid boundaries aligned with political borders, every grid required a statesponsored clearinghouse of survey data, and using grid coordinates bolstered traditional jurisdictional divisions and stable property rights. But this was a territoriality different from that of the map. Instead of fostering the kind of territorial control typical of high modernism— centralized, removed from local conditions, accessible only to a select few— grids worked by creating a territorial framework that would be widely accessible by everyday people, used for everyday projects, and seamlessly enforce mutual coordination between local, regional, and national space. By no means does this mean that grids are benign, however. Compared to a map, domestic grids are in many respects a much more radical proposition: rather than being a tool of state territoriality operating from afar, they instead install that territoriality into the everyday fabric of geographic space.

THE INTERNATIONAL POLITICS OF THE GRID: INTERWAR DEBATES ABOUT SCIENCE AND SOVEREIGNTY At the same time that military grids were adapted for domestic civilian uses, they also became the subject of international discussion and debate, and the 1920s and 1930s saw many proposals for transnational collaboration and standardization, especially from the French and the Americans. These proposals were not restricted to topics like survey monuments or conformal map projections, either; they very quickly opened up to some of the largest problems in geodesy, including scientific questions about the size and shape of the earth and the possibility of combining the survey systems of different countries into a single whole. These ideas engaged some of the most prominent cartographers, geodesists, and survey officials of the time— both military and civilian— and raised crucial questions about territoriality, sovereignty, and transnational exchange. The main lesson of these projects, however, is that during the interwar period the spatial politics of grids had very clear limits. Almost all proposals that threatened to upstage national autonomy— or, even more radically, directly span international borders— were failures. Even the most successful of them had relatively little direct transnational impact. But these failures still performed important work, since they clarified and policed the boundaries of national territoriality and set crucial precedents for the global projects pursued during and after World War II— especially those of the US military, which often directly contin146

Chapter Three

ued earlier plans. In other words, the simple fact that these interwar projects failed is much less important than the particularities of how and why they failed. The most prominent venue for these discussions was the geodesy section of the International Union of Geodesy and Geophysics (IUGG), the postArmistice successor to an earlier German-inflected organization known as the Internationale Erdmessung, support for which did not survive the war. The interwar IUGG was an archetypical scientific organization: although it did have permanent officers, it was essentially a traveling conference series, with meetings (conducted mostly in French) every three years in a different major city. And similar to other interwar sites of scholarly exchange, at the IUGG the scientific and the political were thoroughly entwined: membership was organized by country— with Germany and Austria purposefully excluded— and countries were usually represented by high-ranking (and scientifically prominent) staff from national survey departments. The business of the IUGG generally consisted of sharing information and issuing official “wishes [vœux] and resolutions” to try to steer international work. Most decisions were made by consensus, but in theory each participating country had one vote.63 The topics of map projections, grids, and survey systems were discussed almost continuously from the first meeting, in Rome in 1922, to the seventh meeting, in Washington DC in 1939, and the IUGG passed several resolutions relating to the international standardization of coordinate systems. The recurrent political theme that most guided IUGG debates was a crucial separation between two kinds of work. As explained in 1930 by William Bowie, a world-renowned American scientist-engineer at the US Coast and Geodetic Survey and later president of several national and international societies (including the IUGG), there were two parts to geodesy. First was a “practical phase,” where “almost every civilized nation in the world” conducted the “regular work of surveying, mapping, and charting.” But then there was a secondary— and from his point of view, more important— “scientific phase” devoted to “problems beyond the scope of the work of any single country.” These “truly international” questions included the determination of the size and shape of the earth, studies of the earth’s gravity, and various problems in continental geology.64 This idea of a one-way relationship between a nationalpractical sphere and an international-scientific sphere is why the IUGG mostly just issued international standards rather than itself directly sponsoring scientific projects. The hope was that the right international agreements could inform the methods and priorities of each individual country’s work and thereby coax scientific data out of practical surveys. But, crucially, this rubric also implied that the international sphere should never impinge upon national autonomy and should never have its own practical ramifications. The question, then, is how this divide between national and international was actually determined and how seriously it was regarded in practice. After all, this could easily be a case of scientists giving abundant verbal homage to national sovereignty while quietly pursuing their interests wherever they Aiming Guns, Recording Land, and Stitching Map to Territory

147

might lead. There were four IUGG projects in particular that help answer this question, all with direct links to grids and territoriality: two in the early 1920s and two in the 1930s. Analyzing these projects makes it clear that interwar scientific internationalism was not just rhetorical. It was instead very serious indeed: not only could it shut down debate and remove a proposal from meaningful consideration, but it was also installed directly into the mathematics of the projects that did move forward. Scientist-engineers like Bowie had a dual allegiance to their country and to science, and their defense of national interests against international interference was genuine— and enforceable. The following subsections analyze each of these four projects in turn. The first two emerged directly out of the geodetic problems faced during World War I: one was a proposal for a universal procedure for making national grids, and the other was an international agreement about the size and shape of the earth. The two projects of the 1930s moved away from this grand internationalism and focused more tightly on continental integration: the first was a proposal to consolidate all the survey networks of Europe, and the second was a grid system that could unify all of Africa. Of these four, only the second might be considered even modestly successful on its own terms, but all of the proposals were subject to vigorous debate and, through their failure, successfully kept national territoriality safe from international intervention.

1: Roussilhe’s International Grids At the first meeting of the IUGG in 1922, a French hydrographic engineer named Henri Roussilhe made a bold suggestion that the international geodetic community should endorse a standard template for creating national grid systems. His proposal was motivated by a mix of straightforward international altruism and Bowie’s linear model: not only would an internationally standardized system help less-well-equipped countries by eliminating the need to work out the complicated math of conformal projections from scratch, but it would also be a useful tool for making high-precision coordinates scientifically comparable from country to country. Roussilhe was perfectly situated to head such a project. He had worked alongside Courtier in the survey of Madagascar and helped to develop the système Lambert in 1915; after World War I, he was given charge of the reconstruction of the tax cadastre in northeast France. But after pushing for his scheme in both 1922 and again in 1924, it met a rather anticlimactic end in 1927, when it was officially endorsed in an almost unrecognizably watered-down form. Roussilhe’s scheme was unquestionably rooted in concerns of national territorial consolidation— it was, in fact, exactly the scheme that he had developed for repairing the French cadastre. It worked much like the grid systems of World War I or the State Plane system, except that instead of using a Lambert or Transverse Mercator projection, which are useful for areas of east-west and north-south extension, it used a stereographic projection that excelled 148

Chapter Three

Figure 3.15: Henri Roussilhe’s 1922 scheme for an international grid system. The circles— each 1,120 kilometers in diameter— show the possible coverage of grids based on the stereographic projection. Roussilhe proposed that the equations for these grids be standardized and precomputed so that each country could adopt one of these grids for its territory. Note that countries like France and the UK (including Ireland) would have their own grid, but some grids would be shared between countries. Sweden and Norway would have to use two grids. Image from H. Roussilhe, “Emploi des coordonnées rectangulaires stéréographiques pour le calcul de la triangulation dans un rayon de 560 kilomètres autour de l’origine,” Travaux de l’Association internationale de géodésie 1 (Paris, 1923 [presented 1922]); shading added.

at mapping a circular area. (Today, the stereographic is the main projection used for mapping the poles.)65 The mathematics of applying the stereographic projection to a country like France were remarkably complicated, but Roussilhe found that by increasing the sophistication of his calculations, the stereographic could give twice the precision of other projections, and he could in fact cover all of France with just one unified grid.66 Roussilhe was highly critical of the fragmented state of French cartography— at the time, there were four different survey systems in use, which together required twenty separate projections— and his great dream was to “replace all the[se] systems . . . with a single system of conformal coordinates” that would enforce spatial coordination throughout the country.67 In turn, it was exactly the usefulness of the stereographic for “un grand pays” like France that prompted Roussilhe to recommend it as an international solution. Figure 3.15 shows how his circular grid zones could be copied Aiming Guns, Recording Land, and Stitching Map to Territory

149

throughout Europe and North Africa: while Sweden and Norway would have to use two zones, every other country could be fully coordinated with only one. As he put it, even though the problem of grids was largely “a question of purely national interest, it will certainly be perfected if one undertakes a general study of the question and if one is concerned above all with using international tables of calculation.” That is, once the math had been worked out for a particular latitude, the same values could apply around the world, and the IUGG was the perfect organization to endorse and publish such tables. Roussilhe even suggested that the tables could be coordinated with the sheets of the International Map of the World.68 From the point of view of national sovereignty, however, this was a somewhat ambivalent solution. The concern here was explicitly national, but not every country received its own grid, and certain small countries would have to share a grid with their neighbors. For example, Roussilhe’s map suggested that for the sake of international efficiency, a single grid would be shared by Greece, Turkey, and Bulgaria, and likewise by Poland, Czechoslovakia, and Hungary. For practical cartographic purposes, this meant that these countries would essentially share a single territorial space, with easy-to-use coordinates extending seamlessly across international borders. The reception of Roussilhe’s scheme by geodesists from other countries shows that even this small mixing of the practical and the international was a step too far. When his proposal finally came up for a vote in 1927, there were two main objections to his plan. First was the general worry that grid systems and cadastral mapping were simply too practical to warrant any discussion at all. Roussilhe had himself acknowledged that map projections were “more of practical than purely scientific interest,”69 but the reaction from others was rather more direct. For example, the Swedish delegate, Karl Rosén (a professor in Stockholm and the official representative of the Swedish national mapping office), suggested that even the word “cadastral” should be eliminated from the text of the proposal, since the IUGG “must not deal with questions which are more of a practical than scientific order.”70 The more serious objection, however, was that a circular stereographic grid might not be appropriate for every country. Responding again to criticism from Rosén, whose country would have to be split into two grids under the plan, Roussilhe was forced to admit that “the shape of the territory” was indeed crucial, and that his scheme failed entirely for very large countries like Canada.71 As a result of these objections, the resolution that finally garnered support was hardly what Roussilhe had wanted. Instead of recommending a specific mathematical solution for use in all countries, it simply called for each country to adopt “a system of extended coordinates, unique for each country. . . . When the size and shape of a country do not permit the adoption of a single system, the use of a minimum number of systems is recommended.” The practicalities were left for each country to decide on its own, and Roussilhe’s detailed mathematical tables were never endorsed by the IUGG.72 This sentiment was only reinforced in subsequent meetings. When in 1933, for example, 150

Chapter Three

the British delegate Arthur Hinks suggested that the IUGG might undertake the publication of tables for the Transverse Mercator projection, the unanimous decision was that such a project would not be, “properly speaking, of an international order.”73

2: The International Ellipsoid Rousshile’s grid scheme titled heavily toward the practical, but it was very closely linked to an allied project of the highest scientific order— that of the size and shape of the earth. The precise dimensions of the earth had been studied since at least ancient Greece, and with special interest ever since Isaac Newton had argued that the spinning earth would not be a perfect sphere but would instead bulge outward near the equator to form an ellipsoid. And in the two hundred years after Newton’s calculations, many precise measurements had been made of the so-called “figure of the earth”—in Europe, South Asia, Africa, and the Americas— in order to give exact values to the earth’s radius and its deviation from sphericity. As shown in figure 3.16, there began to be general agreement among these measurements from the late nineteenth century.74 However, there was no consensus about which particular results should count as the most authoritative, and so in 1922 a proposal was made by three countries together— France, Belgium, and the United States— that the time had come to vote on standard values that could be accepted around the world. Unlike Roussilhe’s grids, this proposal was acted upon quite quickly, and in 1924 the IUGG officially endorsed what is still known as the International Ellipsoid, with a radius of 6,378,388 meters, flattened at the poles by a factor of exactly 1/297. The  International Ellipsoid, however, was not adopted without controversy, and the debates that took place show just how complex the exchange between scientific and practical work could be. The scientific value of having an International Ellipsoid was relatively obvious: if geodesists around the world could agree to use the same values in their calculations, then it would be much easier to collaborate on a wide array of unresolved problems, including those of the earth’s gravity field, its internal structure, and even its geological history.75 But values for the size and shape of the earth were not just of scientific value. They were also a crucial component of any national coordinate system. In order to perform a survey that takes into account the curvature of the earth— that is, any survey that covers an area larger than about five miles in radius— it is imperative to know exactly how much the earth is curved. Thus every national survey must use specific values for the size of the earth’s ellipsoid. In the nineteenth century, however, nearly every country had adopted a different set of values, often based on measurements from its own territory. So the problem was not just that individual geodesists were using different values in their scientific work; it was that scientists were constrained by the particularities of each country’s practical surveys. Aiming Guns, Recording Land, and Stitching Map to Territory

151

Figure 3.16: Values for the size and shape of the earth, 1800– 2000. Because the earth is flattened at the poles, making it an ellipsoid rather than a perfect sphere, both values are necessary. Note the smaller spread of results after 1880, which helped justify the 1922 proposal to agree on standard values for an International Ellipsoid. The values ultimately chosen were those determined in the United States by John Hayford in 1909. Large squares show the values that have been used by national mapping agencies to compute their surveys. (See note 74 for sources. For raw data, see www.afterthemap.info.)

Figure 3.17: Junction of the national triangulation surveys of France and Belgium before World War I. Precise angles are measured between the points in each network, as shown with connecting lines. (The points are often church steeples, since they are easy to see from far away.) These various angles, combined with at least one precise length measurement on the ground, give coordinates for every point in the system. The difficulty, however, was that the French and Belgian surveys had found different latitude and longitude values for the two shared points, and it was impossible to stitch the two networks together. There were several reasons for this mismatch, but one important problem was that the origin of the French survey was in Paris, the origin of the Belgian survey was in Brussels, and both countries used different values for the size and shape of the earth in their calculations. Even if the triangulation had been perfect, the French distance from Paris to Brussels would have been about one hundred meters greater than the Belgians’ result. Map adapted from Report on Survey on the Western Front, 1914– 1918 (London: HMSO, 1920), appendix I, diagram 1. (Only the primary French triangulation is shown; distance calculated using Vincenty’s method.)

Perhaps the most immediate drawback to having each country use a different ellipsoid was that it became impossible to stitch together high-precision triangulation networks across international borders; indeed, this is the problem that seems to have most animated the sponsors of the initial proposal. But yet again this was both a scientific and a practical problem at once. In 1922, for example, the eminent French military geodesist Georges Perrier recounted the “terrible situation” shown in figure 3.17, where the French and Belgian triangulations shared points in common but could not be joined because each country had used a different ellipsoid. He argued that this was a scientific problem that international standardization would have prevented: “It might be naively said with some exaggeration that had the round figure of 1/300 [for the flattening of the poles] been permanently adopted some 150 years ago, a great service would have been done to Science and much of the present-day Aiming Guns, Recording Land, and Stitching Map to Territory

153

confusion would have been avoided.”76 On the side of practicality, however, the main problem was that neighboring countries would disagree about the precise location of international boundary markers, since the same physical monuments would have different latitude and longitude values when surveyed using different ellipsoids. This was the problem stressed by the United States, which had recently convinced Canada and Mexico to connect their triangulation systems into a single North American whole. William Bowie made the case very clear, stating that “it is desirable that, for each continent, a single [ellipsoid] be adopted” in order that “discrepancies in latitude and longitude of the international boundaries are eliminated.”77 A similar case was made by Roussilhe as well, who saw the International Ellipsoid as an integral part of his grid project. In Europe, coordinates measured by neighboring countries could differ by as much as seven hundred meters. The choice of ellipsoid was not the only reason for these misalignments, but it was a crucial factor.78 If international standardization would lead to both scientific and practical benefits, why the controversy? There were at least two reasons. The first, which was only implicit in the public debates, was that stitching together neighboring countries’ triangulation systems would— similar to Roussilhe’s grids— essentially constitute a territorial merger, at least from a geodetic point of view. This could be of great scientific benefit, but it could also benefit an enemy’s artillery. The mismatch between French, Belgian, and German coordinates, after all, had been a major cartographic problem in World War I. The second problem was more straightforward: changing ellipsoids could be a very costly endeavor, since all prior work would have to be recalculated— by hand. The British delegation addressed both of these concerns head on, declaring publicly that the IUGG should have “no desire to impose a change of ellipsoid upon any Geodetic or Cartographic Service now in operation, or to suggest that any country should recalculate its whole triangulation.” As one delegate explained in private, the logic was simple: no country should adopt standards that might “give their rivals an undue advantage.”79 International benefit should never trump national autonomy. These objections forced the International Ellipsoid proposal to retreat from any hint of practicality. As the debate evolved toward a final vote in 1924, the original idea of having one ellipsoid per continent, as suggested by Bowie and officially endorsed in 1922, was dropped in favor of having only one global value. This was explicitly meant as a way to move the debate from the realm of national territory to that of international science.80 Even the precise values that were eventually chosen had political resonance. Both were taken from a 1909 treatise by the American geodesist-engineer John Hayford, who had made a new determination of the figure of the earth using triangulation data from the United States. Scientifically, there was much to recommend Hayford’s methods; Perrier in particular was a great admirer. But Hayford’s values were also attractive simply because they were new enough that they were not yet being used in any national surveys (not even those of the United States), 154

Chapter Three

and therefore no country would see its existing system chosen as an international standard.81 In other words, the successful vote for the International Ellipsoid also successfully kept it away from the national-practical sphere and confined it to the international-scientific domain— or, as the British happily put it, to “results of general interest.”82 And indeed, even though the ellipsoid would soon become an important part of international gravity work, there was almost no way that it could become useful for stitching together preexisting national surveys. All the IUGG could do was publish a set of precalculated tables and encourage their voluntary use, which was hardly a coherent strategy for overcoming cross-border coordination problems in countries with well-established surveys. And in the end the IUGG did not even go this far, as the official recommendation was only that the tables might be useful in “countries recently opened to Geodesy” where no high-precision survey had been done at all.83 Before the advent of electronic computers, only two countries— Portugal and the USSR— ever seem to have changed ellipsoids, and both were explicitly projects of national consolidation rather than international cooperation.84

3: Bowie’s Recalculation of the European Triangulation Both of the projects of the early 1920s had emerged directly out of the practical experience of World War I. The two projects of the 1930s instead had entirely interwar origins, but they continued the familiar tension between the national and the international. The first of these was a proposal with thoroughly American roots. In 1927, the US Coast and Geodetic Survey had successfully recalculated the entire North American triangulation (although without a change of ellipsoid), thereby reducing coordinate errors throughout the continent.85 At the 1930 IUGG meeting in Stockholm, William Bowie suggested that a similar project be undertaken for Europe— that is, to treat all European triangulation as part of one large continuous network. As preposterous as this may have seemed to countries without the calculational capacity of the United States, Bowie’s plan could not be ignored, and it was quickly endorsed, once again, by Georges Perrier. Even though no actual calculations could be carried out before the outbreak of World War II, the planning that took place over the course of the 1930s shows an even more radical and irreconcilable separation of the national-practical and international-scientific than found in either of the earlier projects. Bowie’s proposal clearly did have scientific merit. At the 1933 meeting in Lisbon, for example, all delegates agreed the project was of great “theoretical interest” for the determination of the best-fit ellipsoid for Europe and that it could address similarly “fundamental” problems in gravity as well. As summarized three years later, the project was seen to have “a strictly scientific goal.”86 The obvious practical problem, of course, was that a full and rigorous recalculation would effectively usurp all European countries’ existing survey work. It Aiming Guns, Recording Land, and Stitching Map to Territory

155

would stitch together every country’s coordinate system and make it easy to put the survey points of potential enemies onto the same grid. This, however, was not actually much of a pressing concern, since a full recalculation was completely out of the question. Given the size of the European triangulation network, the amount of labor required to solve a system of several thousand simultaneous equations would be outrageous. But here is where the distinction between the practical and scientific really gained traction. Although early discussions of Bowie’s plan assumed that the European triangulation network would have to be treated in the traditional, computationally intensive manner,87 counterproposals soon emerged that would yield enough precision be to give important scientific results without having to compute precise coordinates for every survey point in Europe. As the Finnish geodesist Ilmari Bonsdorff put it in 1933, the main task of the project would not be to perform heroic amounts of brute-force math, but “to perfect a simple method” for the calculations— one that might even be an important scientific contribution in its own right.88 The outline of such a method could in fact be found in the work just performed under Bowie’s supervision in the United States— known to this day as the “Bowie method.” Rather than treating a triangulation network either as a series of strong primary axes (as in France) or as a diffuse mesh covering the entire territory (as in most other European countries), the Bowie method saw triangulation as a series of loops, as shown in figure 3.18. Its great virtue was its computational efficiency: the adjustment of the western United States, for example, required only ten (human) computers working for fifteen months, which was substantial but not prohibitive.89 While the IUGG never reached an official decision about the method that should be employed for Europe, the one that gained preliminary endorsement took this basic logic and removed it quite decisively from the kind of practicalities found in the US. Developed by Niels Nørlund (director of the geodetic institute of Denmark) and James de Graaff Hunter (former director of geodesy for the Survey of India), this scheme would have reduced the triangulation of Europe to a grid that was similar in appearance to a Bowie net— see figure 3.19— but that in fact only considered points spaced roughly every two hundred kilometers or so. The mathematics would also be simplified: the calculation would start just with four main rectangles (the thick lines in the figure), each of which would then be subdivided and iterated in turn. This meant that accuracy at the level of individual triangulation stations would be sacrificed for the sake of accuracy at a continental scale, but since the main astronomical observatories of Europe would end up being located rather well, the lesser inaccuracies were not seen as a major problem. And the calculational payoff was huge: the initial work would be reduced to a system of as few as seventyfour equations.90 In other words, if this plan been carried through— as it may well have been if war had not broken out— it would have created two parallel and mutually incompatible triangulation systems in Europe, corresponding to the differ156

Chapter Three

Figure 3.18: Schematic illustration of the “Bowie method” of adjusting a national triangulation survey. The original high-precision surveys cover almost the entire country, but William Bowie simplified the western part of the network into a series of loops. Before electronic computers, the mathematics of combining raw survey measurements was often incredibly daunting, and Bowie’s method was famous for its calculational efficiency. Image from Oscar Adams, The Bowie Method of Triangulation Adjustment, US Coast and Geodetic Survey special publication 159 (Washington DC: USGPO, 1930), 10.

Figure 3.19: Diagram showing how the national triangulation surveys of Europe could be consolidated and recalculated using a technique similar to the Bowie method. This abstract diagram does not show Europe in particular, but it does show how nine large national surveys— each roughly the size of a European country, shown separated by double lines— would be abstracted into a series of loops, each about two hundred kilometers square. This would greatly simplify the necessary calculations, but unlike Bowie’s calculations for the western US, the level of abstraction here would be so great that precise coordinates would only be available for certain key points. Image from N. E. Nörlund and J. de Graaff Hunter, “Note on the Adjustment of the European Network of Triangulation,” Bulletin géodésique 61 (1939): 269.

ent needs of practicality and science. National systems, each calculated to full precision on its own ellipsoid, would have remained unchanged and discontinuous across international borders. But the same data would also have been watered down to create an international system which was only useful for answering certain scientific questions. The latitude-longitude coordinates of the same monuments would have ended up different in each system. And although the national systems would have still provided the backbone for domestic grids (both military and civilian), the much more diffuse IUGG system would probably have been unable to support any kind of grid at all.

4: Tardi’s International Grid for Africa The final grid-related project brought to international attention during the interwar era was in some respects the most ambitious of all. At the 1936 meeting in Edinburgh, a French astronomer and military geodesist named Pierre Tardi— successor at the IUGG to the now-retired Roussilhe and a strong supporter of Georges Perrier— unveiled a potential grid system “applicable to the whole African continent.”91 But even though standardizing Africa would have been an enormous undertaking just on its own, his presentation evoked even greater plans still. Figure 3.20 shows how the system would have worked: instead of designing a separate grid for each country, Tardi simply divided the continent— or perhaps the entire world— into uniform north-south strips, six degrees wide, each of which would receive a grid based on a Transverse Mercator projection. (Tardi even numbered his zones from 1 to 60 to match the sheet layout of the International Map of the World.) At first glance, this proposal seems to ignore national territory altogether; indeed, it almost perfectly foreshadows the global system installed after World War II by the United States. Perhaps surprisingly, his design received official and immediate endorsement in 1936, although it remained understandably stagnant in the years to come.92 What is remarkable about this scheme, however, is not its hints of globalism, but rather the degree to which its global tendencies were overlooked or even outright suppressed. A close analysis in fact makes it clear that this was again a project primarily concerned with national— or rather, colonial— consolidation. The goal in particular was to strengthen the case for the British consolidation of East Africa. While this might seem an unusual cause for a French geodesist, Tardi shared the leadership of his IUGG subcommittee with Malcolm MacLeod, a British scientist-engineer who had been in charge of worldwide military mapping for the UK before being recently appointed the director of the civilian Ordnance Survey. Tardi’s scheme was clearly indebted to British precedent. Since the mid-1920s, British survey officers in England had been trying to create a centralized East African Survey Department to put its old and new colonies, recently won from Germany, on a single geodetic framework and a shared grid system.93 The specifics of projections and grid-zone boundaries, however, were a serious source of conflict between the British homeland 158

Chapter Three

Figure 3.20: Pierre Tardi’s illustration of a possible grid system to unify surveys and coordinates in Africa. Above is a view of his entire system, which would consist of a series of north-south grid zones six degrees wide, each based on a Transverse Mercator projection. (Shading has been added to one such zone.) Although Tardi’s diagram spans the entire world, the mathematics of his scheme were specifically calibrated for Africa and could not have been applied to any other continent. Note the numbers running along the bottom, which correspond to sheet divisions of the International Map of the World. At right is an enlarged detail of eastern Africa, showing borders between colonial powers rather than borders between individual colonies. Image from Pierre Tardi, “Étude d’un système de projection de Gauss en fuseaux de 6° d’amplitude, pouvant s’appliquer à l’ensemble du continent africain,” annex to his “Rapport sur les projections,” Travaux de l’Association internationale de géodésie 14 (1938 [presented 1936]): facing 24. The Gauss projection was the name used in Europe for the Transverse Mercator.

and the various colonial survey departments. In 1934, for example, MacLeod’s office had called for precisely a system of six-degree-wide Transverse Mercator grids that would be applied uniformly from Egypt to the Cape Colony, but this scheme was rejected by colonial surveyors who already had well-established systems that were better suited to their local needs.94 Having the international community endorse MacLeod’s preferred solution— especially as developed by a disinterested third party— could only help it advance. Indeed, Tardi’s seemingly grand scheme was tailored to the particularities of Africa both rhetorically and mathematically. Echoing the conventional wisdom of the time, Tardi stressed that projections and coordinate systems should always be suited to the size and shape of each country individually. He wrote that there is “no place for conceiving of an international system of projection,” and that— quoting one of the leaders of the French hydrographic survey in Madagascar— a country’s coordinate system should be “not a uniform of the clothing industry, but a custom-fit outfit.”95 He suggested that African countries might even adjust the boundaries of his grid zones as they saw fit. He likewise specifically exempted the French colonies of northwest Africa from his proposal, since they already had perfectly suitable grids. And although it may seem that this verbal rhetoric was entirely contradicted by the visual rhetoric of his global map, notice that his map does not show the boundaries of individual colonies in Africa, but only the boundaries between areas controlled by different European powers. If the eastern spine of Africa is considered as one country, eight thousand kilometers in extension, then a system of north-south grid zones is perhaps as custom an outfit as one could imagine. The particularities of Tardi’s mathematics tell a similar story. Because of the ellipsoidal shape of the earth, the calculation of the Transverse Mercator projection is notoriously complex, and the equations that were in general use in the 1930s, which prioritized accuracy at European latitudes, required substantial computation. Wanting to make his mathematics more compatible with mechanical calculation machines, Tardi developed equations that were markedly simpler. The catch, however, was that they would only give acceptable results for latitudes less than 36° away from the equator.96 So not only could his system not be applied to the entire world, or even the entirety of other continents— all of which extend beyond 36°—but it was not even a perfect fit for the whole of Africa, since the most populated parts of the French colonies of Algeria and Tunisia lie at roughly 37° north. Ultimately, the political legitimacy of Tardi’s proposal was justified on grounds similar to the International Ellipsoid more than a decade before. Rather than asking the IUGG to officially endorse a global scheme that would force its members to recalculate their national surveys, Tardi made it clear that his grids were only meant to apply to “the immense territories which have yet to be conquered by Geodesy,” where “no system of representation . . . has yet been defined.”97 (Evidently the preexisting coordinate systems of the separate colonies did not count.)98 In other words, in the late 1930s just as in the early 160

Chapter Three

1920s, a system of representation that could count as “defined” had to be one that would promote territorial consolidation under the direction of a centralized mapping agency on the European model. This was the true goal of Tardi’s seemingly grand plan. And given the mathematics, it in fact could not have been used for anything else. * Taken together, these four projects show a clear pattern developing over the course of the interwar period. Two points in particular are worth stressing. The first is simply about the less-than-effective internationalism of the IUGG, since relying on the politics of international resolutions, international tables, and the voluntary initiative of individual states put serious constraints on what could actually be accomplished across international borders. Granted, this system was not without its small successes. By 1930, for example, systems similar to Roussilhe’s stereographic grid had in fact been adopted in Syria, Lebanon, the Netherlands, Bulgaria, Romania, and Poland.99 Likewise, a 1936 inquiry found that the IUGG resolution on the International Ellipsoid had led to its use in China, Denmark, Egypt, Portugal, Romania, and Siam.100 But for the most part these countries did not share borders, and the serious disconnects between national systems did not lead to noticeable action. Indeed, the norm was for national agencies to ignore IUGG recommendations altogether. Only a few years after the vote on the International Ellipsoid, for example, the director of French military mapping (who did not participate in IUGG meetings) wrote to his British counterparts that “I don’t use the word ‘abandon’ with regard to the Hayford ellipsoid, but I don’t pursue ‘adoption.’” The British, for their part, continued to publish complaints about the ellipsoid well into the 1930s.101 And when Perrier sent official inquiries concerning the European recalculation to every country on the continent in 1937, he received cooperative replies from all but three agencies, but only after a year of delay and several reminders.102 Tardi’s scheme for Africa was likewise a dead letter in the colonies. The second point, however, is more specifically about interwar territoriality. The pervasive distinction between a national-practical realm and an international-scientific realm was not just an empty gesture designed to elevate the intellectual standing of geodesy. It was instead a very real— that is, mathematically consequential— separation between the territorial and the nonterritorial. The threat posed by the four IUGG proposals was that international considerations would be imposed not just upon individual scientists, but upon the domestic administration of tax cadastres, topographic mapping, and national defense. In order for a proposal to receive official endorsement, it either had to be emptied of any territorial specificity (as with Roussilhe’s grids, the International Ellipsoid, and the European triangulation) or it had to be so closely tailored to a particular domestic project that it was hardly international at all (as with Tardi’s African system). And ultimately there was an unbridgeable divide between the national, territorial use of Roussilhe’s and Tardi’s grids and Aiming Guns, Recording Land, and Stitching Map to Territory

161

the scientific value of the other two projects. The International Ellipsoid proved quite useful for coordinating global gravity measurements, and the European triangulation held the promise of doing the same. But neither could have led to continental consolidation or a transnational artillery grid.

CONCLUSION: GRIDS AND TERRITORY BEFORE WORLD WAR II Considering the early history of grids as a whole— from 1915 to 1939— perhaps the clearest theme is the simple trajectory that grids took from the artillery problems of trench warfare to the political concerns of territorial states. Before World War II, it was always assumed that grids would reinforce— or perhaps help create— a specific (and relatively static) political geography, and the more that grids could be matched one to one with a centralized jurisdiction, the better. The one possible exception is the original système Lambert, which was centered in Germany and used on some maps of non-French territory. But immediately after the war it was dismantled; its center was moved to the Paris meridian and the new coordinates were used only as a domestic system for northern France. During the interwar period, the most striking feature of grid systems— all grid systems— is their clean and definite boundaries: no overlap, no ambiguity, and no transnational reach. There are many reasons for this general pattern— related to mathematics, surveying practice, and administration— but ultimately all such reasons were closely intertwined and self-reinforcing. Mathematically, there was almost no point in wrestling with the equations needed for larger grids— such as those of Roussilhe or Tardi— when such grids had almost no chance of being adopted. There was also no political reason (or funding) to recalculate national survey networks to match neighboring systems, despite the interest of a few elite scientists. Instead, each country generally pursued the cheapest and simplest solutions necessary for its own domestic purposes, and nearly all surveyors and scientists pitched their own professional ambitions as part of these state projects. (The one major exception— the consolidation of North America by US geodesists— not only proves the rule, but shows that American precedent had relatively little impact on established practice elsewhere.) This national focus would shift quite dramatically during and after World War II— with the US taking a remarkably activist stance— but in the 1920s and 1930s there was no trajectory toward everlarger grid systems or eventual transnational consolidation. In other words, the national project was not only firmly entrenched, but it was also potentially quite stable. And indeed, almost all of the domestic grids created before World War II still exist today and still serve their original purpose. Grids were a new way of organizing space— one that took a step away from traditional mapping and toward a full-scale framework of points— but on their own they did not inherently destabilize existing territorial norms. They would have to be pushed. 162

Chapter Three

CH A PT E R F O UR

Territoriality without Borders: Global Grids and the Universal Transverse Mercator, 1940– 1965

Before World War II, the printing of grids on maps— and their installation in the world as networks of survey monuments— proceeded largely in line with political boundaries. Providing each political space with its own Euclidean coordinate system was a way to simplify engineering surveys, stabilize property records, and enforce coordination between all manner of different activities. Making these easy-to-use coordinates span international borders, however, was seen as an unacceptable violation of national autonomy. The Second World War was a decisive break. Suddenly the misalignment of coordinates between neighboring countries was a serious problem, and the focus turned from problems of national consolidation to problems of transnational or even global scope. But this transition was more than a simple scaling-up of what had come before. The visual technology of the grid remained exactly the same— just straight lines on large-scale maps— and the experience of gridded space remained largely unchanged. The territorial stakes, however, were new. Extending precise coordinates across boundaries presented a serious challenge to the national organization of geographic space. The first worldwide grid schemes appeared just months after the start of World War  II; they were created by several different countries, each with a slightly different design. None, however, were found entirely sufficient, and after the war mathematicians at the US Army Map Service designed and promulgated the global grid system that is still used today: the Universal Transverse Mercator, or UTM. The design of the grid was explicitly modeled after the sheet layout of the International Map of the World and was similar to earlier global proposals as well. As shown in figure 4.1, it consists of a series of sixty north-south belts each the width of one IMW sheet, plus circular areas for the poles. These belts completely ignore political boundaries and together make the entire world seem flat. They give every point on the earth an easy-to-use Eu163

Figure 4.1: The US Army’s Universal Transverse Mercator (UTM) system, shown here as if all gridded maps were laid together like an unfolded globe. The name of the system comes from the use of separate Transverse Mercator projections for each of the north-south grid zones. (That is, Mercator projections rotated ninety degrees so that there is no distortion along a north-south line of longitude.) Each tiny square— roughly onethird of a millimeter in size— is one hundred kilometers on a side.

clidean coordinate, expressed in meters, that can be used to enforce coordination and simplify calculations of all kinds— for everything from transnational infrastructure projects to the aiming of intercontinental ballistic missiles. In the decades after the war, army geodesists made a strong push to have UTM installed as widely as possible around the world. It was accepted for army operations in 1947 and joint US operations in 1949; by the early 1950s it was adopted by NATO, and other US allies followed shortly thereafter. As shown in figure 4.2, UTM spread over the course of the Cold War to unify most of the non-Communist countries. It was also matched by a very similar Soviet system that unified the Eastern Bloc.1 Installing UTM, however, was a major undertaking that involved much more than the printing of new maps. In order to be useful in the field, the lines on the map had to be backed up by consistent, high-precision surveys, and survey boundaries between countries could only be eliminated with staggering amounts of new calculation. The US poured huge resources into UTM and used a variety of strategies to enroll other countries, ranging from offers of free mapping work and mutual exchanges of data to the outright coercion of captured German mathematicians. But even though the confrontation between UTM and the Soviet system was certainly dramatic, the main contrast is not actually between these two grids. Instead, 164

Chapter Four

Figure 4.2: Use of UTM (and its Soviet counterpart) by the end of the Cold War. This map shows countries that had adopted each system for at least some of their own maps, either military or civilian; it does not show actual map coverage. Both the US and the USSR also fully mapped each other’s spheres using their own systems. (See note 1 for sources; there is no information for Cuba, North Vietnam, or North Korea.)

the more important tension was between these transnational grids and other ways of indexing geographic space. This includes not just the large number of national and subnational grids— more than 3,200 of which had been defined by the end of the century— but especially the reigning global system of latitude and longitude.2 The question that animates this chapter is simple: What kind of global space is created by UTM? Again, this is both an experiential and a political question. Experientially, UTM is more subtle than one might think from its rigid and abstract world diagram. To the user, the most important feature of UTM is not its global reach but the lack of sharp discontinuities— no coordinate mismatches at political boundaries, and relatively easy transitions from one north-south zone to another. When using UTM, its global space thus does not actually feel global; it is instead mostly a regional space, where the user is centered within a horizon of a few hundred kilometers. This is an intentional feature of its design, and it promotes an embedded geographic subjectivity— one that is similar to earlier national grids, but with less fragmentation. Politically, the installation of UTM as a transnational system was also pursued primarily at a regional level. UTM promoted the spatial consolidation of several major geopolitical regions— including Europe, Southeast Asia, southern Africa, and South America— and the US used different strategies for each. Cartographically speaking, however, UTM did not turn these areas into larger Territoriality without Borders

165

versions of territorial states, with geographic knowledge overseen by a centralized mapping agency. Centralization was only partial, and often temporary, and responsibility was rarely allocated based on geography alone. The territoriality of UTM thus broke down the strong interwar dichotomy between a national sphere of practical work and an international sphere of purely scientific activity. The regional and global spaces of UTM were just as practical and consequential as any national system, and they could easily coexist with existing national coordinates. In other words, UTM was fully territorial, but it was not territorially exclusive. This chapter is divided into four parts, all of which address these questions of globalism, boundaries, and new kinds of territory. I first analyze the geographic logic of the various grid systems used in World War II, with a particularly strong contrast between American and British strategies. The second section then introduces the postwar design of UTM and analyzes its experiential implications for a soldier in the field. Finally, the last two sections track the installation of UTM over the next several decades, in two phases: first as a primarily military (and regional) effort from the late 1940s through the early 1960s, followed by growing civilian (and global) interest thereafter. The chapter ends by zooming out to compare UTM with the traditional logic of latitude and longitude. This final comparison between UTM and the lines of latitude and longitude is central to my overall argument about the geo-epistemic transition from the authority of representational maps to the pragmatism of systems like GPS. While UTM is certainly an important part of postwar geodesy and Cold War military strategy, my main interest is still to understand how the new spatial technologies of the twentieth century differed from earlier ways of interacting with geographic space. In the case of UTM, the novelty here is therefore not simply that it was global in ways that earlier geodetic or military grid systems were not. After all, the coordinates of latitude and longitude have been global for more than two thousand years.3 What matters is how the gridded Euclidean space of the US military created a new kind of global and installed this new rationality as an intrinsic feature of geographic space. Unlike latitude and longitude, this is a rationality that puts the needs of everyday users above any overarching interest in the earth itself. It is a rationality of calculability, regional consciousness, and a retreat from the cleanly delimited space of the territorial state.

GRIDS IN WORLD WAR II: BORDERLESS TERRITORY, SHIFTING SCALE When the US Army created UTM in the late 1940s, its design was primarily a reaction to geographic problems that were encountered for the first time during World War II. In particular, UTM had to respond to two new problems that had not been especially relevant during World War I or the interwar pe166

Chapter Four

riod. The primary issue was related to mobility and boundaries. Compared to the static trench warfare of the 1910s or the neatly bounded spaces of civilian grids, World War II was a war of movement that turned the entire globe into a potential battlefield. Mathematically, however, any individual grid must stay within certain geographic limits in order to be useful, and although there were several strategies to make gridded space global, boundaries remained a persistent issue. The other problem was related to scale, and it stemmed from the new importance of coordination between land, air, and naval forces. Not only did each of these cover vastly different geographic areas, but they also had different requirements for balancing geographic precision with a wider regional consciousness. With both of these main problems, the inadequacy of existing strategies led to the creation of many new grid systems, each in conflict with the others. Ultimately, the problem was thus not that grids were unable to provide useful solutions to the new geography of the war, but that they tended to provide too many solutions, none of which satisfied all needs. And each new system inevitably made the main benefits of gridded space— unification, coordination, and simplicity— all the more elusive. Indeed, the most immediate difference between grids in the two world wars was that in the second, many of the logistical problems of using grids were solved and there were relatively few impediments to their widespread use. The basic logic was the same as before— grids enabled “map firing,” which meant that artillery could aim using only coordinates and could hit hidden or faraway targets with relative ease— but in World War I, grids had been relatively inflexible. An attack could be planned in advance, but changing plans in the middle of a battle was difficult. During World War II, this was addressed by the creation of centralized command centers— first by the US, but then by other countries as well— that could receive up-to-date observations, often directly from aircraft, and quickly send out new coordinates to the artillery, which could then hit a target only minutes after its location was first reported.4 This made grids much more powerful; it also meant that grids were often built directly into observation and targeting equipment. In 1943, for example, all American heavy coast artillery was realigned to use an artificial “grid north” instead of true geographic north. Radar equipment was likewise designed to compute range and direction in grid values, and radar screens were ruled with grid lines.5 Grids, in other words, moved even further away from the paper map: during World War II they were a pervasive geographic language, not a cartographic curiosity. All told, roughly half of all artillery fire in World War II would use a grid, and the distinction between map firing and simply firing became increasingly blurred.6 As successful as these logistical improvements were, however, the expansion of grids— both geographically and into the new domains of air and naval use— was generally done without serious study, and this is where the problems arose. The ad hoc nature of wartime grids is most clearly seen in the various attempts to make them cover large areas. In keeping with the reputation of Territoriality without Borders

167

Figure 4.3 (see gallery for color version): The British and American grids of World War II, as of April 1943. Each shaded area is a separate British-designed coordinate system; the idea was to match these systems to the probable theaters of the war as much as possible. The US system covered the rest of the world— the white areas of the map— with a series of narrow north-south strips, the boundaries of which are shown here with double vertical lines. Map from Army Map Service, Grids and Magnetic Declinations, Memorandum 425, 2nd ed. (Washington DC, 1943); shading added.

World War II as the first “truly global war,”7 continental or global systems were created by the UK, US, Soviet Union, and Germany— all within the first few years of the war— but these systems were essentially just expansions of whatever national or colonial systems had been in place by the late 1930s. Almost uniformly, these grids were ill suited to wider use, and some even seemed to forget the basic lessons of World War I. Taken together, they make it clear just how big the break was between the prewar geography of national space and the new globalism of the war. There were two main strategies that were used to make grids global. The mathematics of grids were the same as ever: a grid still consisted of a map projection applied directly to the earth to make a full-scale Euclidean coordinate system, and its geographic extent was still limited by the inherent distortions of the chosen projection. Since the two most useful projections— the Lambert and the Transverse Mercator— are best suited for east-west or north-south strips roughly six hundred kilometers wide, making a global grid system simply means deciding how to cover the earth with a series of these narrow strips.8 The two strategies can be seen in their simplest form in the systems of the UK and the US, both of which are shown figure 4.3. The British approach covered 168

Chapter Four

Figure 4.4 (see gallery for color version): The jigsaw puzzle of British grids in Europe. The “Nord de Guerre” zone was an expansion of the original système Lambert.

Europe, Africa, Australia, and most of Asia. The basic idea was to create a jigsaw puzzle of independent grids wherever they might be needed. Most of these grids had prewar origins, especially in Commonwealth areas, but many were hastily installed or expanded in 1940, roughly conforming to probable theaters of war. Figure 4.4, for example, shows the “Nord de Guerre” zone, which was simply a reinstallation and expansion of the original système Lambert from World War I, except that now the same coordinate system would be used not just in northeastern France but throughout northern Germany as well. It was also only the exigencies of war that finally allowed the imposition of uniform Transverse Mercator zones in East Africa.9 This piecemeal approach was largely in keeping with the logic of the interwar period, since each area was given a grid that suited its geographic alignment, and the overall international system was conceptually nothing more than a multiplication of national systems. The resulting whole therefore lacked any consistent logic. Most grids spanned international borders, but some did not. Most were measured in meters, but many used yards. Some used the east-west logic of the Lambert projection, while others used north-south Transverse Mercator (or even Cassini) projections. And the benefit of having Territoriality without Borders

169

many small grids, each of which would have less mathematical distortion and therefore be better at maintaining the fiction of the flat earth, was always balanced against the hassles of having too many disruptive boundaries between adjacent systems. In short, the British approach was messy— but it was also sensitive to local conditions.10 In contrast, the American solution was much more uniform and much less tied to existing political geography. As again shown in figure 4.3, the US covered all non-British areas with a rigid system of north-south grids, each stretching from pole to pole, spaced eight degrees apart. These strips ignored boundaries of all kinds, both physical and political— the only exceptions were for Maine, Hawaii, and the Philippines, which had preexisting grids that remained in place. This system was also an expansion of prewar precedent, but instead of creating a patchwork of new grids, the US approach simply used the north-south strips of its domestic military system (which, again, was separate from the civilian State Plane grids) and repeated them around the world. Calculations began soon after the US entered the war, and the full system entered operational use in 1943.11 Despite its apparent simplicity, this system was no less a product of wartime exigency than its British counterpart, and there was no attempt made to make it particularly coherent. Most glaringly, the prewar US system had used a nonconformal projection that introduced significant angular distortion into artillery computation, but the middle of the war was seen as an inopportune time to switch to a better alternative. (The projection was the polyconic, and the overall system was known as the World Polyconic Grid.)12 The overall spacing and numbering of the grid zones was also tied to the continental US rather than the Greenwich meridian or the various British grids. The boundaries of the American system were in fact only determined in 1942, when US and British mapping authorities divided the world into two distinct areas, and the American grids were simply applied to all areas that did not yet have any system in place.13 In other words, the US system spanned the world, but its design was not especially global. The wartime grids of the USSR and Germany both lie somewhere in between these two extremes. The Soviet system, which was first installed for domestic use in 1928, consisted of north-south strips, similar to those of the US, but spaced every six degrees and constructed using proper Transverse Mercator projections. The Soviet grids were arguably much more appropriate for global use— the basic design was in fact almost identical to the later UTM— but during the war they only seem to have been applied outside the USSR in areas immediately adjacent to Soviet-controlled territory, with sheets often published in sync with the westward Soviet advance that began in 1944. It was thus a transnational system, but with no apparent worldwide ambitions.14 The German strategy was rather less systematic and combined features of both main approaches. Early in the war, Germany used preexisting domestic systems both at home and abroad, similar to the British. But at the same time, 170

Chapter Four

they also began expanding their own national system— known simply as the Gauss Grid, after both the projection and the polymath— into neighboring countries. This grid had been designed with a certain global consciousness, as it consisted of north-south zones that were aligned with the equator and the Greenwich meridian, but the zones were only three degrees wide, and when these narrow grids were applied in new locations they often created as many problems as they solved.15 In 1942 Germany thus took a new approach and decided to copy (and rename) the Soviet domestic system for its own use throughout Europe. The resulting Deutsche Heeresgitter—German Army Grid— replaced the Gauss Grid uniformly on the eastern front (stretching all the way east to the Urals and south to Turkey), but conversion of grids at home and in the west was never complete.16 The Germans thus expanded domestic grids in the American manner, but there was always tension with preexisting systems. Between the two main approaches— the ad hoc British grids and the more systematic American style— it would be wrong to see one as necessarily simpler or more suitable than the other, and during the war there was no general consensus about which was better. The basic trade-off was a question of where the boundaries between grids would fall. The British approach allowed these boundaries to be placed in convenient places— in the oceans, along mountain ranges, or otherwise away from the main action. The British were especially critical of the uniform German system and its often-disruptive boundaries. One report saw these grids providing “an interesting sidelight on German expansionist psychology,” as the narrow zones “los[t] simplicity in an effort to attain great (and unnecessary) accuracy, and to be of world-wide application.”17 But in large part, accuracy and expansibility were not the main goals of the uniform approach— much more important was calculational efficiency. The goal of the US artillery, for example, was to have no more than a thirtysecond delay between receiving target information in a command center and relaying final coordinates to the field, and uniform grid zones meant that the same mathematical tables and coordinate conversions could be used anywhere in the world. Having fewer tables would lead to fewer errors and faster relays between observation and attack.18 The British, in other words, tried to push boundaries aside; the Americans tried to make them less important. The war largely ended up justifying the American position, as high mobility meant that encountering grid boundaries was almost inevitable. For one, British predictions about the probable theaters of war were often incorrect. The boundary between the “Nord de Guerre” and the “Lambert I” zones in northern France, for example, was especially problematic, as it ended up directly in the path of Allied armies advancing east from Normandy after D-Day. Shooting across this boundary not only required extra math but also raised worries about ambiguity between coordinates on either side of the line.19 But the more serious problem was deeper, since the truly problematic boundaries were not those between the grids themselves, but those between the underlyTerritoriality without Borders

171

ing national survey systems. At the western border of Germany, for example, there were discrepancies with French and Belgian surveys of several hundred meters, just as there had been during World War I. And although special equations were devised to bridge these gaps, adjusted coordinates could still give discrepancies of thirty or forty meters. (The amount of discrepancy need not be large to be problematic. For radar, sound-ranging, and long-range artillery, positional errors of around two meters were considered cause for concern.)20 At times grid-junction problems and survey problems could even overlap. The closely spaced colonial surveys of Egypt, Palestine, Trans-Jordan, and Lebanon, for example, were particularly problematic, since these four countries were covered by five different triangulation networks, some of which were only loosely connected, and prewar coordinates had been calculated using two different models for the size and shape of the earth. Ultimately two sets of coordinates had to be published for the Sinai Peninsula, giving different grid values on the Egyptian and Palestinian surveys.21 These problems were compounded by the fact that mobility across survey boundaries actually led grid coordinates to become less reliable over time. As each side inevitably captured the other’s survey information and converted it to its own system, previously reliable data would often be miscorrected, leading to major problems. Data captured by the Allies in southern Germany, for example, was often on a local coordinate system that first had to be converted to the main Prussian system before being distorted to match up with the neighboring systems of France, Switzerland, Austria, and Bohemia and then converted in turn to the British “Nord de Guerre” grid. By late 1944, a British survey officer complained that the coordinates being published by the newly liberated French army were “a jumbled hotch potch” of data drawn from multiple systems. His French colleague was forced to agree, saying that data in western Germany— one of the most surveyed areas on earth— “will never be reliable. . . . It will only be a patchwork which will invariably involve variations and which will offer an ensemble of non-agreeing coordinates.”22 This was serious business, since surveyors reported that when they found errors in the published grid values of even a handful of survey monuments, this was often adequate reason to distrust the entire list.23 The overall lesson, then, was that the only solution to boundary-related problems would be to eliminate the boundaries altogether. For grid junctions, this could be done by making grid-conversion equations ever more simple. During the war, for example, US army mathematicians spent much of their time devising new tables and new graphical techniques that could allow artillery to fire from one grid into another with increased accuracy in less time.24 For the survey junctions, however, there was no easy solution. The only strategy would be to recalculate everything from scratch. The second main problem of the war— that of coordination across geographic scales— proliferated relatively independently from the problems of boundaries. The major issue here was the conflict in all countries between 172

Chapter Four

artillery grids and the coordinate systems used by each country’s air force and navy. In England, for example, the British army used the National Grid (also known as the Military Grid), while the Royal Air Force— and consequently British radar installations— used the so-called Fighter Grid, which was based on a different projection that spanned the English Channel. The Germans’ Gauss grid competed not just with the later Heeresgitter, but also with three different systems used for communicating with the air force: the generalpurpose Meldenetz, the Gradmeldenetz (for bombing) and the Jägermeldenetz (for antiaircraft reporting). And the US Army’s World Polyconic Grid was used alongside the Air Defense Grid and the Joint Army-Navy grid. Each military also used several reference systems that were sheet specific or relied on gridded acetate overlays.25 Different coordinate systems, in other words, did not just multiply around the world, but were also layered on top of each other. In 1941, for example, a US Army training manual included a discussion of just two grid systems. Three years later, the 1944 revision of the same manual detailed eight different coordinate systems which were then in common use by the Allies, with all the British grids counted as one.26 By the end of the war, points in northern France could be referenced by no less than seven different overlapping systems. Most of the systems used by air and naval forces were not technically grids at all, but rather just new ways of making latitude and longitude references easy to communicate. The American Air Defense Grid, for example, was a five-layer system for subdividing the world into ever-smaller areas based on latitude and longitude. Figures 4.5 and 4.6 show this system; it was not Euclidean, and the size of the areas varied around the world. The Joint Army-Navy grid also used latitude and longitude, but instead of fixing coordinates for the earth as a whole, its coordinates could be moved, rescaled, and redefined for each battle. As shown in figure 4.7, the system was typically pegged to a specific map, using a similar logic to the simple locator grids found in road atlases. Figure 4.8 shows how it could be extended to adjacent map sheets, but the basic idea remained the same.27 There was thus almost no functional overlap between these systems and the local Euclidean requirements of artillery grids, and wartime attempts to reduce the number of systems were consistently unsuccessful, since no system addressed the needs of other users. Beginning in late 1942, for example, a series of meetings were held in the United States to try to choose either the Air Defense Grid or the Joint Army-Navy system for use in combined air-land-water operations, but it was decided that restricting all forces to one system would be worse than having multiple options.28 At its root, the incompatibility here was about how to deal with multiple geographic scales at once. Army operations were resolutely local: artillerists did not need to refer to points more than a few dozen kilometers away, and they sometimes needed to know their coordinates accurate to a few centimeters. And while it was possible to give an unambiguous global position on an artillery grid, in practice coordinates were almost always abbreviated. British Territoriality without Borders

173

Figure 4.5: The first two reference levels of the US Air Defense Grid, which gave letter values to large blocks of latitude and longitude. Since it is not Euclidean, it properly not a “grid” at all. From War Department, Orientation for Artillery, Technical Manual TM 44-225 (Washington DC, 1944); shading added.

grids, for example, were usually subdivided into a series of large squares with two-letter codes, and these codes would be used as a prefix when reporting a location. In a large grid, however, the same two-letter codes could be used more than once; at times they could also be confused with the codes of a neighboring grid.29 The grids of the other armies tended to give references in numbers only, but these were quickly truncated when ambiguity was unlikely. For example, a point whose full coordinates were 1457310, 2892005 might be reported using only the last four digits of each number, making them completely meaningless without adequate context. These systems of abbreviation worked perfectly fine in the field, but communicating with higher command was cumbersome at best. This meant that even global artillery systems like the World Polyconic Grid were quite local in practice. In contrast, air and naval operations were almost never local. They ranged over much larger areas, and accuracy on the order of centimeters was completely irrelevant. And since navigators often relied on the sun and stars (in both planes and ships), being able to refer to latitude and longitude was imperative. With the Air Defense Grid, for example, only four letters were needed to refer to an area the size of large city, anywhere in the world, but it was impossible to refer to areas small enough to be helpful to an artillerist. The navy likewise never found it necessary to refer to areas less than two hundred yards 174

Chapter Four

Figure 4.6: The most precise Air Defense Grid references would include both letters and numbers; “Point X” on the west coast of Florida has coordinates of EBPW 5518, which is only accurate to about a kilometer. Same source as figure 4.5.

on a side.30 But there was even incompatibility between the air force and the navy: the air force always wanted global references, while the navy system was inherently regional and could not give global references at all. In other words, there was no system that could address space at the local, regional, and global scale at the same time. Army mathematicians had made local geography intuitive and Euclidean (albeit chopped up by inconvenient boundaries), while traditional navigation took place in a global space that was spherical, continuous, and relatively imprecise. In between was a no-man’s-land of ambiguous and incompatible regionalisms. Neither the problem of boundaries nor the problem of scale had been especially relevant during the interwar period. For the designers of domestic grid systems, considerations of scale and continuity were largely dictated by political boundaries, and the territorial state had likewise been taken as the natural unit in international discussions. If anything, discontinuities at international boundaries were seen as a good thing, since they protected domestic space from foreign interference. But a mobile, multicontinental war and new imperatives of coordination between land, air, and water made earlier systems seem Territoriality without Borders

175

Figure 4.7: The US Joint Army-Navy reference system. This system was applied to a single map sheet (at any scale); its geometric properties therefore depended on the projection used for the underlying map. New systems would be created for each battle. It can refer to points or areas; square A is “JAN area 47,” and point D is “JAN point 8359.” Same source as figure 4.5.

increasingly unworkable. There was suddenly no such thing as a conveniently located boundary, and strategy needed to be able to address the megaregional and the microlocal in the same coordinated space.

DESIGNING THE UNIVERSAL TRANSVERSE MERCATOR SYSTEM: THE SOLDIER’S NEW REGIONAL HORIZON Planning for the Universal Transverse Mercator system began even before the war had ended. Initial meetings were held within the army in 1944, and the final system was made operational in early 1947. It was accepted by the Canadian and British armies the next year, and by the time the Korean War broke out in 1950, it was in use for joint land-air-marine operations as well.31 Its main designer was a mathematician-astronomer named John O’Keefe, who had degrees from Harvard and the University of Chicago and had spent World War II working on map calculations for the army. From the beginning, he presented 176

Chapter Four

Figure 4.8: The Joint Army-Navy system can be extended to adjacent map sheets using letters. The original map becomes the M coordinates, and the point labeled Z, on the sheet just to the southwest of the original map, is “JAN point G7025.” Same source as figure 4.5.

UTM as an eminently logical continuation of earlier ideas, not a radical break. Not only did it echo the sheet lines of the International Map of the World, but it was also nearly identical to the international grid system for Africa endorsed at the IUGG in 1936 and the Soviet and German systems used during the war. Civilian mathematicians had even recommended it as an alternative to the World Polyconic Grid as early as 1941.32 But O’Keefe’s design was also clearly a reaction to the two main problems of World War II. First and foremost, it sought to minimize boundaries— both between grid zones and between national survey systems— and to make those boundaries easier to accommodate. It was also explicitly designed to upstage the coordinate systems of the air force and the navy, not just to allow better coordination between artillery, aircraft, and ships, but also to support the new requirements of long-range missile targeting. Put in somewhat grander terms, however, the larger goal of UTM was not just to make a new grid system, but to make a new kind of soldier on a new kind of earth. This soldier would be just as geographically embedded Territoriality without Borders

177

as with earlier grids, but UTM was also designed to make global space into a seamless whole, continuous both across boundaries and across scales. Although the basic idea behind UTM was indeed well known and widespread, there were two important ways that it departed from earlier systems, both of which give insight into the kind of space—and the kind of subjectivity— sought by the US Army. First was the way that coordinates were labeled (and therefore experienced) using a rubric known as the Military Grid Reference System, or MGRS. Second were its mathematical details and the way that UTM would interface with preexisting national survey systems. In both cases, the overarching sensibility was the same: although UTM was a uniform global system, it was inflected with an important regionalist sensibility. This move was similar to the changes in cartography described in chapter 2, where regionalism was not an alternative to globalism, but rather the way that the global world was actually experienced. From the soldier’s point of view, this meant that the global system would seem everywhere local, with the user centered within a regional horizon. From a strategic point of view, this meant that UTM would be mathematically calibrated for the probable (regional) geography of future global wars. The most obvious sense of regionalism can be seen in the Military Grid Reference System. Although UTM divided the world mathematically into sixty narrow north-south strips, each six degrees wide (plus small circular areas for the poles), this is not how a soldier would actually understand the system. Unlike the global systems used during the war, which generally labeled coordinates with reference to the equator and the Greenwich meridian,33 the new MGRS used three types of references at once, none of which followed this pattern. First, large areas bounded by lines of latitude and longitude were given a “Grid Zone Designation” of letters and numbers, as shown in figure 4.9. The zone containing Boston, for example, was labeled 19T. (Each of these large zones covered the same area as two sheets of the International Map of the World.) Second, every north-south strip was divided into a grid of onehundred-kilometer squares, and each square was given a two-letter code; these are the smaller letters shown in figure 4.10. These letters would be added after the Grid Zone Designation, such as 19T-CG for the Boston metro area. Finally, numerical coordinates would be found within each of these one-hundredkilometer squares by measuring from the southwest corner. As with earlier systems, precision was determined by the number of digits used, with more numbers being used for a more precise reference. To locate downtown Boston within one kilometer, for example, the reference would be 19T-CG-30-91, while the coordinates of, say, Paul Revere’s gravestone— accurate to one meter— would be 19T-CG-30179-91536. (In practice, references are always given without punctuation, such as 19TCG3017991536.)34 The payoff from this somewhat baroque scheme was that UTM could accommodate global, regional, and local scales at once, but the soldier on the ground would stay rooted in local coordinates measured in meters. This was especially important as a way of blocking competition from the latitude-and178

Chapter Four

Figure 4.9: Top-level “Grid Zone Designations” for UTM. The north-south bands are numbered 1 through 60, just the same as the International Map of the World, and rows of latitude are given a letter C through X. (Boston is in zone 19T, shown in gray.) At the global level, references are thus quite similar to those of the latitude-and-longitude systems preferred by the air force and the navy. From Army Map Service, Grids and Grid References, Technical Manual 36 (Washington DC, 1950); shading added.

longitude systems preferred by the navy and the air force. (In the late 1940s in particular, the US Joint Chiefs of Staff chose UTM over a new air force system called GEOREF that could finally give local references, but still only on the graticule.)35 The preference for the local was shown, as in earlier systems, by the way that coordinates were abbreviated: soldiers would almost always use short, local coordinates, and only rarely the full global version. And although the two-letter codes were similar to those used earlier by the British, in the American system they were designed to run continuously across neighboring grid zones, ensuring that the same code would not repeat within several hundred kilometers.36 In Korea, for example, regional references were the most common, and a typical coordinate would be something like YD3621 (central Pyongyang), since the nearest square with an identical reference was in western China. At the regional scale, this meant that there were finally no boundaries. That is, there was nowhere on earth where a soldier would have to use a global reference to avoid regional ambiguity. There were other ways that UTM reinforced these assumptions about scale and regional consciousness. Perhaps the simplest was that official policy dictated that the full UTM grid would be printed on all maps at scales of 1:250,000 and larger, while those at smaller scales would only have the global Grid Zone Territoriality without Borders

179

Figure 4.10: The large codes are the top-level grid zones. The smaller areas are one-hundred-kilometer squares, each of which is given a two-letter code. These two-letter codes run continuously from one grid zone to the next and do not repeat within several hundred kilometers. This means that the global references can usually be omitted. Same source as figure 4.9.

Designations. This policy was explicitly designed to divide the purview of the army from that of the air force; it thus effectively defined the ground soldier’s regional horizon as an area roughly the size of Connecticut, which is the extent of a single map sheet at 1:250,000. This was the outer limit of the local.37 In turn, the alignment of the top-level UTM grid zones with the sheets of the International Map of the World was a comment on the outer limit of regional scale, since the 1:1,000,000 scale of the IMW was seen as the useful limit of strategic topographic mapping. (O’Keefe even distributed UTM grid tables to other countries in order to encourage them to make their IMW sheets.)38 UTM, in other words, was not a perfectly scalable global system. It gave several specific cutoffs between local geographic embeddedness, a wider regional view, and synoptic globalism. The inherent regionalism of UTM’s underlying mathematics is not as obvious as these more explicit markers of scale, but it was no less important. At first glance, UTM seems to completely ignore natural features like rivers, mountains, and deserts, and it seems to apply the same mathematical gridiron to all parts of the world equally. Indeed, there were several ways that the mathematics were adjusted to heighten the uniformity of the system and to make boundaries as unproblematic as possible. Adjacent grid zones, for example, were made to overlap by twenty-five miles— greater than the maximum range of conventional artillery in the 1940s39—and the Army Map Service provided relatively simple equations that could easily convert coordinates from one zone to the next. In the 1950s, army mathematicians developed additional equations that could convert coordinates across as many as nine zones— fiftyfour degrees of longitude. This meant that UTM could be used to aim all but the longest-range missiles, speeding up calculations by as much as a factor of five. For almost all tasks, UTM thus really did make the world flat, replacing latitude and longitude with simple Euclidean coordinates at a continental scale.40 And over its entire lifetime, only three modifications have ever been made to the basic grid structure to accommodate local conditions; these were all made in the mid-1950s in the vicinity of Norway, in order to keep the Norwegian Army from abandoning the grid. Otherwise, the basic diagram has not yielded.41 But despite this appearance of a total withdrawal from geographic specificity, there were two ways that UTM was designed with real-world geography in mind. First was a subtle Eurocentrism. The specific mathematical parameters of the UTM projection were chosen with the assumption that the location of future land wars would be in European latitudes, where, according to O’Keefe, “the most violent battles are usually fought.”42 Only north of the Mediterranean would the distortions of the projection remain within artillery requirements; closer to the equator, errors at the edges of the grid zones would quickly compromise accuracy if left uncorrected. UTM was still better at minimizing errors than the earlier systems for the USSR or Africa, but as shown in figure 4.11, in the 1960s its inherent distortions happened to fall squarely along the north-south axis of southern Vietnam.43 Territoriality without Borders

181

Figure 4.11: Errors at the edges of UTM grid zones in Europe (above) and Southeast Asia (below). The dark gray areas show where measurements on the grid differ from real-world measurements enough to cause problems for aiming long-range guns and performing field surveys. UTM was specifically designed to fulfill all artillery requirements in Europe, but closer to the equator, accuracy is compromised in order to keep the grid zones as wide as possible. In South Vietnam, errors are almost two and a half times the allowed maximum. The small squares are again one hundred kilometers on a side. For GIS layers, see www.afterthemap.info.

Figure 4.12: Ellipsoids to be used for UTM coordinates. Rather than allowing each country to maintain its own system, the Army Map Service sought large regional coherence. (Actual recalculation of triangulation systems, however, would come later.) Over time the idea was to have the European block, shown here using the International Ellipsoid, expand to include Africa and Asia, while the North American block would expand to include South America. Map from Army Map Service, Universal Transverse Mercator Grid Tables, Technical Manual 7 (Washington DC, 1949?).

The second and more important way that UTM was adjusted to the specifics of the earth had to do with the relationship between grid coordinates and national survey systems. The fact that almost every country calculated its surveys using a different model for the size and shape of the earth— that is, a different ellipsoid— was one of the main reasons that coordinates did not align at international borders. This was the same issue that had been addressed (unsuccessfully) in the 1920s and 1930s. Figure 4.12 shows how UTM first confronted this problem: instead of using one ellipsoid for the entire earth, the Army Map Service divided the world into seven large blocks, each of which would use one of the five most common ellipsoids. In North America, for example, UTM coordinates would assume that the earth was substantially larger and less spherical than it was in the Soviet Union, since this is how the surveys of Territoriality without Borders

183

the US and the USSR had originally been computed. Notice as well that the boundaries between these large regions were hardly arbitrary. They generally followed major physical or political features: the Sahara Desert, the Caucasus and Himalayan Mountains, or the southern border of former Japanese Manchuria.44 This did not immediately solve the problem of mismatched national surveys, but it did announce that the US Army would push for continental consolidation. Over time the idea was that the European block, which used the International Ellipsoid, would expand to incorporate Asia and Africa, while the North American block would expand to cover the entire Western Hemisphere. Eventually the hope was to connect all surveys into one global system, so that artillery would be able to operate anywhere in the world without fear of unreliable coordinates, and long-range missiles would be able to hit precision targets on the other side of the ocean. Taken together, both the labeling of UTM coordinates and these subtle tweaks to its mathematical structure point in the same direction: UTM was certainly a global system, but its globalism was of a particular flavor. Instead of treating the earth as an abstract mathematical object, the goal was to make UTM coordinates correspond with a real-world regional horizon that would actually be felt by the soldier on the battlefield. And although there were several ways that UTM defined the extent of this regionalism— from an area the size of Connecticut up to the entirety of North America— the ability to operate differently at different scales was also an explicit goal. Artillerists needed different coordinates than pilots, and long-range missile targeting required coordinates that stretched smoothly not just across international borders but across entire continents. UTM thus promised to do what no other system had done before: to create a single coordinate system that could accommodate the needs of all user groups, and all tasks, everywhere in the world.

INSTALLING UTM: A NEW RELATIONSHIP BETWEEN COORDINATES AND TERRITORY The design of the Universal Transverse Mercator system was innovative and ambitious— but earlier transnational coordinate proposals had been, too. What separated UTM from its predecessors was that it was in fact successfully installed as a working and self-sustaining system. This required a delicate strategy that balanced unilateral and multilateral initiative. Not surprisingly, the US ended up shouldering most of the up-front burden of recalculating and redrawing existing maps, but the Americans’ ultimate goal was to create a shared— even “universal”—geographic language, and this meant that other countries would have to be enrolled. These other countries would have to see UTM align with their own interests, and final responsibility for the system would have to be decentralized. Indeed, this balance between American muscle and global openness is what most distinguished UTM from the very 184

Chapter Four

similar Soviet system, which likewise expanded after World War II but was always managed as a covert program, where even the existence of the finished paper maps was kept secret.45 Granted, the American strategy could at times involve a certain amount of coercion— including even the literal conscription of captured German geodesists— but by the mid-1950s UTM had begun to shift the long-standing relationship between national coordinates and national territoriality. Despite its strong hand, the US military did not become the mapping office for the entire non-Communist world. Instead, by successfully persuading other countries to adopt and maintain its universal system, the US ended up problematizing the very idea of centralized geographic knowledge. The main figure behind the installation of UTM was a geodetic engineer named Floyd Hough, who began his career with the US Coast and Geodetic Survey just after World War I and was appointed head of the Geodetic Division of the US Army Map Service just after serving in Europe during World War  II.46 In keeping with UTM’s regionalist design, Hough’s main postwar task was to oversee the massive consolidation of national survey networks at a continental scale. This required acquiring high-precision data from preexisting surveys (or making new surveys where existing regional connections were not adequate) and having it recalculated as part of a greater whole. In a 1953 memo, John O’Keefe— who worked under Hough— called UTM a “vehicle” for these other projects: although the cartographic goal was simply to replace the wartime British grids with the uniform American system, the grid lines on the map were only the public face of the more fundamental work of collecting, recalculating, and stitching together national surveys. This, after all, was the work that would finally allow artillery and long-range missiles to be targeted across international borders.47 The region of immediate concern was Europe, which was successfully consolidated in the late 1940s and early 1950s. The goal was then to attach other regions to this European core (and to do the same in the Americas), eventually creating a single global system.48 Although this larger plan would not proceed as smoothly as hoped— mostly due to unexpected new measurements for the shape of the earth— regional consolidation still continued, and by the 1960s new transnational coordinates had been installed in dozens of countries, layered on top of existing national systems.

Step 1: Europe Hough was well aware of the long-standing American interest in the survey networks of Europe. When the eminent William Bowie first presented the idea of recalculating the European triangulation in 1930 as an international scientific project— one that would not interfere with the practical work of national mapping agencies— Hough was working directly under Bowie as a young engineer.49 World War II, however, was a decisive turning point, and it was exactly the practical problems of coordinate misalignments at international borders Territoriality without Borders

185

that motivated the Army Map Service to finally achieve what international collaboration could not. The first steps toward a practicable recalculation came as early as 1944, when Hough was given command of a special military unit of seventeen men— known as the Hough Team— that followed behind the American advance through Germany scouring libraries and archives for map data. The success of this mission was overwhelming. Just a month before the end of the war in Europe, the Hough Team ended up locating the chief engineer of the German national triangulation— the world-renowned geodesist Erwin Gigas— holed up in a small village in what would soon become Soviet-controlled territory, along with some of his top-ranking staff. A week later, Hough discovered a warehouse in a nearby town containing the entire map archives of the German Army. This archive included data not just for Germany, but also for eastern Europe and even the original surveys for the Trans-Siberian Railway, which had been performed by German contractors before World War I. All told, the Hough Team ended up shipping a total of 180,000 pounds of captured material back to the US.50 This captured German data, along with the captured German geodesists, formed the nucleus of the army’s massive recalculation effort. The actual work, however, proceeded in two steps, with two very different political strategies. The first step was simple coercion. After promptly moving his German captives to the American zone, Hough used POWs to increase the size of Gigas’s staff from seventeen to sixty and then directed them to begin recalculating the triangulation of central Europe using his preferred technique—a full, rigorous version of the Bowie method, using the (American-derived) International Ellipsoid. Two years and 1,305 simultaneous equations later, the calculations were complete, and survey boundaries were finally eliminated within an area stretching from the Netherlands and northeastern France east through Germany, Austria, Czechoslovakia, and Poland into the western USSR.51 This entire zone could now be treated as a single unified space, with high-precision coordinates ready to facilitate everything from bombing to bridge building. The second phase— connecting the rest of Europe to this newly rectified block— required much more diplomatic finesse. Just after the war, one of Hough’s colleagues visited various western European survey agencies and received positive reactions to the larger plan; Hough also argued in print that no country should object to the project on “political” grounds, because much of the high-level triangulation data had already been published in official national reports. Even interwar Soviet data was freely available.52 But when Hough convened a special session of the International Union of Geodesy and Geophysics in Paris in 1947 (under the direction of Pierre Tardi, designer of the most important interwar precedent to UTM) and offered the calculational services of his captives, several countries balked at the idea that German geodesists would be given access to western European data and entrusted with a project which clearly exceeded the realm of disinterested science. Hough’s counterproposal was a win-win proposition. Rather than continuing to use 186

Chapter Four

Figure 4.13: The US Army’s recalculation of the European triangulation. The darker lines are the Americanstyle Bowie loops selected from the continental network to make the calculations workable. The thick gray ring stretching from Germany east to the USSR is the original central European block that was calculated just after the war by the German geodesists captured by the Hough Team in 1945. Once the full recalculation was complete in 1951, all of Europe was able to be unified using UTM coordinates, with no mismatches between countries. Image from Floyd Hough, “The Readjustment of European Triangulation,” Transactions, American Geophysical Union 28 (Feb 1947): 63. The final calculations deviated slightly from this plan; see note 53.

German labor, he proposed to have all calculations done instead in Washington DC by computers at the Coast and Geodetic Survey (both human and mechanical). The final high-precision coordinates would then only be made publicly available to an accuracy of about twenty meters, with the full results being communicated in confidence to each country individually. Figure 4.13 shows Territoriality without Borders

187

how the calculations proceeded, with American-style Bowie loops carved out of the existing triangulation network. Although each country would continue to retain its existing national system, by the middle of 1951 new continental coordinates were available from North Africa to the Arctic Ocean. By that time Hough had also succeeded in getting UTM adopted as the preferred grid for NATO.53 With this heroic effort complete, the next task— no less monumental— was to redraw and reprint all NATO maps of Europe using the UTM grid. Since a gradual transition would only result in confusion, with some maps showing UTM lines while others still used the old British war grids, the Army Map Service coordinated a massive replacement of all maps in most of western Europe, all at once, on a single day— “C” Day, March 26, 1952. John Ladd, the commanding officer of the Army Map Service, called this effort “perhaps the largest single project in the history of military map making.” It required the conversion of more than twelve thousand maps and the printing of nearly ninety million sheets— more than half of which were shipped from the United States. These maps were accompanied by fifteen hundred new editions of trig lists that together gave precise UTM coordinates for roughly four hundred thousand survey monuments.54 As large as this effort was, however, it only replaced two of the earlier British grids: the original “Nord de Guerre” and the “North Italy” zone (see again figure 4.4). Over the next few years, similarly massive changeovers would take place for maps throughout Europe, replacing British grids one by one. (The grids for the British Isles themselves, however, would remain in place.)55 As map replacement proceeded apace, the Army Map Service also extended the underlying European survey block even further. For calculating positions in the USSR, the International Ellipsoid replaced the Soviets’ own ellipsoid in 1952, and the next year the captured German material was used to (secretly) connect the European triangulation through Siberia all the way to the border with Manchuria.56 In the Middle East, new surveying was done to extend the European system south through the Arabian Peninsula, and eventually east through Iran as well. In this case, the conversion of paper maps was purposefully delayed to allow time for the new geodetic work, since the army was worried that the premature appearance of UTM-gridded maps would only make recalculation seem less urgent. Finished sheets were available by the end of the decade.57 For this first phase, the overall strategy was thus quite similar to what the United States had done with the World Aeronautical Chart and the International Map of the World. The US essentially upstaged the politics of voluntary internationalism and completed what had been conceived of as an enormous, long-term collaborative project in just a few years, all on its own. But the US was also anxious to maintain the legitimacy of the project as an international endeavor, and so other countries were asked to contribute as much as possible and the results were, for the most part, freely shared. Again similar to 188

Chapter Four

the cartographic projects, the details of the collaborative arrangements reveal a subtle shift in the way that mapping responsibility was divided between allies. Rather than responsibility being allocated based on geography alone, as had been done during the war, the American role in European map conversion mostly varied by cartographic scale. In general the Americans took responsibility for smaller-scale maps at 1:250,000 and 1:100,000, preferring to leave larger scales to each country individually. But the US also made special arrangements as necessary; it did, for example, agree to convert large-scale maps in Denmark, to use its captured German data to produce maps of Spain and Albania, and to do supplemental aerial survey work in Greece.58 In other words, although the project was unquestionably an American one, the Army Map Service tended to act less like a primary national mapping agency than as an agency of last resort. It agreed to take on work only when local capacity was insufficient, and it sought to pass responsibility to local agencies whenever possible.59

Step 2: Asia, Africa, and Latin America Even before the European project was finished, the Army Map Service began to shift its attention to the tropics. Existing surveys outside Europe were generally quite disconnected, and many countries had no unified national system at all, meaning that extending the survey blocks of Europe and North America would require work at both the transnational and the national level at once. This larger project thus followed much the same logic as before, but with even more sensitivity to international legitimacy, accompanied by ever subtler forms of persuasion and new divisions of labor. As a result, however, seeing this new phase as nothing but thinly veiled American self-interest is not quite right— at stake here is a more profound reconfiguration of international mapping work. To be sure, the public debate was often accompanied by more than a healthy dollop of lofty rhetoric. In 1951 and 1952, for example, Hough introduced UTM at regional survey meetings for Africa and South America by presenting the “universal” in UTM to mean not just universally applicable, but universal in a much more philosophical sense. To British African surveyors, he explained that the name of the system came because “we believed [UTM] to be so logical that it was not unreasonable to assume that the people on such planets as Mars and Venus were using also this system.”60 Likewise, even more than he had with his European counterparts, he stressed the historical continuity of UTM with the interwar scientific goals of the IUGG and described the recent recalculation of the European triangulation as “a monument to wholehearted international cooperation . . . among scientific men of goodwill,” which seems disingenuous at best.61 And indeed, Hough received a somewhat skeptical response from survey officials suspicious of American intentions, especially in Africa. The chief mathematician for British colonial surveys, for example, not Territoriality without Borders

189

only thought that Hough’s push for global uniformity was mathematically overambitious, but he also questioned the larger project, warning that “in the present circumstances of atomic war-heads and guided missiles . . . happy is the country that has no [coordinates] fixable with reference to UTM.”62 One certainly did not need a weatherman to see who would benefit most from Hough’s project. But behind these somewhat stark public exchanges, both the specifics of Hough’s proposals and the way that work actually proceeded were not so black and white, and together they show the larger territorial implications of the UTM project— in Europe as well as the wider world. From the point of view of national autonomy, perhaps the most important part of Hough’s presentations was his assurance that UTM should not be considered as a replacement for nationally defined grids, since each served a very different purpose. The goal of a global system was to minimize the total number of grid zones, even at the expense of distortion, while the goal of a national system was to keep distortion everywhere low enough for precise cadastral and engineering surveys. Not only was there “no disadvantage in the customary practice where a country retains its cadastral grid along with the UTM,” but the United States itself would continue to support the interwar State Plane system without change. European countries would maintain their established domestic systems as well.63 The Army Map Service backed up this pledge in practice, too: at the same time that Hough was lobbying for the adoption of UTM in Central and South America, his staff was helping most of these Latin American countries design national grids that were custom fitted to local geography and discontinuous at national borders.64 Similarly, when the US was confronted with a British proposal for a new custom grid for Malaya in 1955, an internal American memo noted that, “as a strictly local problem, no one can successfully refute . . . the superiority of the [proposed British] projection over UTM.” Local suitability, however, was something different from the “correct solution on an extended-area basis”— they were separate spheres that should be treated separately.65 Hough’s assurance that his international military grid would remain disconnected from domestic civilian grids made perfect sense, both mathematically and politically, but it represented a sharp departure from the European ideals of large-scale mapping over the previous two hundred years. Since the late eighteenth century, the ultimate goal of national and colonial mapping agencies had been territorial unity: the connection of national triangulation, topographic mapping, and local cadastral records into one overall system. Historians of cartography have shown that most of these attempts at integration did not succeed, but the repeated failures only underscore how much the history of modern mapping— including the history of grids, from the first carte de France to the 1930s— consists of attempts to achieve this singular aim.66 In contrast, Hough made it clear that UTM would create a world with two overlapping and independent systems, just as it had in Europe— one national, and one transnational. Coordinates in the two systems would not even necessarily 190

Chapter Four

be the same, since they were often based on different calculations, using different ellipsoids. And unlike William Bowie’s project of the 1930s, the international system was not confined to “purely scientific” purposes: both UTM and national grids would be used to provide continuous local coordinates based on high-precision surveys.67 In other words, national territory would continue to be important, and would still be characterized by tight correspondence between political and cartographic boundaries, but layered on top of this was a new transnational space that was arguably just as territorial. This was the space of regional military pacts— NATO in Europe, or the Rio Treaty in the Americas. For example, even before South American consolidation began, O’Keefe wrote to an officer in Peru: “I think that the question of the best [grid] system cannot be settled apart from political considerations. The great problem is the defense of the Western Hemisphere as a whole.”68 The territoriality here was therefore ambiguous: considered from the point of view of an individual country, UTM was a tool of foreign relations, but considered from the point of view of regional defense, UTM was essentially domestic. The actual work that followed over the next few years further problematized the separation of distinct national and international spheres. The Americans’ preferred strategy for enrolling other countries, as it had been in Europe, was to make offers of serious support and resources, and these offers were structured in ways that were easy to accept. The most straightforward simply had the United States perform survey work free of charge. In Africa, for example, British colonial surveyors had been working for decades to create a unifying axis of triangulation stretching from Egypt to South Africa, but work had stalled in the swamps of Sudan. In the early 1950s the British gladly deferred to the Americans, who quickly completed the job.69 In Central and South America, the approach was even more systematic. Just after the war, the Army Map Service began an outreach program called the Inter-American Geodetic Survey, which was structured as a series of bilateral agreements between the US and nearly every country in Latin America. Initially the US simply offered to perform high-precision surveys and aerial photography, free of charge, as long as it could keep a copy of the results, and through the early 1950s army engineers proceeded to extend and stitch together many formerly disconnected systems. In 1952, however, the US expanded the program by opening a free survey training school in Panama, with classes conducted in Spanish and enrollment open to all participating countries. The graduates of this school eventually took over the bulk of the fieldwork in their own countries, but results were still shared with the US.70 By the mid-1950s, similar offers of free training or consultation in exchange for survey data were quite common and were extended not just to national survey agencies but also to corporations— to Shell Oil in Alaska just as much as to Brazil in the Amazon.71 The most common way that the US enrolled other countries, however, again made use of the overwhelming computing power of the Army Map Service and the Coast and Geodetic Survey. Even before the advent of programTerritoriality without Borders

191

mable computers, mass calculation had defined the American approach to geodesy, and the resources and accumulated expertise of these two agencies together dwarfed that of most other countries combined.72 American offers of free calculation were often a trump card. In the 1930s, for example, a British survey officer had estimated that converting Ugandan coordinates to a new system could take as long as twenty worker-years. When Hough offered free computation for UTM in 1951, this was also an opportunity to finally pursue national or colonial projects that had been previously unworkable, and much of Africa joined the project despite a preference for British-style grid design.73 The next year, the Army Map Service also became the proud owner of the third UNIVAC machine, which could perform calculations that were an order of magnitude more precise than what was possible with paper tables. With this added incentive, even the very same British official who publicly raised concerns about missile targeting ended up sending data to Washington. His results— to millimeter accuracy— were ready in just three months.74 The army also processed all the raw data coming from Latin America, and in the mid1950s it recalculated the surveys of Southeast Asia, consolidating the British triangulation of India with those of Thailand, Cambodia, Laos, and Vietnam.75 These exchanges certainly worked to the advantage of the United States, since they allowed the Army Map Service to assemble a worldwide archive of survey data. But even though the US positioned itself as a global center of calculation, this was not a traditional colonial arrangement. The complex calculations typically only needed to be performed once, and afterward the results were passed off to be enacted, maintained, and extended by autonomous national agencies. Indeed, rather than seeing the spread of UTM as evidence of simple American imperialism, it is perhaps more productive to see it as the clearest signal that the interwar dichotomy between national and international domains had shifted. Regional consolidation did not replicate the duties of a centralized mapping office at a larger scale, and US sponsorship rarely extended beyond the big expensive projects. This gave the US Army a global view, but its involvement was temporary and did not imply any ongoing responsibility. And unlike other forms of US presence abroad, its geodetic work was genuinely welcomed— perhaps exactly because it remained detached from local concerns.76

Step 3: The World Throughout the 1950s, the final goal was always to assemble national survey systems not just into regional blocks but also into a single global whole. The need for a worldwide system was felt by the newly formed US Department of Defense as early as the late 1940s, since only a global system would allow for accurate calculation of trajectories between continents. The regional and global projects, however, were not always directly related, and it was possible to pursue one without the other, or both at once. In the early 1950s, the global 192

Chapter Four

Figure 4.14: High-precision survey data acquired by the Army Map Service by the end of the 1950s. The heavy lines show the huge arcs of intercontinental triangulation that were assembled— in part by US Army surveyors working abroad— to recalculate the size and shape of the earth in 1956. A few months later, Irene Fischer also included triangulation in the shaded areas for a more complete calculation; this resulted in the new “Hough Ellipsoid.” By the end of the decade, Fischer would extend the calculation into Japan as well. (Note that this map does not show the full extent of the US archive; it only shows the data used for the ellipsoid calculations.) Image from Irene Fischer, “The Hough Ellipsoid, or the Figure of the Earth from Geoidal Heights,” Bulletin géodésique 54 (Dec 1959): 50.

problem was assigned to both the Army Map Service and the air force’s Aeronautical Chart and Information Center, and the two agencies took rather different approaches. Although solutions began to be calculated by the late 1950s— the first tentative framework was created by the army in 1956, which was then expanded and combined with the air force system in 1959— these initial results left much to be desired, and over the course of the 1960s the difficulties of the global problem led to a further territorial layering, with separate stand-alone systems not just for national and regional space, but for a more detached global space as well. Just as before, the regional space was what ultimately mattered for UTM. By the middle of the 1950s, the army had amassed a huge amount of survey data, including several intercontinental connections, but the original plan to use Europe and North America as anchors soon ran into problems. Figure 4.14 shows the extent of these connections: enormous arcs of triangulation Territoriality without Borders

193

Figure 4.15: Fischer’s explanation of the problems encountered when the Army Map Service tried to extend the European and North American survey blocks south to include Africa and South America. She described this diagram as “the Potato and two Eggs,” with the irregular surface of the potato-like earth conflicting with the smooth egg-shaped ellipsoids used in surveying. Although the discrepancies here are shown greatly exaggerated, notice how the International Ellipsoid fits Europe quite well but is too large for the world as a whole. The ellipsoid for North America is instead too small (and too flattened). These mismatches threatened to introduce major errors into coordinate calculations in the Southern Hemisphere. Image redrawn for clarity from Irene Fischer, “Is the Astrogeodetic Approach in Geodesy Obsolete?,” Surveying and Mapping 34 (June 1974): 128.

connected Alaska to the southern Andes and Scandinavia to South Africa, and east-west systems stretched across North America and Eurasia. As these large systems were calculated, however, an army mathematician named Irene Fischer— an Austrian Jewish refugee who had studied with the famous Vienna Circle and would go on to be one of the major geodesists of the twentieth century— discovered that the ellipsoids used in Europe and North America could not be extended south without serious errors. Figure 4.15 shows how she later explained the problem: each ellipsoid was a good match for its continent, but the allegedly International Ellipsoid was “much too big” for Africa, and the North American ellipsoid led to “intolerable” coordinate errors in Chile— up to 140 meters.77 Assembling a global system by extending preexisting systems simply would not work. This problem provoked two responses. The first, which was pursued in different ways by both the army and the air force, was to simply make an entirely new global system. Fischer, for example, started in 1956 by making a new calculation, using the UNIVAC, for the size and shape of the earth from all available data; she named the result the Hough Ellipsoid, after her boss. It was soon used to guide the first American satellites, and satellite data was used in turn to improve the calculations further.78 The air force pursued the same goal using global gravity measurements instead of survey data (gravity was especially important for missile targeting), and soon presented its own results. Although both army and air force geodesists made a strong case that their approach was the more rigorous of the two— with the main figure behind the air force project even arguing that his approach signaled a “new era of geod194

Chapter Four

esy”—in 1959 the two solutions were simply averaged together to create the first Department of Defense system, the so-called World Geodetic System of 1960, or WGS 60.79 Later revisions to this system (and similar nonmilitary versions) would integrate ever more global data to directly tie continents together using satellites, without relying on ground triangulation at all. But while this approach would eventually lead to the eclipse of traditional surveying altogether, its global focus meant that it was initially quite disconnected from the actual maps and coordinate systems in use around the world. The second response to the ellipsoid problems was simply to continue the original regionalist project as an end in itself. Faced with the inadequacy of extending the North American ellipsoid south, for example, in the 1960s Fischer created a new system tailor made for South America that was quickly embraced by the Pan American Institute of Geography and History, the same intergovernmental organization where Hough had first presented UTM. The Army Map Service also changed course in Africa. Instead of hemispheric or even continental consolidation, it ended up supporting two regional systems: one stretching from Sudan to Senegal, and the other connecting most of the countries south of the equator. A new ellipsoid was also adopted in Australia as part of its conversion to UTM in the mid-1960s.80 While these initiatives might seem to contradict the US’s larger global ambition, they were motivated by the same pragmatism as before. As Fischer put it, each of these regional projects was simply “a coordinate system chosen for its suitability for a specific purpose”— and that purpose was “internal consistency,” not globalism for its own sake.81 Taken as a whole, the history of UTM in the fifteen years after its initial design is thus not as straightforward as the smooth uniformity of the grid itself might suggest. UTM did not simply replace all existing coordinates with a ready-to-wear global system— not only did it spread quite gradually, but it also operated primarily at a regional scale. Indeed, of all the geodetic work associated with UTM in the 1950s, global projects like the World Geodetic System were perhaps the most tangential. Similarly, while the Army Map Service certainly saw the gridding and mapping of the entire world as within its purview, its more ambitious goal was to create a geographic system that was more than simply American, and this likewise required a more incremental approach. The global territoriality of UTM, in other words, exceeded the bounds of individual states, but it was meant to stay rather close to the ground. It relied on a dense network of ground survey and was adopted by each country individually as part of large contiguous blocks, with the larger global diagram only relevant for a few tasks of truly intercontinental scope.

UTM SINCE 1960: CIVILIANIZATION AND UNIVERSALISM By the time that Floyd Hough retired from the Army Map Service in 1957, the installation of UTM as a common military system was well underway. It was Territoriality without Borders

195

in use throughout Europe and would soon be available in the Middle East, Africa, Southeast Asia, and the Americas. A few large countries would not be converted until the 1960s and 1970s, and a number of well-established systems of colonial British and French origin would be retained through the end of the Cold War, but otherwise the success of UTM was undeniable.82 Yet the story of UTM after 1960 is not primarily a military one. Instead, UTM continued to expand its reach by slowly becoming a civilian system as well, where it was adopted for unforeseen uses, often without any urging from the US Army. By the 1990s UTM was being used as a straightforward, neutral system in places that might never have joined an American military project. The end result of this gradual diffusion was that UTM became a genuine alternative to latitude and longitude in most of the world. The earliest and most obvious way that UTM was civilianized was through its adoption by national survey agencies as a domestic grid system. In the early 1950s, Hough had insisted— for both political and mathematical reasons— that UTM was not suitable for engineering or cadastral surveying, but within a decade his advice was routinely being ignored, since maintaining one system was always easier than maintaining two. UTM was especially beloved by those involved in economic development projects that spanned international borders, since UTM was transnational by design. (The fact that many of these countries were located in the tropics, where UTM is most error prone, was an irony not lost on disappointed American mathematicians).83 In the early 1970s, even the United States moved toward convergence, when civilian officials from the US Geological Survey proposed fully replacing the interwar State Plane system with UTM, which had already been included in the margins of their maps for more than ten years and was continuous into Canada and Mexico. Although State Plane was retained (albeit recalculated in meters rather than feet), UTM lines began to be drawn more prominently and gained official sanction as an alternative all-purpose civilian grid.84 UTM was newly adopted for oil prospecting by countries in the Middle East around the same time, and even China began using UTM for some purposes by 1980. Within a few years of the breakup of the USSR, several former satellite states had likewise converted their coordinates from the Soviet system.85 As a result, the ideal of layered territory described by Hough has been increasingly replaced by a single global system, and the norms of the interwar period have been entirely upended. In the early twentieth century, detailed coordinates had been the exclusive domain of national states. During the Cold War, there were generally two overlapping systems: national and transnational (either UTM or the Soviet alternative). But by the turn of the twentyfirst century, many of the national systems had faded away and been fully upstaged by the global coordinates of UTM. Although the US military took a strong activist position in the middle of the century, the later shift happened largely as a result of individual states adopting the new systems of their own accord. 196

Chapter Four

But the story of UTM also stretches beyond the confines of individual states, and it has been found to be quite useful for international projects of all kinds. Archaeologists, for example, began using UTM in the late 1960s. Echoing earlier grid language, an American doing fieldwork in California described it as “a means of permanently describing the location of an archaeological site,” one that was more reliable than existing domestic systems. Another went further, championing UTM because it “cover[ed] uniformly those areas not having any system of political boundaries or having various systems of political boundaries.”86 Botanists, zoologists, and demographers began using UTM around the same time because it provided a framework of standardized square units that could be used to compare distributions of flora, fauna, and people around the world. Unlike areas bounded by latitude and longitude, which were not the same area at different latitudes, the main UTM squares were always one hundred kilometers on a side.87 In both cases, what most recommended UTM for widespread use was that it was already available on most maps, hiding in plain sight. And for many scientists, its military origin only made it more appealing, since it was therefore unlikely to disappear any time soon. More recently, UTM coordinates have also begun to appear on international boundary treaties. They were used for the 1994 peace treaty between Israel and Jordan, as well as the boundary delimitation between Ethiopia and Eritrea in 2006.88 The fact that both of these treaties gave ultimate legal priority to a list of coordinates rather than to physical boundary stones was notable but not particularly novel.89 What was novel about both treaties, however, was that their lists of coordinates included two columns: one for latitude and longitude, and one for UTM. These coordinates were presented as perfectly interchangeable. Given that boundary treaties are designed to admit only one reading and to clearly set out an “order of precedence” in boundary description (the Israel-Jordan treaty in particular has been praised by surveyors for its lack of ambiguity), the assumption of complete equivalence between the two systems is especially striking.90 Finally, the Internet and GPS have enabled UTM to be used in nearly every situation where latitude and longitude once prevailed. Almost all GPS receivers give the option of displaying UTM coordinates, and the benefits of UTM have been publicized on homemade websites since the earliest days of the World Wide Web. One such guide, first posted in 1995, counseled that “if UTM gridded maps are available, don’t waste time fooling around with Lat. & Long, at least on land. . . . For ease of conversion from map to GPS and back, you can’t beat a grid system.” Another, from 2003, gave the difference a rather millennial spin: “First there was ‘latitude’ and ‘longitude’ map coordinates . . . then came flat thinking grid technology!”91 Since 2006, UTM coordinates have also been a standard option for both Google Earth and Wikipedia. Far from being confined to the margins of paper maps, UTM is now more accessible than ever.92 Territoriality without Borders

197

How should we understand the success of UTM? The simplest explanation would simply point to the enormous resources poured into the system. To make a global grid, the US Army had to undertake massive calculation projects that had eluded geodesists for decades before the war, leveraging not just the latest cutting-edge technology but also the forced labor of defeated Germans; they also had to use a variety of strategies to get access to data that had traditionally been treated as nationally privileged. And making this system stick required assertive diplomatic coordination, shouldering the huge expense of redrawing and reprinting military maps throughout the world, and training millions of soldiers in new techniques. Given how much capital has been invested in the system— financial, political, and human— it is hardly surprising that it has remained essentially unchanged for almost seventy years.93 But the success of UTM can be seen in a rather broader light as well. Comparing UTM to all the alternatives that have been used (or proposed) since the 1940s, it seems that what is at stake here is not just the diffusion of a certain technical system, but the persistence of a much wider territorial sensibility. For example, although the Soviet system was never civilianized, its postwar trajectory paralleled UTM quite closely. In 1952 its use was mandated throughout eastern Europe for military purposes, and the USSR applied it to its own maps around the world; it was also used in Mongolia and Communist China. Except for a few mathematical details that kept it incompatible with UTM, the basic design of both systems was essentially the same.94 Likewise, since the late 1950s a variety of articles have appeared in specialist journals about ideal designs for grid systems. With only a few exceptions, cartographers have overwhelmingly found systems of north-south Transverse Mercator belts to be the best solution, for small and large areas alike. There has been almost no interest in jigsaw-puzzle systems like the US State Plane or the British war grids.95 In other words, even though the American military was responsible for the particular details of UTM, the larger question being asked has remained the same throughout the world since the late 1940s— namely, what is the best coordinate system to use on a continental or worldwide scale, ignoring political boundaries? If that is the question, then a system like UTM has seemed like the perfectly logical choice. And this has been the goal not just of global militaries, but also of international development planners, oil prospectors, archaeologists, botanists, demographers, peacekeepers, and even recreational hikers. There is indeed something universal about the Universal Transverse Mercator: a universal goal, shared by states, private organizations, and individuals alike, to make geographic space continuous across borders.

CONCLUSION: GRIDS AND GLOBAL SPACE In the time since they were first invented during World War I, grid systems have become ubiquitous. They have only rarely been noticed or appreciated— 198

Chapter Four

even by map users or historians of cartography— but their geographic power is undeniable. They ease the frictions of space for both the novice and the expert, they organize territory across international borders, and they make faraway lands accessible— often literally so, in the case of artillery and longrange missiles. They emerged entirely from the paper world of maps and the mathematics of map projections, but they ended up replacing the traditional god’s-eye view of the map with a hybrid system that exists just as much as a full-scale landscape of physical monuments as it does as lines on a sheet of paper. And indeed, by the time of World War II or UTM, almost none of the mathematical discussion of grids is about maps at all. Instead it is about practical utility for the everyday user, in the real world. Here the comparison to latitude and longitude is again apt. During the debates in the late 1940s between the US Army and Air Force about the benefits of their respective systems— debates which were often quite heated— much of the disagreement hinged around a conceptual difference about the nature of coordinates. The air force (and the navy as well), coming from a long tradition of celestial navigation, saw the graticule of latitude and longitude as a natural system and simply could not imagine a map without it. They argued that only latitude and longitude provided the “simplicity, regularity, and uniformity” necessary for global operations on a spherical earth and dismissed UTM as fragmented and geographically counterintuitive— in short, “unsuitable in every way.”96 This understanding of latitude and longitude is common; it is the one presented in textbooks, encyclopedias, and popular histories of exploration. Latitude and longitude are real, while grids are just a mathematical trick used to pretend that the world is flat.97 Mathematicians at the Army Map Service, however, saw the problem in entirely different terms. Evaluating the objections raised by the air force, John O’Keefe wrote in his notes that the debate between grid and graticule was “not an argument between two entirely different sorts of things, but between two systems of coordinates.” And evaluated in those terms, there was very little to recommend latitude and longitude as a unifying system: “stripped of its historic incrustations, legacies from Sumerian astrologers and English sea-captains, the graticule is merely a badly designed grid.” It relied on nondecimal units (some positive, some negative) that made calculations unnecessarily complex; it was also nonconformal and became unusably distorted only a few dozen kilometers from the equator. These were mistakes that “no competent modern designer of grids would permit.”98 For O’Keefe, the lines of latitude and longitude were simply one of many map projections that could be applied to the earth— namely, the equirectangular— and by no means the best. In a later journal article, one of his colleagues expressed this view in somewhat broader terms: “A coordinate system provides a convenient way to express mathematical properties. There is a wide choice, depending on what one considers ‘convenient.’”99 In other words, neither artillery grids nor the lines of latitude and longitude Territoriality without Borders

199

are natural systems; both are artificial human inventions that made certain tasks easy and others difficult. The hundred-year history of grids lends much more support to O’Keefe’s position than to the air force view, especially since grid systems and the graticule are equivalent not just mathematically, but practically as well. While it is certainly true that (some) explorers and navigators locate themselves by the sun and stars, the way that everyone else finds their coordinates— including most postwar captains and pilots— is through some kind of governmental system: a paper map, a list of grid monuments, or the readout of a radio receiver. And for precision on the scale of meters, these governmental systems are the only way to find a reliable coordinate.100 The question for the soldier (or the archaeologist, or the hiker) is not which of these coordinates is real or true, but rather which of these is useful. Knowing that my house is located a few miles northeast of 41°N, 73°W is helpful for situating myself with respect to the equator and to the capital of nineteenth-century naval power, but not much else. Knowing instead that I live at 18T-XL-7419-7568, while not as familiar, can be quite helpful indeed. (Especially if I need to call for reinforcements.) By the 1960s, then, there is no reason not to see UTM as fully interchangeable with latitude and longitude, both mathematically and administratively. With this in mind, the history of grids can be told in two different ways. One story would stress the relentless geographic expansion of grids as they moved from organizing artillery on the western front of World War I to organizing a wide mix of activities all around the world. This narrative, which has largely structured the last two chapters, is a globalizing one: grids moved from being relatively local to being national to being transnational. This story also puts the expansion of American power front and center, as the US Army Map Service aggressively upstaged the patchwork of interwar domestic grids (and the British grids of World War II) with a single universal system that, like many other postwar American universalisms, advanced American interests by blurring the earlier distinction between domestic and international space. But considering grids from the perspective of several centuries rather than just a few decades, the other interpretation would instead see one global system— latitude and longitude— being challenged and partially displaced by another: UTM, along with its Soviet twin. To be sure, this displacement was an enormous undertaking, requiring sustained attention not just from geodesists and other scientists, but also from dozens of national governments (the US above all) that had to provide real-world infrastructure and negotiate the boundaries of national autonomy. Yet in the grander context of epistemology and the human experience of geography, this work is perhaps less important than the basic shift from one way of organizing space to another: adding centimeter-level precision, creating a recentered and embedded subjectivity, and seeing the miniature world of maps as secondary to the full-scale system of coordinates. Zooming out to this broader narrative also underscores 200

Chapter Four

exactly what was left behind as authoritative representational mapping lost its privileged status as a tool of international legibility. For the cartographers of the early twentieth century, universalism was about timeless truth, scientific objectivity, and the progress of (European) civilization. For US Army mathematicians, these goals were largely irrelevant, and transnational universalism was much more closely linked to transnational intervention. What mattered, after all, was convenience.

Territoriality without Borders

201

PA R T  I I I

Electronic Navigation and Territorial Pointillism

CH A PT E R F I V E

Inhabiting the Grid: Radionavigation and Electronic Coordinates, 1920– 1965

On the cloudy night of October 30, 1944, a British bomber leaves northern England on its way to Germany. On board, the navigator is primarily occupied with two objects. The first is a black box with knobs and a circular display, as in figure 5.1. The navigator spends most of his time fiddling with these knobs trying to make two wavy signals align on the screen; doing this results in two numbers— something like 49.1 and 4.8. At least one of these numbers is constantly changing and the navigator’s box requires ongoing attention. The second object is a simple map known as a lattice chart that shows a dense network of colored curves, as in figure 5.2. Each of these curves corresponds to one of the numbers from the machine; finding the intersection of the lines labeled 49.1 and 4.8 puts the bomber just north of Leeds. From here, it will head east over the North Sea to join in formation with more than nine hundred other aircraft, all carrying similar equipment, before turning south for yet another bombing raid on German civilians— this time, in the suburbs of Cologne. Hidden by the clouds, not a single plane will be lost.1 This combination of a black box and a map was known as the “interpretation system” for a radionavigation system codenamed Gee that first came online in March 1942. The system was an urgent necessity, since the British had realized that most of their nighttime bombers weren’t able to navigate anywhere near their intended targets, let alone bomb accurately once they arrived. Gee stood for grid; in the words of Robert Watson-Watt, the leader of the British radar project, the goal was to “unfold [an] electronic grid over Germany.”2 Gee was only the first of many radionavigation systems to use the language of the grid. In 1965, for example, an American engineer presented a newly developed system in exactly the same terms— it was “a repeatable, predictable, electronic grid”—and descriptions of the Global Positioning System from its 205

Figure 5.1: Mock-up of the black boxes used for radionavigation in an Avro Lancaster, one of the principal heavy bombers of the British Royal Air Force during the second half of World War II. The large circle on the left is an oscilloscope display that would show two wavy lines; using the knobs below to adjust the position of the wavy peaks would result in a pair of numbers. Photo courtesy Peter Zijlstra, http://www.pa0pzd.com.

initial design in the 1970s into the early twenty-first century have likewise described it as a “worldwide common grid” for all military operations.3 Spatially, these grid-like navigation systems are remarkably similar to the cartographic grids described in the previous two chapters. In all cases, the goal has been to create an invisible framework of unique, easy-to-determine coordinates operating at full scale. Even when radionavigation systems have used nonrectangular coordinates— such as the curved lines of Gee or the latitude and longitude displayed on a GPS receiver— the epistemic concerns are still convenience and efficiency rather than objectivity or truth, and the user’s experience of space is an embedded one. Artillery grids and radionavigation also share a similarly American trajectory of global expansion: just as the US Army installed the global grid of the Universal Transverse Mercator on top of a patchwork of national and regional systems, a few decades later GPS was specifically designed to render dozens of less ambitious systems obsolete and create a single coordinate grid for the entire earth. (And this American system was, once again, matched by the Soviets.) Perhaps more important, both UTM and GPS have 206

Chapter Five

Figure 5.2 (see gallery for color version): Lattice chart showing central Great Britain and eastern Ireland. The two numbers from the black box in figure 5.1 correspond to two of the (colored) hyperbolic lines crisscrossing the map. The intersection of 49.1 (purple) and 4.8 (red) is in the upper right, just north of Leeds. This is a postwar reprint of a British map dated February 1944 (NARA Cartographic Records, RG 77, box 54 of 215, folder “907.2: Raydist, Decca”).

also exceeded their military origins and become widely adopted as universal, neutral, even quasi-natural systems. Since the 1960s, UTM coordinates have been found useful for things like international development work, archeology, and boundary treaties. GPS, of course, has been adopted much more widely still. It is used not just by pilots and scientists but also by commuters and teenagers, and GPS receivers have been incorporated into everything from shipping containers to stoplights. The goal of these final two chapters is to understand the radicality of GPS within the long history of radionavigation and the gridding of space. In hindsight, GPS seems like such a simple, obvious technology— albeit a sophisticated, almost magical one— that it is difficult to see it as only one possible solution that emerged in response to a particular set of political-historical problems. After all, there is no necessary connection between radionavigation and coordinates: radio can be used in many different ways, and for certain navigational tasks a grid can be rather unhelpful. Similarly, as tempting as it might be to see a straightforward linear trajectory between the limited coverage and Inhabiting the Grid

207

esoteric interface of Gee and the final perfection of GPS, the reality is much more complex. For example, when GPS was first designed, its overall approach departed sharply from the prevailing sensibility of most navigators and engineers. And even though it has ended up displacing nearly all other systems, the reasons are as much political as technological. These are the historical problems to be explained. What kind of departure was GPS? How should we understand its success? What possibilities did it shut down? To answer these questions, I have divided my analysis both thematically and chronologically. This first chapter focuses on the intersection between radio and coordinates, from the earliest systems of the 1910s through World War II and into the first decades of the Cold War. The next chapter analyzes the turn toward global systems in the 1960s and the eventual creation of GPS. My argument follows closely from this division. I locate the major territorial shift in the decades surrounding the war: this is when navigation systems, similar to both paper maps and cartographic grids, shifted from a focus on the bounded space of the territorial state to a much more fluid sense of permeable boundaries and regional consolidation. This is also when the logic of points, grids, and coordinates first emerged with consequence. Although the global systems that arrived in the 1960s did indeed imply a rather assertive political vision, their globalism was (and is) only relevant for a very small number of tasks and did not represent a fundamental break in the spatial logic of radio. Instead, what ultimately set GPS apart when it came online in the 1990s was its universalism: unlike all earlier systems, it was monopolistic— and often forcefully so. Compared to its predecessors, it thus did not inaugurate an entirely novel experience of geographic space, a new approach to territory, or a newly global sensibility. It did, however, eliminate a great number of alternatives. The most important of these alternatives was a much more structured way of using radio to organize space. For most of the twentieth century, radionavigation systems can be divided into two main types: those that enable point-to-point navigation— so-called “track guides”—and those that provide “area coverage.” Point-to-point systems create strictly defined routes, while area techniques allow for navigation anywhere within a wide expanse. Geographically, the difference is thus between a network of one-dimensional lines and a two-dimensional field of points.4 Although this distinction is somewhat blurry from an engineering point of view, it has nevertheless been quite powerful at an operational, geographic, and regulatory level, and much of the history of radionavigation can be told as a competition between these two techniques. The gradual eclipse of point-to-point navigation by area systems like Gee and GPS, however, should again not be seen as a simple function of technological progress or US military expansion. Track-guidance systems were often much more accurate and reliable, and the most important of these systems have been solidly American. Instead, this transition should be seen more generally as a part of the shift from bounded to unbounded territory. Maintaining a network of track guides requires building an antenna at every 208

Chapter Five

junction, while area coverage can easily be used for navigation beyond the limits of political control. These techniques are not strictly incompatible— and even today some track guides continue to exist— but they imply widely different territorial goals. This chapter is divided into four sections. I begin by analyzing the various approaches to radionavigation pursued before 1939, especially in the fledgling commercial aviation industry. During this time, two distinct technological strategies emerged— a point-to-point system in the United States and an area system in Europe. Both, however, reinforced the political split between national and international space. My second section then traces the development of new systems by the US, UK, and Germany during World War II. Since most of these systems were designed to guide bombers to their targets, it suddenly became necessary to navigate over unfamiliar, featureless, or otherwise hostile terrain and to extend territorial legibility across international borders. The distinction between track-guidance and area systems was important here as well, even as the politics changed quite dramatically. The second half of the chapter analyzes the postwar use of these new systems in two domains: surveying and civilian navigation. In the first, new radiosurveying techniques were an important part of the American-led project of consolidating survey networks at a regional or continental scale; at the same time they also gained wide popularity for commercial surveys, especially offshore surveys, in both the US and Europe. As a result, these systems ended up disrupting boundaries of all kinds— not just boundaries between countries, but even the boundary between land and water. In civilian navigation, the new systems were part of a similar project of transnational consolidation, but here the politics were much more fractious. Immediately after the war, the hope was that international agreement would recognize the best systems and that these would then be installed around the world; the main conflict was between American and British proposals, and there were again major clashes between point-to-point and area navigation. By the mid-1960s, however, it was clear that global space would be organized neither through the politics of standardization nor through a network of stable pathways, as both were superseded by new computer-heavy techniques for combining signals from multiple systems into a single whole. Surveying and navigation each ended up constructing a similar transnational space— one that was continuous across political and geographic borders without being uniformly global. My analysis here thus relates to my overall argument in two ways. First, the close analogy between cartographic grids and certain kinds of radionavigation makes it clear that the geo-epistemic shift from representational maps to a more pointillist approach to space was a relatively broad change in attitudes, goals, and practices across the mapping sciences as a whole; it was not just about the invention of satellites, digital computers, or any other technological device. Second, the long history of radionavigation also makes it clear that the spatiality of GPS was not unrolled all at once by a single monolithic power. European deInhabiting the Grid

209

signs were just as important as those from the US, and there have never been sharp lines between military, civilian, and commercial technologies. If anything, the transnationalization of space in the decades before GPS was driven as much by political failure and commercial competition as by any top-down military project.

RAILROADS OF THE SKY VS. THE AIR OCEAN Before World War II, there were two main approaches to radionavigation. Air travel in the United States relied on a point-to-point system called the Radio Range, while navigation in Europe used a wide-area system known as radio direction finding, or D/F.5 These were not just different technological solutions; they were also different ways of understanding air travel and managing geographic space. The Radio Range created stable paths that were seen explicitly as a kind of aerial railway, while D/F had much in common with the nautical logic of lighthouses and was used by both ships and aircraft alike. By physicalizing these two analogies— the railroad and the ocean— radio technology channeled the much broader use of these metaphors in early aviation. These metaphors were in common use in both the US and Europe, but they implied different practical goals, different relationships to the state, and different ways of making aviation a stable, dependable civilian industry. The two radio strategies of the US and Europe thus participated in two very different political and geographic projects. The railway-like Radio Range was a project of territorial consolidation, domestic services, and subsidies. The lighthouse-like D/F, in contrast, was part of a wide-ranging discussion about international coordination and the limits of sovereignty. Both strategies, however, shared a similar respect for the ideal of state autonomy. Wherever it appeared, the railroad metaphor was invoked to stress the importance of ground installations; it was most directly at work in the idea of an “airway.” The implicit argument was that regular air service was not just as simple as flying from point A to point B. Instead it required stable pathways composed of things like radio beacons, rotating searchlights, meteorological facilities, emergency airfields, refueling posts, and wireless telegraphy stations. As an American airline executive put it in 1926, “an airway is just as truly on the surface of the earth as is a railway.” Figure 5.3 shows this same sentiment in visual form: the “railroads of the air” required serious construction.6 As a result, the politics of the railroad analogy were aligned with state sponsorship of ground support and explicit governmental attempts to establish new commercial and administrative links. In France, for example, aviation pundits borrowed railroading vocabulary when arguing for state support for signals, weather services, and land acquisition.7 The United States and Canada likewise each organized their air systems around state-supported “transcontinental” airways (with a “terminus” on both coasts), and in both Africa and 210

Chapter Five

Figure 5.3: The railroad metaphor is accompanied immediately by photographs of the physical structures required to support aviation: buildings, runways, hangars, and towers. This article, meant to drum up support for the airport in Buffalo, New York, goes on to emphasize the importance of regular weather broadcasts along a pilot’s route, along with adequate lighting for night flying. At the airport itself, his recommendation is for “oil lanterns of about the same size and shape as those used by railway brakemen.” John M. Satterfield, “Railroads of the Air,” Buffalo Journal of Commerce 21 (July 1927): image on 4, quote on 28.

the Americas ambitious continent-spanning airway schemes were seen as the direct heirs of Cecil Rhodes’s abortive Cape-to-Cairo line and the unbuilt PanAmerican Railway.8 And when keeping tallies of the progress of aviation in various countries, the usual unit of comparison was, in the tradition of railroad one-upmanship, miles of airways.9 The airway, in other words, was a physical structure— an expensive kind of territorial engineering— with distinctly national or imperial overtones. Over the course of the 1920s and 1930s, the Radio Range gave physical form to this rhetoric. Figure 5.4 shows how it worked: a set of radio antennas sent out audible Morse code signals in four directions, with slight overlaps. A pilot listening on headphones would hear either an A (dot-dash-pause) or an N (dash-dot-pause) depending on the quadrant; when flying “on the beam,” the A and N would merge to create a steady tone that indicated the correct course— either directly toward or directly away from an antenna.10 By adjustInhabiting the Grid

211

Figure 5.4: The directional beams of the Radio Range, developed at the US National Bureau of Standards in the 1920s. A single antenna (shown here as a single dot, but actually composed of four interconnected, closely spaced aerials) transmits Morse code signals for “A” and “N” in four directions, with varying power; the shaded circles show signal strength. The dark gray beams are an artifact of adjacent signals blending into one another to create a constant tone; the diagram in the lower right shows how the Morse code “A” and “N” signals interlock. The edges of the beams are thus somewhat indistinct— hence the “twilight” names for the edges— but note that these beams can be quite narrow even though none of the actual transmissions are at all focused. By changing the power in the aerials, the beams can be oriented in almost any direction.

ing the power and orientation of the four directional transmissions, these narrow equisignal zones could be pointed in almost any direction, and multiple beams could be strung together to make a stable airway, with each antenna typically sited near an airfield. These kinds of directional pathways were a continuation of earlier American navigational techniques— before radio, pilots would commonly “steer a range” by using the visual alignment of two distant landmarks— but by 1941, a Radio Range primer noted that pilots, especially inexperienced ones, would often “think of a beam as a railroad track . . . us[ing] the published course as if it were a pair of rails.”11 212

Chapter Five

The Radio Range was an explicitly nationalizing project. Its development was wholly sponsored by the US government, and the persistent goal was to create a national network of airways for internal services, especially airmail and domestic military aviation. It was designed in the 1920s by engineers at the National Bureau of Standards and the Army Air Corps, with funding coming first from the Army Air Service and then from the Department of Commerce.12 One of the major features of the Radio Range was that the only equipment needed in the aircraft was a simple radio receiver; not even a transmitter was required. This was especially important in the mid-1920s, since the delivery of airmail— the first major civil use of aviation in the US after World War  I— was moving from direct government operation to contract flying, and the private planes used were generally small and cheap. The Radio Range was thus a way of subsidizing these routes: its designers explained that “the complicated and expensive apparatus is on the ground . . . maintained by the Government.” This balance between public and private again followed closely from railroad precedent, with mail used in both cases to make passenger service more economically viable.13 As a government service and a strategy of domestic consolidation, the Radio Range was remarkably successful, and the Radio Range network defined American aviation for decades. It was first installed on airways around 1929, starting with the Transcontinental.14 By 1933 there were eighty-two Radio Range stations in operation; by the end of the decade there were more than 250 Ranges covering the entire country, with routes often closely following existing rail lines (see figure 5.5). At its peak around 1950, the network included almost 400 stations, the last of which was turned off only in 1974.15 But even as the Radio Range itself became obsolete and was gradually replaced by its successor (a more flexible and reliable system known as the VHF Omni-Range, or VOR), American air traffic control continued to be structured as a linear network of stable radio pathways. The oceanic metaphor was not incompatible with these domestic concerns, but it conjured a rather different political vocabulary. While it could sometimes be used to make a case for public investment, it was more commonly part of debates about international air law and the limits of national sovereignty. In contrast to railway talk, these debates often assumed that the atmosphere was inherently navigable: what mattered was the “natural state” of the air, not ground services.16 Especially before World War I, oceanic analogies appeared widely in legal debates about the extension of territorial sovereignty into a country’s airspace, with advocates for the “freedom of the air” invoking ships’ long-standing right of innocent passage through territorial waters. But even after the peace talks of 1919 definitively settled the question in favor of total sovereign control, questions about the inherent “freedoms” of civil aviation continued to be debated well into the 1950s, and the seventeenthcentury theorist of international waters Hugo Grotius continued to be cited in arguments about unrestricted competition, cabotage rights, and the potential Inhabiting the Grid

213

Figure 5.5: The railroad-like airway system of the United States in the late 1930s. The top map shows the orientation of individual Radio Range stations; the bottom map shows the routes as actually used. Note how the east-west “Transcontinental Airway” stretching from San Francisco to New York followed much the same route as the original transcontinental railroad of the 1860s. Top map from Ronald Keen, Wireless Direction Finding, 3rd ed. (London: Iliffe and Sons, 1938), 484; bottom map from Civil Aeronautics Authority, First Annual Report of the Civil Aeronautics Authority (Washington DC: USGPO, 1940), appendix B.

for military espionage by overflying airplanes. The oceanic metaphor was thus immediately aligned with problems of foreign relations rather than internal development.17 Direction-finding technology channeled much of this discussion; it provoked debates about pilot autonomy and relied more on international standardization than state-sponsored construction. As the name implies, D/F equipment simply indicates the direction to a radio source; it works by exploiting the directional reception properties of certain kinds of antennas. With a ring-loop antenna, for example, the signal from a distant radio source will be greatest when the antenna is aligned edge-on with the transmission and at a minimum when the open ring faces the source; direction can be found by simply rotating the antenna. For aircraft, this principle was turned into a navigational system through the creation of a network of direction-finding stations on the ground that could track transmissions from planes flying overhead. A pilot would simply use a radiotelephone or wireless telegraphy to ask ground stations for D/F readings of his18 transmission; staff at these stations— at least two, but usually three— would coordinate among themselves to combine their readings and then radio the result back to the pilot in whatever form was most convenient.19 (What might seem like the simpler solution— D/F equipment in the plane— was actually more complex, for both electrical and navigational reasons.)20 Like a mariner using lighthouses near shore, a pilot was thus not confined to predetermined routes— at times even to the chagrin of air traffic planners.21 The use of D/F for aviation in fact followed directly from marine precedent, and the same systems were used for sea and air alike. The first experiments with direction finding dated to the 1890s and early 1900s— before the Wright brothers had even taken their first flight— and the first D/F stations were established for shipping in the early 1910s. The UK Post Office began a coastal D/F service in 1912, and stand-alone equipment was also installed on large ships like the Mauritania. The first use of D/F in aviation came during World War I, when Germany used ground stations to direct not just its warships and U-boats, but its Zeppelins as well; the Allies in turn installed a network of D/F stations in Great Britain and northern France to track the German fleet and shoot down the airships.22 Even the basic method used to coordinate D/F readings carried an oceanic flavor: D/F ground staff would typically combine multiple bearings by drawing intersecting lines on a map in exactly the way that ship navigators were taught to use lighthouses for determining their location at sea.23 The growth of the European D/F network roughly paralleled the expansion of the Radio Range, but both its geography and its structure were irreducibly international. Although marine services were installed in both Europe and North America soon after World War I, the European aeronautical network was first organized in the late 1920s, primarily to facilitate travel across the English Channel (see figure 5.6). By 1938 there were roughly a hundred aeroInhabiting the Grid

215

Figure 5.6: Aeronautical direction-finding (D/F) stations in western Europe as of 1931. The aircraft in the lower right has sent out a request for its location to Brussels and Le Bourget, and these two stations have used directional antennas to determine their bearing to the plane. After coordinating among themselves, one of the ground stations would radio back to the pilot the name of a nearby town that the pilot could find on a map. Station locations from Gerald C. Gross, “European Aviation Radio,” Proceedings of the Institute of Radio Engineers 19 (Mar 1931): 346.

nautical stations throughout Europe.24 Although each station was financed by its host country and often built to national designs, D/F communication protocols— along with various other navigational standards— were regulated by the International Commission for Air Navigation (which, again, did not include the United States). In the late 1930s, the ICAN map committee even discussed plans to standardize D/F maps as another adaptation of the International Map of the World.25 This internationalist approach was very well suited to the air network of Europe, which consisted of a tangled and shifting web of short hops between closely spaced cities— often rearranged as countries changed their airspace and customs policies— rather than the durable trunk lines of the US.26 The interwar period was thus defined by two sharply contrasting understandings of how radio could organize space. The US was crisscrossed by 216

Chapter Five

railroad-like beams defining stable paths as part of a project of domestic consolidation, while western Europe was dotted with lighthouse-like beacons that located pilots using points and allowed flexible navigation across (wellrespected) international boundaries. The one-dimensional Radio Range required only headphones, while the two-dimensional European solution relied heavily on maps.27 The overall goal in both cases was to ease the frictions of space, but neither solution was easily universalized— they were responses to different political geographies, different strategies of state support, even different types of aircraft. By the end of the 1930s there were signs that this clear distinction might gradually fade, but no sweeping changes took place before the start of the war. European engineers did begin adopting American beam technology (especially as blind-landing equipment on individual runways), and dozens of installations were made in the UK and Germany by subsidiaries of the American multinational ITT. There was also great progress in Europe on airborne D/F equipment that could be used for flexible point-to-point flying in a network of radio beacons, and these so-called “radio compass” receivers were soon mandated in the US as well. The system that would eventually replace the Radio Range— which was already under development at RCA by 1936— was likewise presented in both the US and Europe as another possible compromise between fixed paths and free flying.28 However, this inchoate convergence, which may well have blunted the contrast between national and international navigation had it come to pass, was sharply diverted by the war. Rather than any gradual stabilization of radio techniques, the 1940s instead saw a great multiplication of new systems.

THE INVENTION OF ELECTRONIC GRIDS IN WORLD WAR II World War II provoked a rapid and somewhat haphazard reconsideration of radionavigation strategies. Both D/F and the Radio Range continued to be used extensively, but there was also wide development and experimentation with new techniques. At least a dozen new systems were initiated by the US and the UK, while no less than twenty-five were pursued by Germany. (Almost none, however, were developed by Japan, Italy, or other countries.)29 Unsurprisingly, these systems rarely lived up to engineers’ expectations (or postwar rhetoric), but the confrontation between new technological systems and the realities of war led to a significant reorientation of radio away from the realm of national or international space and toward a profoundly transnational territoriality. Wartime radionavigation strategies can be divided into three types, each with a different geographic logic and postwar trajectory. The first two were primarily offensive and were used extensively for blind bombing; one used intersecting beams and was deployed by Germany, while the second used various distance-measuring systems and was developed mostly by the Allies. The Inhabiting the Grid

217

third kind of radionavigation was a new class of area-coverage systems, including Gee, that created a lattice of electronic coordinates. These systems were used for basic navigation and did not push any limits of precision or lethality, but their supporting role was quite substantial. (A common quip among British radar engineers was that D-Day should have rightly been called “GDay.”)30 None of these three techniques worked entirely as planned; the latter two, however, did end up leading to new geographic projects. The distancemeasuring systems, while never a bombing panacea, were quickly adopted for use in wartime survey and reconnaissance; the result was that radio became seen as a “precision yardstick” of unprecedented length and precision.31 The area-navigation systems, in turn, were a new kind of geographic infrastructure that combined the territorializing presence of the Radio Range with the flexibility of D/F. Following the metaphor of the grid, these coordinate systems ended up being treated as a semipermanent, even politically neutral feature of the landscape, despite their transnational scale. Neither of these two characteristics— permanence and neutrality— was entirely expected, but they were crucial for the grid systems’ postwar longevity. The German intersecting-beam systems are perhaps the best known navigational technology of the war; they were conceptually quite simple and played a dramatic role in early battles. These beams came in several varieties— the best known were Knickebein and the X-Gerät, both developed in secret during the 1930s32—but they all used principles similar to the Radio Range. Figure 5.7 shows how they worked: German transmitters were installed along the west coast of Europe, and narrow beams were aimed to intersect over various targets in England. In the first months of the war these beams were responsible for unprecedented destruction; the surprise bombing of Coventry in November 1940 was especially horrific. As dramatic as these beam systems were, however, they were disabled relatively quickly by British countermeasures. The original beams were essentially useless by early 1941, while the more advanced systems deployed by Germany in 1942 were neutralized even before they were turned on. Indeed, it is precisely the beams’ inflexibility that has made them so well known. In his memoirs, Winston Churchill famously dubbed the raids over England the “Battle of the Beams”; they were a heroic moment when a ruthless German offensive was thwarted by the “scientific intelligence” of British boffins.33 Later in the war, German engineers continued to develop new beam systems for deployment elsewhere in Europe, but these systems ended up being used less for offensive operations and more in ways similar to the original Radio Range.34 The second type of radionavigation— based on precision distance measurement— was more versatile but likewise had trouble living up to expectations, at least offensively. The logic here was similar to that of radar, except that instead of a ground station emitting pulsed signals to measure the distance to an unknown object, radar equipment was installed both on the ground and in the aircraft. The distance between the plane and the 218

Chapter Five

Figure 5.7: The German beam-based navigation systems of the Battle of Britain (as of summer 1941), showing Knickebein beams aimed at Derby and X-Gerät beams aimed at Coventry. The beams were created using the same principle as the US Radio Range, but they were designed to be much narrower to allow for highaccuracy bombing. With Knickebein, bombs could be dropped within an area about a mile square; with the X-Gerät, bombs were consistently dropped within one hundred yards of the main beam. Adapted from R. V. Jones, Most Secret War (London: Hamilton, 1978), 203.

ground station could thus be known on board the aircraft and used for precision bombing.35 Germany combined this technique with its directional beams to create two systems— the Y-Gerät (stations shown in figure 5.7) and Egon— that could supply both direction and distance from a single ground station. Like the pure beam systems, however, these were quickly disabled by countermeasures and ended up being used mostly for controlling defensive fighters.36 The British and American systems— known as Oboe and Shoran, respectively— were instead designed using two different ground stations. Figure 5.8 shows the basic idea: a bomber would fly a constant distance away from one transmitter and then release its bombs at a predetermined distance from Inhabiting the Grid

219

Figure 5.8: The British Oboe system, as first deployed against targets in western Germany. The equipment measured the distance between the aircraft and two ground stations. The pilot would keep a constant distance from the “cat” station, and bombs would be released at a certain distance from “mouse.” The accuracy of the system was very high, but because it relied on two-way communication it could only be used by one aircraft at a time, and there were constant worries that the bomber’s signal could be used by the enemy for tracking. Adapted from R. V. Jones, Most Secret War (London: Hamilton, 1978), 276.

the second transmitter. In ideal circumstances, the performance of these systems was quite impressive: Oboe could place bombs within a circle only a few hundred yards in diameter, and one of the first uses of Shoran was to destroy bridges in northern Italy.37 But similar to other precision-bombing technology (especially the Norden bombsight), these systems were hardly foolproof. Equipment problems, operator error, and uncooperative atmospheric conditions all conspired to ensure that these systems were distinguished as much by their potential as by their actual record. The Ninth US Bomber Command, for example, felt that Oboe was “oversold,” while its experience with Shoran was “discouraging.”38 220

Chapter Five

But even though these systems did not inaugurate a brave new world of error-free warfare, they did introduce a new fluidity between navigation and mapping. The most obvious symmetry was between bombing and precise aerial photography: instead of using radio to drop a bomb, a plane could take a photo of the ground below and use radio to know exactly where it had been taken. During the war, radiolocated photography was used extensively for documenting bombing raids, but it was also used for conducting stand-alone surveys. Both Oboe and Shoran (along with three Oboe-like systems known as Rebecca-H, Gee-H, and Micro-H) were used in this way, everywhere from western Europe to the jungles of Southeast Asia.39 But there was a more profound symmetry as well. Instead of using two well-located ground stations to guide a bomber (or a camera), a plane could also be used to measure the unknown distance between two antennas. Although neither Oboe nor Shoran were originally designed for this kind of survey work, the connection between bombing and mapping emerged quite forcefully during their development. Since their calculated accuracy was so high, it was difficult to distinguish equipment error from errors in the maps used to guide the bombers, and both Oboe and Shoran ended up discovering map errors during testing. Oboe, for example, was so accurate that the problem of mismatches between adjacent national triangulation networks— a problem that had plagued artillery since World War I— began to be relevant for aircraft as well, and the British ended up having to remeasure the distance across the English Channel using trial bombing runs on Belgium. It was the discovery of similar errors in the Bahamas that prompted the designer of Shoran, an RCA engineer named Stuart Seeley, to describe his system as a “radio yardstick.” By the end of the war, the complementarity of surveying and navigation was seen as “obvious.”40 The third type of navigation— coordinate-based area-navigation like Gee— was hardly free from technical glitches, but these kinds of systems were much more amenable to mass deployment than any of the blind-bombing techniques. Gee was the first such system; its debut in March 1942 sent British bombers to destroy Essen. A German system known as Sonne (“Sun”) came online that June, and the Americans’ Loran (LOng RAnge Navigation) began transmitting in October. A second British system— codenamed QM during the war but later named the Decca Navigator after its original corporate sponsor (of gramophone and record-label fame)— was ready just in time for the D-Day landings.41 These systems were less accurate than the blind-bombing systems, but they tended to be more reliable and covered vastly greater areas. Figures 5.9, 5.10, and 5.11 show the coverage of the three major systems at the end of the war. Gee spanned from northern Scotland to Tunisia; it was used not just by the British, but also by all US forces in Europe (the Eighth US Air Force installed it in 80 percent of its planes). Sonne blanketed nearly all of western Europe and was being expanded eastward. Loran receivers— more than seventy-five thousand of which were built during the war— were used in every major theater.42 Wherever these systems provided coverage, aircraft and Inhabiting the Grid

221

Figure 5.9: Gee coverage as of late 1945 for planes flying at an altitude of a few thousand feet, with names and station locations for each “chain” of three synchronized antennas. Coverage in the UK from John Hall, ed., Radar Aids to Navigation (New York: McGraw Hill, 1948), 61; continental coverage from sketch map included in “Gee Cover— Europe— Phase II,” 2 July 1945 (NARA, RG 331, entry 268, box 77, folder “Continental Cover Plan: Gee and G-H”).

Figure 5.10: German Sonne coverage as of mid-1945. Areas where two stations are available to give a full fix are shaded dark gray; in lighter areas only one line of position is available. (Note that although the range of each station is shown ending rather abruptly; in reality the signal would simply fade with distance.) Map based on a sketch map in “Radio Navigation Systems and Equipment,” an August 1945 Allied translation of a captured German original written sometime after November 1944 (NARA, RG 165, entry 79, box 1954, folder “Radio Navigation Systems and Equipment”), after 16.

Figure 5.11: US Loran coverage as of late 1945. Range increased dramatically at night due to radio reflections off the earth’s ionosphere, though with some loss of accuracy. For Skywave-Synchronized Loran, transmitters were placed so far apart that stations could only be coordinated using these reflected signals. This technique allowed for coverage over nearly all of Europe without building any stations on the continent itself; it was also important for navigating over “the hump” separating India and China. For GIS layers, see www .afterthemap.info. Station locations from J. A. Pierce, A. A. McKenzie, and R. H. Woodward, eds., Loran: Long Range Navigation (New York: McGraw Hill, 1948), appendix B; day and night coverage from US Navy pamphlet, “Loran: Long Range Radio Navigational Aid,” Aug 1945 (ICAO, box “Com— Sub. 1, 2, & 3: 1945– 1949”), 4– 5, with domestic training-chain coverage added; SS Loran coverage derived from sketch maps in a letter from UK Coastal Command, “Gee and Loran Accuracy Charts,” 3 May 1945 (PRO, AVIA 7/2316).

ships could locate themselves (and their targets) on a stable grid of electronic coordinates. Technologically, there were two approaches to creating these electronic grids. The Allies’ systems— Gee, Loran, and Decca— were all based on time measurements, again similar to radar. But instead of measuring the round-trip time delay between a single transmitter and a receiver, the important measurement was the difference in the time delay of two signals sent from two coordinated transmitters. Figure 5.12 shows this idea in the abstract, while figure 5.13 shows how three stations could be combined to create a full coordinate system. Figure 5.14 shows the actual lattice. (Because of the shape of the grid lines, this technique is known generically as “hyperbolic navigation.”)43 In contrast, the German Sonne system was a direct outgrowth of the earlier beam systems. The main difference was that instead of static beams defining fixed paths, the Sonne beams slowly rotated. This meant that a user listening on headphones would periodically hear a station identification tone, then a se224

Chapter Five

Figure 5.12: Hyperbolic navigation. Two transmitter stations— A and B—send out time-synchronized radio signals. Since a user— at X, for example— only receives these signals passively, it cannot measure the absolute distance to either station, but it can determine the difference in the distance between the two. (That is, it does not measure q or p, but rather q minus p.) This establishes its position somewhere along a hyperbola. Although this may seem cumbersome, the great advantage of a passive system is that the same signals can be used by an unlimited number of users and cannot be used for tracking. Note that with only two stations, X and Y cannot be distinguished, since q – p is the same as w – v.

ries of dashes, the constant tone of the beam as it passed, and finally a series of dots. Simply counting the number of dashes and dots could give a remarkably accurate measure of the direction to the station. To create a two-dimensional coordinate system, Sonne stations were simply positioned so that navigators could locate themselves at the intersection of two bearings. These bearings were printed on lattice charts, and the operational result was essentially the same as the hyperbolic systems.44 There were significant differences between these systems— not just differences of accuracy, range, or equipment, but also origin, sponsorship, and strategic goals— but in all cases there was a close alignment between new operational requirements and the metaphor of the grid. Gee, for example, was first invented in the late 1930s but only found sponsorship in June 1940 when British Bomber Command approached the Telecommunications Research Establishment (TRE, the main British radar laboratory) in search of a remedy for their abysmal bombing performance. The designer of Gee, a television and radio engineer named Roger Dippy, later recalled that the requirement was for “a sort of grid reference” that could be used as a common system by all aircraft at once.45 The other systems were prompted by similar needs and understood in similar terms. Loran— which was developed along with American radar at the MIT “Rad Lab” but largely derived from Gee46—was needed to help route convoys through the vast and notoriously cloudy North Atlantic; its primary Inhabiting the Grid

225

Figure 5.13: Three Loran stations— in North Carolina, Nantucket, and Nova Scotia— create two intersecting sets of hyperbolas. In every group of stations, one is designated the “master” (here, Nantucket) and the others are “slaves” that use signals from the master for synchronization. Map from J. A. Pierce, “An Introduction to Loran,” Proceedings of the IRE 34 (May 1946): 219; shading added.

Figure 5.14 (see gallery for color version): A Loran lattice chart of the area around Nantucket. Grid lines are only shown in the ocean, but some limited coverage was also available on (and over) land. More than two million of these Loran charts were published during the war. This detail is shown at actual size; map is Loran Chart, Atlantic Coast: Cape Sable to Cape Hatteras, chart 1000-L (Washington DC: US Coast and Geodetic Survey, 1948).

purpose was likewise “grid-laying.”47 Sonne, in turn, was first deployed for Uboat navigation off the west coast of France, and Decca— another invention of the late 1930s— eventually found the sponsorship of the British navy in mid-1941 for offshore minesweeping (and mapping) in the English channel. In preparing for the D-Day landings, a manager at TRE summed up these navigational requirements with a succinct, comprehensive catchphrase: what was needed was a “gridded battlefield.”48 In other words, what was needed was a way to turn a large and featureless expanse of water, clouds, or darkness into something legible, something coordinated. In contrast both to the domestic metaphor of the railroad and the international metaphor of the ocean, the grid thus suggested a profoundly transnational geography. It was a physical construction, but it was not limited by political borders.49 The epistemic problems addressed by the new coordinate-based systems were thus quite similar to the problems addressed by artillery grids, and they were used in similar ways. Perhaps most obviously, electronic grids were presented as direct replacements for latitude and longitude. During the war, for example, commanding officers in the British Royal Air Force found that the best way to avoid errors between pilot and navigator was to have them “speak the same language”; this meant using Gee coordinates for all communication, without reference to latitude and longitude.50 One of the lead engineers for Loran made a similar suggestion, noting that “there is little that is inherently more desirable in latitude and longitude than there is in the loran co-ordinates themselves.”51 The colored hyperbolas shown on lattice charts likewise came to be seen less as lines drawn on a map and more as full-scale features of the world. With Gee, for example, memos written by air force officers initially described the system as “an aid to map reading”—that is, Gee coordinates were a way of locating oneself on a map. But soon this language was deliberately inverted by central command, and instead lattice charts began to be described as an “aid in the interpretation of Gee readings”—that is, the maps became a way of locating oneself in the Gee grid.52 As with artillery grids, maps were useful but no longer primary. And for some tasks, such as following a single hyperbola all the way to a target, the electronic grid could simply be used on its own. The stubborn tendency of these grids toward geographic stability is best seen in the contrast between Gee and the other three systems. For the most part, the deployment of Sonne, Loran, and Decca was essentially cumulative. Stations were built and then remained in place, and the same coordinates (and maps) were used for the duration of the war. Gee, however, was designed to be reconfigured over time. Figure 5.15, for example, shows a British plan for Gee installations on the European continent after D-Day: it calls for a series of mobile Gee stations which would be advanced in leap-frog fashion as territory was won from the Germans. But this kind of planning was plagued with ongoing problems, since once the Allies finally broke from their foothold in Normandy, their eastward progress was much faster than expected. This led 228

Chapter Five

Figure 5.15: Plan for advancing Gee coverage in France and Germany after D-Day. Each set of open-jaw lines indicates one chain of three synchronized mobile Gee stations; chains would be moved east or north as territory was captured from the Germans. Of all the wartime grid-like systems, only Gee was made mobile; Loran, Decca, and Sonne used permanent installations. From “Advance Proposals for the Use of GEE and GH on the Continent,” Aug 1944 (NARA, RG 331, entry 276F, box 128).

to a bitter rift between tactical forces and central command. In late 1944 and early 1945 a British air marshal in charge of tactical operations wrote a series of letters to headquarters expressing “grave concern” about the rigidity of Gee planning and delays in adapting to new conditions, arguing that there should never have been “any preconceived plan” at all— what was needed was opportunistic siting of a “local nature.” But the charting office responded that greater latitude in moving the antennas would in fact render Gee useless, since the primary cause of operational delay was the need to resurvey the antennas and recalculate all the necessary lattice charts with every change of plans. This tense situation continued throughout the rest of the war, with ongoing complaints from all sides.53 The lesson here is not just about the specifics of Gee; indeed, a similar scheme involving mobile Loran stations was proposed for Southeast Asia but was eventually abandoned for similar reasons.54 Instead, the lesson is that the new coordinate systems were territorial in much the same way as cartographic grids. They were invisible and seemed to have a rather light presence on the ground, but they were nevertheless quite geographically specific. And because they relied on a complex system of paper charts, intensive calculation, accurate surveying, and difficult equipment, their permanence was essential to their success. The political neutrality of the wartime grids resulted from a similarly material set of concerns; it is best seen in the mutual appropriation of Sonne and Gee. Almost immediately after the British discovered the existence of Sonne Inhabiting the Grid

229

in late 1943 by capturing a lattice chart, the archboffin Reginald Jones— who had earlier been responsible for intelligence against the German beams— suggested that instead of destroying the enemy transmitters, it would be better to find their precise location so that the British could issue their own set of charts. The head of navigation for the British Coastal Command agreed, later saying that “what was good for the U-boat prey was invaluable for the hunters”: if Sonne was helping German U-boats hunt Allied ships, then the system— once rebranded as “Consol” to strip it of its Teutonic associations— would help the Allies hunt the U-boats in turn.55 As the war progressed, the British made an ongoing effort to keep Sonne/Consol as a mutually beneficial aid. For instance, when the Germans modified one of their transmitters so that it would no longer give coverage over German territory, it was promptly bombed by the UK— and then subsequently rebuilt in its original configuration. And when a station in Spain began experiencing maintenance problems, it was the British who supplied the spare parts. After the war, Watson-Watt described Consol as a “delightful product of German-British co-operation.”56 The Germans ended up reaching a similar conclusion with Gee. Although they began jamming Gee transmissions over Germany soon after they discovered the system in August 1942, the British subsequently developed antijamming countermeasures and had largely restored the system to full use by the middle of the next year. Beginning in late 1943, however, the Germans reversed their strategy and began taking steps to exploit (and likewise rename) the system for their own use, including manufacturing more than a thousand of their own receivers and building new ground stations in Poland and western Russia.57 Both Britain and Germany continued to deploy countermeasures wherever possible, but by the end of the war such measures were increasingly selective. In early 1945, for example, British pilots found that their Gee sets were being jammed only by directional antennas, and rather inconsistently at that.58 Thus even during the war, the spatial politics of radionavigation were more nuanced than they might first appear. The obvious effect of the war was that radionavigation was reimagined as a way to project geographic legibility into areas beyond the traditional limits of territorial control. German beams, Allied measuring systems, and electronic grids were all means to this invasive goal, and all these systems ignored the clean national/international dichotomy that had structured the distinction between the Radio Range and D/F. At the same time, however, it is difficult to clearly distinguish military from nonmilitary goals. There was no sharp discontinuity, for example, between wartime reconnaissance and postwar mapping: the same systems were used in the same way both before and after 1945. The politics of electronic coordinates were even more ambiguous, since even though they gave a clear advantage to their sponsors, the overall geographic result was to reduce spatial friction for aggressor and defender alike. As early as 1943, Robert Watson-Watt began convening Commonwealth-wide conferences to position Gee and other British systems as civilian technologies for commercial shipping and aviation. The designers of 230

Chapter Five

Loran made similar moves, with the US Navy, for example, issuing a pamphlet just before the fall of Japan that advertised their system to an international audience and made it clear that its coverage would only expand in the years to come.59 By the end of the war, not only were radio mapping and navigation both positioned as nonaggressive technologies, but both had acquired an institutional momentum that transcended any immediate military goals.

POSTWAR CONSOLIDATION, PART I: THE MANY USES OF RADIOSURVEYING By the end of the war, the two new types of radionavigation were thus both poised for wider adoption: the distance-measuring systems developed for blind bombing held great potential in high-precision surveying, and the various electronic grid systems promised to transform civilian navigation. Simply from an engineering perspective, the grid-like navigation systems are the most clearly similar to GPS, especially as purveyors of stable, boundary-crossing wide-area coverage. But in the decades immediately after the war, radiosurveying was no less important as a technology of spatial consolidation; from a wider political and geo-epistemic perspective, radiosurveying can even be seen as the more immediately successful of the two. The new radio-mapping methods were embraced with great enthusiasm for military, civilian, and commercial projects alike, and they played a particularly important role in the American-centric politics of postwar geodesy, the Universal Transverse Mercator grid, and new kinds of territorial claims. In all cases the main appeal of the new equipment was that it could extend high-precision survey across previously unbridgeable gaps— through jungles, into the frozen Arctic, or well out of sight of land— and could therefore connect small-scale or isolated surveys to larger national or transnational networks. This extensive survey work is just as much a part of the prehistory of GPS as is navigation. After all, GPS itself was used for surveying long before it became dominant in navigation, and creating a “worldwide common grid”—or indeed a global positioning system in general— is largely a question of providing a stable framework of precise coordinates. In the late 1940s and 1950s, radio continued to support the kind of aerial photography and hydrographic surveys pursued during the war. Decca, for example, was used for difficult surveying everywhere from the Sahara to Greenland, and surveying quickly became the only use for Oboe and the related H systems.60 The postwar significance of radiosurveying, however, lies more in the advent of two major new strategies, both of which appealed to a wide variety of users and blurred the line between national and transnational space. The first technique was high-accuracy geodetic measurement, and here the American Shoran was unique among the new systems. Shoran continued to be used as a blind-bombing system— especially during the Korean War, where it knocked out dams, rail lines, and other precision targets— but it had a much Inhabiting the Grid

231

longer life as a survey tool.61 It was especially useful for measuring the distance between very widely separated ground stations. As long as both ground stations could see the same airplane at the same time, their spacing could be measured within a few feet. Stations could even be several hundred miles apart, and the total error would still be comparable to the best ground surveys. By measuring the distances between a large network of stations, Shoran could thus be used for high-precision surveys in areas where traditional triangulation would have been either instrumentally or economically impossible. Because this technique relies on distances rather than angles, it is known as “trilateration”; figure 5.16 shows how it works. (The accuracy possible with this technique was indeed impressive: during initial tests in southern Florida, for example, Shoran not only discovered an error of thirty-five feet in the position of Key West, but also an error of 0.005 percent in the accepted value for the speed of light.)62 Shoran was used to map the vast expanse of northern Canada in the late 1940s and 1950s, and its later, even-higher-accuracy cousins Hiran and Shiran became the preferred tool for connecting the previously separate surveying networks of North America and Europe. Figure 5.17 shows the extent of these projects through the 1960s.63 The politics here were a blend of domestic consolidation and US military globalism. Although some countries sponsored trilateration surveys for purely internal reasons, the majority were undertaken with the support of geodesists at the US Army Map Service or its collaborating agencies as part of the larger project of regional— and eventually global— consolidation of surveying networks. The connections across the Mediterranean and the Caribbean, for example, were part of the attempt to extend European and North American coordinates into the Southern Hemisphere and recalculate the size and shape of the earth; the link across the North Atlantic was likewise part of the US Army’s first World Geodetic System in the late 1950s.64 But even purely national surveys— especially in Ethiopia and Iran— could be just as important for this larger project, since they often filled conspicuous holes in the larger transnational network. Individual countries had significant incentive to collaborate, since in exchange for helping with the American plan, their domestic surveying agencies would receive a new set of high-accuracy points that could be used for mapping, stabilizing international boundaries, or facilitating the discovery and exploitation of natural resources. When presenting their work in South America, for example, American survey engineers made explicit reference to oil concessions in northern Guatemala and to the timber, gold, and bauxite in British Guiana.65 The second main technique of the postwar period was the smaller-scale application of radio to offshore exploration. This work was not dominated by any particular country or technology, and it was only partially related to official state mapping projects. Throughout the 1950s and 1960s, private companies in the US, France, and the UK (including again Decca, which was spun off as an independent corporation soon after the war) developed a vari232

Chapter Five

Figure 5.16: The blind-bombing system Shoran reconfigured for high-precision surveying after the war. Rather than using two precisely located ground stations to guide a bomber over enemy territory, instead the distance between two temporary stations is measured using simultaneous transmissions to a plane flying between them. The resulting trilateration network (below) is designed to have the same number of unknowns and equations as a traditional triangulation system. From Carl Aslakson, “The Influence of Electronics on Surveying and Mapping,” Surveying and Mapping 10 (July–Sept 1950): 167.

Figure 5.17: Radio trilateration performed in the 1950s and 1960s, shown with dark shade. Trilateration was at times used for domestic surveying— as in Canada, Brazil, and Ethiopia— but it was mostly used to bridge gaps between existing survey systems and to connect far-flung islands to the mainland. The majority of this work was either performed or supervised by the US. The connections in the North Atlantic and Caribbean were especially important for the recalculation of the shape of the earth for long-range missile guidance. Locating small Pacific islands was also important, since these islands were home to radionavigation transmitters. Map from Defense Mapping Agency, Geodesy for the Layman, 5th ed. (Washington DC: DMA, 1983; map dated 1971), 18; shading added.

ety of specialized systems for hydrographic mapping and oil prospecting on the continental shelf. Although these systems were generally designed using principles similar to the wartime electronic grids, they differed in being portable, affordable, and much more accurate than their predecessors: they could locate (and relocate) points within a few meters, even dozens of miles from shore. The names used for marketing purposes— Raydist, Decca Survey, the Electronic Position Indicator, Sea-Fix, Hi-Fix, Hyper-Fix— also connoted precision measurement rather than navigation.66 Much like the wartime grids, these systems were inherently multifunctional and were used as much for facilitating coordination as for surveying alone. In the late 1950s, for example, a collaboration between Shell and British Petroleum in coastal Nigeria used Decca not just for surveys (both onshore and offshore), but also for guiding tankers, survey aircraft, helicopters, and even Land Rovers. Similar linkages took place elsewhere— especially in the Middle East.67 234

Chapter Five

The overall effect of the various offshore systems was that the oceans ended up being treated as simple extensions of continental territory, in both a rhetorical and a political sense. Rhetorically, these systems were seen as creating fixed, durable landmarks in the otherwise featureless ocean, similar to both electronic grids and their paper-based cousins. This materiality was often quite explicit, as in the surveying advertisements shown in figures 5.18 and 5.19.68 The new systems also allowed much faster collection of reliable depth soundings, and this huge increase of underwater data meant that the seafloor could often be shown on maps with the same detail and visual language used for land. Not only did mariners compare the contours and coloration of their updated charts explicitly with topographic maps, but the 1950s also saw the advent of new scholarly projects that showed underwater features using the kind of hand-drawn relief typically used for mountain ranges. The most famous of these, the shaded-relief map by Marie Tharp and Bruce Heezen that ended up being used to support the theory of plate tectonics, was cosponsored by the US Navy and AT&T. Figure 5.20 shows the popular version published a few years later in Fortune magazine, drawn by none other than Richard Edes Harrison.69 Politically, offshore radio coordinates were closely connected to the aggressive postwar expansion of national territorial claims to the ocean. President Truman was the first to flout the typical three-nautical-mile limit of a country’s territorial waters when he declared sovereignty over the American continental shelf in 1945; a few years later several countries in South America made similarly ambitious claims to fishing rights within two hundred nautical miles of shore.70 These grand claims did not necessarily require radiosurveying, but as they became codified in the UN Convention on the Law of the Sea— first in the late 1950s, then more aggressively in the early 1980s— reliable electronic coordinates became crucial for placing boundaries, policing fishing zones, and partitioning oil and gas discoveries. In the Baltic and the Mediterranean, for example, mismatches between previously land-bound national surveys were seen as a problem as early as the late 1940s. In the North Sea a few decades later, it was estimated that every meter of error in the maritime boundary between the UK and Norway would translate to two million dollars of natural gas.71 The terrestrial logic of precisely surveyed, hard-edged boundaries was thus extended deep into the ocean, provoked by a political-geographic synergy between offshore coordinates and offshore revenue. Taken together, geodetic trilateration and offshore surveying were pursued by very different actors for very different reasons, but all these projects shared a common goal: connecting new surveys to existing networks. Offshore oil surveyors connected the continental shelf to national land-based networks; national governments connected their coordinates to their neighbors’; and the Army Map Service bridged continents. Unlike traditional municipal or national surveys, these projects did not end at jurisdictional boundaries and did not create stand-alone survey systems. The geographic result was not a perfectly homogeneous global space; it was instead a hierarchical space with Inhabiting the Grid

235

Figure 5.18: The featureless oceans become as legible as the intersection of two city streets. This advertisement for offshore Shoran surveying promises instantaneous and continuous accuracy “within a very few feet,” as well as automatic plotting of a ship’s position on a map. From Surveying and Mapping 19 (Mar 1959): 141.

Figure 5.19: The surface of the sea made solid and ready for inscription. More than a decade after the advertisement shown in figure 5.18, the novelty and importance of extending precision surveying beyond the sight of land had not lost its appeal. The promise is again for accuracy of “a few feet.” From Navigation (US) 18 (Fall 1971): inside cover.

Figure 5.20: The ocean floor as a terrestrial landscape of mountain ridges and valleys, drawn by Richard Edes Harrison. These kinds of maps were first made in the late 1950s; they incorporated the huge influx of reliable depth soundings from radiolocated survey ships. Such surveys were sponsored for both military and commercial purposes. Map originally published in Fortune in 1959; this version from Annals of the Association of American Geographers 51 (Sept 1961).

a relatively smooth continuity between the US military’s global project and the smaller geographic reach of other groups’ interests. Both the politics and the geography of postwar radiosurveying, in other words, had much in common with the postwar regionalism of mapping and cartographic grids. But compared to the International Map of the World or UTM, radio ultimately coordinated a much wider range of activities and more fully mingled military, civilian, and commercial goals into a single spatial whole.

POSTWAR CONSOLIDATION, PART II: THE BATTLE FOR UNIVERSAL NAVIGATION Compared to the history of radiosurveying, the postwar development of civilian radionavigation followed a rather messier trajectory. Instead of a few leading technologies developing smoothly from the war, in 1945 there was no single dominant navigation system, and new technologies continued to 238

Chapter Five

proliferate in the decades afterward, with novel systems created for each new task and user group. And instead of being guided by a few major institutions, the postwar politics of navigation ended up being much more contentious, with especially strong clashes not just between the US and the UK, but even between different agencies within each country. By the mid-1960s, however, the geographic result was remarkably similar. The multiplicity of navigation systems and user groups did not lead to spatial fragmentation, but instead to large-scale spatial unification— again largely at a regional scale. The key development that made this possible was new interest, beginning as early as the late 1950s, in promoting translation between different systems. Known as “integrated navigation,” this technique relied on new kinds of receivers to combine signals from multiple sources and present a unified result to the user. This was a both/and solution: each specialized solution could continue to exist, but it also became possible to navigate freely between incompatible systems. The sharp prewar distinction between point-to-point and area technologies was thus significantly diminished, and for many navigators— especially those operating across borders or oceans— the strong experiential metaphors of aerial railroads and oceanic beacons were replaced by the now-familiar experience of being automatically located on a map, centered in a smooth regional space. It is important to emphasize, however, that this result was something quite different both from the expectations of the immediate postwar period and the later universalism of GPS. It came about largely as a commercial solution to political failure, as airlines and navigation companies— especially Decca— ignored diplomatic stalemate and pursued their own solutions. The path from the grids of World War II to the coordinates of GPS was not a purely military one. Immediately after the war, engineers and officials responsible for radionavigation shared a nearly universal goal: to reduce the number of competing systems through an international process of standardization. Dozens of countries were involved in these discussions, which took place everywhere from one-off marine-navigation conferences to the periodic meetings of international organizations, especially NATO, ICAO (the International Civil Aviation Organization), and the ITU (International Telecommunications Union). In theory, the goal was to have a debate about “technical merits” and reach harmonious agreement about the one best system for each major navigational task, with as much overlap between ships and aircraft as possible.72 In practice, however, these negotiations were not driven by discussions of precision, coverage, or functionality, or even by the overarching logic of railroads, oceans, or grids. Instead they mostly consisted of officials from the US and the UK pushing for their own homegrown systems— systems that had cost millions of dollars and that were already installed in thousands of ships and aircraft. US officials, for example, pushed not only for the successor to the Radio Range— the path-laying VHF Omni-Range (VOR)— but also for the expansive grid of Loran. The British advocated not just for the grid coverage of Consol (which Inhabiting the Grid

239

Figure 5.21: A 1947 British scheme for worldwide installations of Decca and Consol. This is a largely symbolic proposal for postwar UK dominance, one that might be made possible by swift international standardization. For a host of reasons— geographic, technological, and political— this plan had essentially no chance of being followed. From International Meeting on Marine Radio Aids to Navigation: Proceedings and Related Documents, April 28– May 9, 1947 (Washington DC: State Department, 1948), facing 438; shading added.

they had completely appropriated from the Germans) and their own Decca system (which outperformed wartime Gee in almost all respects), but also for a Decca-derived path-guidance system known as Dectra (DECca TRAck) that spanned the North Atlantic.73 The level of chauvinism at times bordered on the absurd. Figure 5.21, for example, shows a British plan presented at a conference in 1947 for worldwide installation of Decca and Consol. The conspicuous omission of Loran, which at that time covered a third of the surface of the earth, smacked not just of technological redundancy, but political fantasy. Such chest-thumping was perfectly matched by the US Coast Guard, which presented similar schemes for worldwide Loran expansion.74 This clash mixed political and commercial goals from the beginning, and much of the competition was driven by the lure of international markets for navigation equipment. For example, one of the most vigorous proselytizers of British systems was Robert Watson-Watt, who, only a few months after the war, began pressing his government to offer free equipment to other countries as a way to gain their support and prime the pump for future orders. In 1946 he reported to the Ministry of Transport that his overall goal was nothing less than “making the world fly and sail British”; at stake was “our exports and our prestige.”75 The United States was just as aggressive. It likewise offered 240

Chapter Five

Figure 5.22: Map presented by the US at the 1959 ICAO meeting showing its low-altitude airway network as planned for 1965. The conflict between this point-to-point logic and the area coverage of Decca was a direct continuation of the split between the US Radio Range and European D/F in the 1920s and 1930s. About half of the roughly 1,100 stations shown here were in place by the time of the ICAO meeting. Map from a pamphlet by the US Air Coordinating Committee Technical Division, “Short Distance Radionavigation: Background Information and Views Presented by the United States of America,” 1958 (ICAO, box “SP/COM/ OPS/RAC 1958”), chart 9.

free or discounted equipment for tests in other countries, and for many years after the war it kept Loran operating in the North Atlantic in open defiance of frequency-allocation treaties. In the 1950s it even sold navigation equipment to Communist countries while publicly supporting trade embargoes that would prevent the British from doing the same.76 In his memoirs, WatsonWatt described the Anglo-American struggle as “the cold war of radio aids.”77 Some international standardization did in fact take place, but for the most part agreements were partial at best; if anything, the protracted battles and messy compromises only encouraged greater pluralism. The overall goal was to standardize three classes of equipment: one for local operations near harbors or runways, one for “short-range” journeys over land or near the coast, and one for “long-range” navigation between continents. Only the first of these was settled with relatively little debate in the late 1940s.78 The other two provoked a series of heated and largely unproductive meetings that stretched over most of the next two decades. For example, one of the greatest US-UK confrontations was a 1959 ICAO showdown over short-range equipment that pitted the American VOR against the British Decca. As shown in figures 5.22 and Inhabiting the Grid

241

Figure 5.23: Expansion of Decca coverage in the decades after World War II. By the time of the ICAO standardization meeting in 1959, Decca was installed throughout Europe and eastern Canada and was used for both civilian shipping and aviation. After the US blocked its adoption as an international standard for aviation, it nevertheless continued to expand for helicopter and maritime use. The US military even used Decca to guide its helicopters in Vietnam. For GIS layers, see www.afterthemap.info. Station locations from a database maintained by Jerry Proc, derived mostly from Decca Navigator News (available at http://jproc.ca/hyperbolic /decca_chains.html). For coverage diagrams of Europe, Canada, and the Persian Gulf, see International Hydrographic Bureau, Radio Aids to Maritime Navigation and Hydrography, special publication 39, 2nd ed. (Monaco: IHB, 1965), sec. II.3. For Japan, see Kazuo Taguchi and Kazuo Sao, “Errors of Decca LOP Due to the Metal Structure of a Ship,” IEEE Journal of Oceanic Engineering 7 (Jan 1982): 59. Other coverage bubbles from Jerry Proc. Dectra tracks from Thomas D. Johnson, “Status of Dectra,” Navigation (US) 5 (Summer 1957): 305.

5.23, these two systems were both very well established in their home regions; both were also primed for greater expansion and enjoyed the strong support of their respective governments.79 The official triumph of VOR— which the British press called a “débâcle” of strong-armed American vote packing that might even threaten the legitimacy of ICAO— did indeed lead to international dominance of the American system in civil aviation, even in Europe. But it also pushed the Decca Navigator Company to focus more aggressively on the coastal marine and helicopter markets that it soon came to dominate. (This rift provoked such bitter tension that it was still being remembered almost fifty years later as “the great area-coverage controversy of the 1950s.”)80 In 242

Chapter Five

Figure 5.24: Expansion of Loran coverage after World War II (known as Loran-A to distinguish it from later versions, especially Loran-C). Coverage in North America and the Pacific did not change much after the war, except for new stations near Japan added in the early 1950s. The biggest expansion was in western Europe, sponsored by NATO. For GIS layers, see www.afterthemap.info. Station locations from a database maintained at http://loran-history.info/Loran-A/Loran-A.htm; coverage following International Hydrographic Bureau, Radio Aids to Maritime Navigation and Hydrography, special publication 39, 2nd ed. (Monaco: IHB, 1965), sec. II.2; coordinates for the 1970s Chinese stations (which were independent of any US or NATO plans) from “Loran-A is Alive and Well . . . and Living in China,” Wild Goose Association Radionavigation Journal, 1985– 1986, 47.

long-range navigation, NATO’s official adoption of Loran likewise did little to alter the allegiance of commercial fishers to Consol, and the result was again that both systems continued to expand. By the 1970s, Consol transmitters were installed not just in western Europe, but on both coasts of the US and even in the USSR. Figure 5.24 shows the parallel expansion of Loran— not just into Europe, but also the western Pacific. Together the major systems ended up offering massively duplicated coverage, especially in those areas with the most international traffic.81 This logic was only reinforced by the development of ever more new systems— especially long-range systems— in the 1950s and 1960s. Most of these were developed by private companies and offered only incremental improvements in heavily trafficked areas. By the early 1960s, for example, there Inhabiting the Grid

243

were six separate aids in use by aircraft flying across the North Atlantic, with other, more experimental proposals presented regularly at ICAO. These systems included both track-guidance and grid-like techniques, from both the US and the UK— and eventually from France and Australia as well.82 At the same time, new systems were also developed with much wider geographic reach, including the first global systems. One of these was a proposal from Decca; two were working systems from the US Navy. Loran itself was also upgraded: in the late 1950s the original wartime system was renamed Loran-A and gradually replaced by a more accurate and even longer-range variant called Loran-C that could be used reliably over both water and land. As shown in figure 5.25, it began by duplicating coverage in the North Atlantic and the Pacific; it eventually connected most of the Northern Hemisphere.83 Government officials saw this prodigious geographic overlap and the wide choice of different systems as major problems. In discussions within the US State Department in 1958, for example, American diplomats and military officials felt that the multiplication of navigation systems— especially by the US itself— was a grave threat to the military potential of the new Loran-C. They concluded that “the strongest possible support . . . at the highest possible level” was necessary for getting Loran-C adopted internationally, since only international standardization would ensure availability of the necessary frequency allocations around the world. The State Department thus set about convincing roughly a dozen domestic agencies to “close ranks” behind a unified proposal.84 The next year, civilian officials in the UK met to discuss their response to NATO’s new Loran-C installations in Europe, which they saw as a clear threat to Decca. The British worried that Loran-C would “graduate from the military to the civil field” and began trying to convince their neighbors that it was simply too “wasteful of frequency space” to be adopted.85 In 1962, both the US and the UK circulated strongly worded papers at ICAO calling for a repeat of the earlier clash over short-range systems. Rather than waiting for a dominant long-range system to emerge on its own, their hope was to standardize transnational radionavigation once and for all.86 Airline companies and equipment manufacturers, however, were much less concerned with technological exclusivity. And in contrast to the geographically driven debates at ICAO, the pages of navigation journals and trade magazines were instead filled with enthusiastic reports about the maturation and quick civlianization of a new class of “self-contained” aids that could track an aircraft’s position without using any outside signals at all. Two such techniques were particularly important. The first, known as Doppler navigation, sent radio signals from a plane to the ground below; the second, known as inertial navigation, used high-accuracy gyroscopes. Neither was a new invention at the time— both, in fact, were first used by the Germans for missile guidance during World War  II— but Doppler was only declassified around 1957, and the first commercial inertial system was available for sale a decade later. (Its major debut was on the new Boeing 747.)87 These were a compelling 244

Chapter Five

Figure 5.25: Coverage of Loran-C, the successor to Loran-A. In addition to the US installations in the Pacific, North America, and Europe, this map also shows coverage of the nearly identical Soviet Chayka system, as well as Loran-C stations built by other countries. This multinational network was more extensive and more continuous than that of Loran-A— especially over land, where Loran-A did not work very well— but the basic geographic focus on the northern rim of the Pacific and Atlantic Oceans was similar. For GIS layers, see www .afterthemap.info. Station locations from a database maintained at http://loran-history.info/Loran-C/Loran -C.htm; coverage derived from maps in Edward Durbin, “Current Developments in the Loran C System,” Navigation (US) 9 (Summer 1962): 140; Walter N. Dean, “How We Got to Where We Are” (paper presented at the first Wild Goose Loran-C Association Symposium, 1972), fig. 12; Robert L. Frank, “Current Developments in Loran-C,” Proceedings of the IEEE 71 (Oct 1983): 1130; Per Enge et al., “Terrestrial Radionavigation Technologies,” Navigation (US) 42, no. 1 (1995): 71.

departure from the usual techniques of electronic navigation, and they were especially useful for intercontinental travel, where outside signals were scarce at best. They also did not rely on any ground equipment that might require international agreement.88 Self-contained aids did not threaten to replace the ubiquitous and reliable ground-based systems; instead they were seen as an important supplement that would only be helpful if combined with existing technology. With Doppler and inertial, the benefit of being able to navigate without external signals was offset by the fact that measurement errors were cumulative and accuracy Inhabiting the Grid

245

degraded slowly over time. But this was exactly the opposite of groundreferenced navigation, where accuracy increased over time with the averaging of multiple readings. Combining the two systems could thus give increased reliability over both short and long time scales, especially over the oceans. Civilian equipment for this “integrated navigation” was developed nearly concurrently with the first Doppler sets. In late 1957, Decca Navigator announced a system called DIAN (Decca Integrated Air Navigation) that combined three signals: self-contained Doppler, area-coverage Decca, and path-based Dectra. Decca was likewise the first to commercialize digital aircraft computers; its first model, known as the Omnitrac, was unveiled in 1960. Within only a few years, manufacturers in Europe and the US had released similar equipment that could combine as many as eight outside systems and five self-contained aids, in addition to data from an onboard compass and wind-speed indicator. By 1970, integrated navigation was commonplace.89 The appeal of integrated navigation was thus both technical and political. For navigators, it allowed for more accuracy and reliability. For government officials, it offered a way out of the gridlocked debates over standardization. Politically, the payoff came quickly: in 1965, the promise of combining Doppler, Loran, and Consol led to détente at ICAO, where the US and UK signed a joint resolution stating that there was suddenly “no requirement for a world standard” for long-distance systems. The issue was never discussed again.90 But in addition to this abrupt end to the internationalism inherited from the war, the popularity of integrated navigation also led to several broader geographic effects, both at the experiential level of day-to-day operation and at the macro level of radio coverage and governmental oversight. Compared to the squabbles at ICAO, these more diffuse changes were arguably much more profound. Perhaps most noticeably for pilots and navigators, the main appeal of the compound systems was a radically new navigational experience. Since the computer combined signals from several sources, it made no sense to give output in the units native to each individual system— Decca or Loran coordinates, for example. Instead, output could be given in whatever form was most convenient: distance and bearing to a destination, latitude and longitude, or even UTM.91 But the approach that received the most attention was to show position using a “pictorial computer” that placed the user directly onto a map. Decca was again the leader. Its first map-display system, known as the Flight Log, was unveiled in 1956; when DIAN was released the next year as an add-on to the same interface, a British flight magazine reported that Decca’s map was already “widely accepted as an ideal form.”92 As other companies rushed to catch up, different display types were developed for different user groups. Figures 5.26 and 5.27 show two common varieties: a cockpit-integrated display for use near an airport (especially popular for helicopters), and an automatic version of the traditional “roller map” for navigation along a route. Some roller maps even came with an “in-flight briefing system” of tape-recorded 246

Chapter Five

Figure 5.26: This kind of integrated map display was designed for use near an airport; a “bug” representing the plane would simply be moved over the surface of the map. It was described as “most helpful in overcrowded terminal traffic.” From Eric S. Guttmann, “Chart Logistics for Advanced System Requirements,” Navigation (US) 12 (Winter 1965– 1966): 340.

Figure 5.27: A simple “roller map” display that would automatically advance as the plane made its way along a route. The black wire stretched across the map showed the current location. This model was especially designed for small aircraft; it could even be run using portable batteries in planes without electricity. A separate tape recorder could also be connected that would give voice commands in sync with the map. Similar roller-map cases had been used for decades in cramped cockpits, but they had been turned by hand. From G. Wikkenhauser, “A Roller Map Equipment,” Journal of the Institute of Navigation (UK), 13 (1960): 100.

voice directions that would play as the map advanced. Rear-projection displays were also available that used microfilm maps of very large areas, mostly for military purposes.93 By black-boxing the infrastructure of hyperbolic grids or radio beams, these displays immediately increased the autonomy of the navigator, who could now navigate freely anywhere within the boundaries of the map. But much like the GPS-enabled cell phones of today, these interfaces also put limits on the navigator’s perspective: instead of seeing an entire route at once, map displays only showed one’s immediate surroundings. Again, the subjectivity of the pilot was a regional one.94 From the point of view of engineers and national radionavigation policy, the change was more gradual but pointed in a similar direction. First, hybrid navigation recast the age-old conflict between wide-area and point-to-point systems. With integration, navigational practice was no longer dictated by the physical logic of the ground equipment, and every system could potentially become either a track-guidance or an area system as needed. Although rigid paths would remain dominant for most purposes, by the late 1960s the aviation agencies of both the US and the UK began making new flight rules that allowed aircraft equipped with integrating computers to take more direct routes through the existing VOR network. These rules— known as RNAV, for aRea NAVigation— were first used by Eastern Air Lines’ shuttle service on the congested route between New York and Washington DC, again using equipment from Decca.95 At the time, RNAV was seen as another win for pilot autonomy: it “put the control of the airplane . . . back in the cockpit.”96 But for engineers it was also an explicit rejection of overreliance on any one technological system, and the availability of multiple radionavigation options changed from being a liability to an asset. Especially at high altitudes, integrated navigation was simpler and more reliable than VOR alone.97 The overall result was that the various systems that were formerly in competition began to be seen as parts of a single whole. At ICAO, this was described as a change in the “underlying philosophy” of navigation: instead of focusing on the accuracy, range, and standardization of individual technologies, officials began to focus on the “total navigational capability” of a particular aircraft in a particular location. The US even circulated a memo arguing that the word system itself needed to be redefined to refer to this larger whole rather than to a discrete network of transmitters and receivers.98 These sentiments were echoed throughout Europe and North America, and the new “systems planning approach” was soon taken up in marine circles as well. As explained by an American engineer in 1969, its basic lesson was that “no one piece of equipment is the answer”—the whole was indeed greater than its parts.99 A Norwegian military officer used a similar logic to explain his country’s new interest in “system considerations” when installing both Consol and Decca for commercial fishing, arguing that “there is not any real need to limit the number of systems, as long as each system adds something valuable.”100 What mattered was not simply covering the earth with radio beacons or a smooth 248

Chapter Five

coordinate grid, but the ability of a navigator to complete the task at hand. These tasks varied by location, and the relatively straightforward goal of reaching a destination was usually much less pressing than maintaining adequate spacing, keeping on schedule, and complying with outside directions. This was especially true for air traffic control, but it was also a pressing concern in congested harbors.101 Ultimately, the geography of radionavigation was thus quite similar to the kind of geographic space constructed through radiosurveying. Most attention was focused on areas of obvious concern, especially western Europe, the eastern US, and the connection across the North Atlantic. The link across the Pacific to Japan and Southeast Asia was also quite important. In these highintensity areas, radio systems provided both accuracy and redundancy, and navigators experienced a smooth transnational space that was continuous across land, water, and political boundaries. And these primarily regional systems— along with the local systems of individual harbors and airports— were beginning to be connected to the global systems of the US military that would become increasingly important after 1970. In other words, space was indeed continuous, and it would soon be global, but it was not homogeneous, and a navigator could experience different levels of precision and reliability from several technologies over the course of a single journey. And although American technologies were the backbone of this global space, the US had no monopoly on navigation. Even its flagship system, Loran-C, required cooperative allies to operate antennas, and allies were not always cooperative. Eventually diplomatic agreements would even be signed with the Soviets to coordinate with their nearly identical Chayka system, much as the UK and Germany had done informally during World War II.102 In short, the radio landscape of the Cold War was complex and lumpy— both spatially and politically. It is worth highlighting, however, that in the 1960s this was not necessarily seen as a problem, and the fledgling global systems of the US military were primarily regarded as additional inputs into hybrid computers, not as a potentially universal solution to all navigational tasks. Global technologies did not immediately threaten any well-established regional or local systems, and satellites in particular held no great appeal. In 1967, for example, an engineer at ITT cataloged thirteen different systems then in use and saw no need for any more; as he put it, “Our theme for the next few decades might well be: ‘Better use of existing systems.’”103 When presenting an integrated computer for civilian ships a few years later, an American marine engineer similarly predicted that no existing aids would be rendered obsolete any time soon and mentioned satellites as but one of several new technologies on the horizon. The Norwegians were even more dismissive, arguing quite simply that it would “be a mistake to over-rate the value of satellites for navigation purposes.”104 These statements are not evidence of technological conservatism. Instead, they underscore that the consensus of the 1960s was that global space was best realized through the creation of universal receivers, not a universal transInhabiting the Grid

249

mitter. This was a solid engineering solution, and it satisfied navigators and officials alike. This stable engineering sensibility was potentially stable politically as well. Although radio signals extended well beyond the boundaries of individual states, ground equipment remained stubbornly national and entrenched systems could never be eliminated through international agreement. Integrated navigation might thus be seen as a kind of United Nations of radio aids: it was a way for many autonomous national systems to coexist for mutual advantage. Urban helicopter pilots and naval convoys could each have their own custom solution, and yet together these could be combined to create a relatively seamless global infrastructure. This was not the smooth homogeneous space of GPS, but it was technologically robust, politically legitimate, and geographically expansive.

CONCLUSION: THE SPATIALITY OF RADIO During World War II, describing the new radionavigation systems as electronic grids carried two connotations— one specific, one general. The specific analogy was to the intersecting colored lines on the lattice chart. In many respects the bomber’s lattice was similar to the artillerist’s grid: it was undeniably real without being tangible, and as crucial as the lines on the paper map were, they were always secondary to the full-scale coordinates found in the field. In both cases, the grid was a stable— and thoroughly physical— framework of dependable, semipermanent reference points. In contrast, the general analogy referred more broadly to the overall task of coordination. The grid was a form of order, a language, a way of locating and being located. The “gridded battlefield” was not just an area of the earth overlaid with lines, but an information space where communication flowed smoothly and decisions could be made with full understanding of the changing dynamics of the war. This more general idea of the grid was again important both in the air and on the ground, and oftentimes the bomber and the artillerist could even find themselves inhabiting the very same information space, despite the geometric differences between the lines on their maps. In the years after the war, however, radio increasingly diverged from the literal figure of the grid, and by the 1960s even the most grid-like radio systems had become much less tied to any particular set of coordinates. In surveying, for example, although the offshore measurements of oil companies and national hydrographic agencies were performed using radio grids much like those of Gee or Loran, the results were almost always translated into a more familiar system— often latitude and longitude, but also rectangular coordinates— before any finished maps were made. The raw radio coordinates were only an intermediary step.105 In navigation, integrating computers and map displays likewise hid the physical framework of radio hyperbolas and 250

Chapter Five

presented only the final result. The navigator still inhabited the very same grid as before, but the intersecting lines had disappeared. Similarly, the waning of the controversy between track guidance and area coverage was not a straightforward triumph of grids over routes; instead, the stalemate was simply bypassed as it became possible to fold both kinds of systems into a single black box of reliable geographic location. A domestic transportation guide like VOR, for example, could continue to function as ever for route-based traffic control— and it still does even today, despite being slowly phased out— while still contributing to the broader project of coordination.106 With a system like UTM, the actual lines of the grid are an indispensable part of geographic space. With radio, however, the geometric grid came to matter rather less than the individual geographic point. As the various regional spaces created by individual countries (or companies) from the 1940s to the 1960s were gradually superseded by the global coverage of the US military, this separation between the grid line and the gridded point would only become stronger. Yet despite this shift away from the visible geometry of intersecting lines, cartographic grids and radio lattices still shared a very similar territoriality. As projects of spatial coordination, both were geographically immersive technologies that were wielded in ways that prioritized regional continuity and challenged the sharp borders of territorial states. If anything, the territoriality of radio was rather more aggressively flexible. Radio not only crossed international boundaries, but it could never itself be bordered. Its spatial extent could only be defined by a fuzzy frontier, one that often changed depending on the time of day and the condition of the atmosphere. It was also deployed by a wider range of actors for a more diverse array of uses. The central players were largely the same— military engineers in the US, the UK, Germany, and France— and their attention was mostly directed toward military targeting and domestic administration. But radio was also quite successful commercially, and entrepreneurs— again, both European and American— developed systems specifically designed for civilian aviation, fishing, and prospecting. Even when radio was used in service of national claims, the borders that resulted were of a new variety. Radio ran continuously across both land and water and made oceanic territory a practical reality, but oceanic borders for customs enforcement, fishing treaties, and mineral rights were never the allor-nothing edges of borders on land. This interplay of boundary crossing and boundary fixing would likewise only become more important in the era of increasing American dominance and the global satellites of GPS.

Inhabiting the Grid

251

CH A PT E R SI X

The Politics of Global Coverage: The Navy, NASA, and GPS, 1960– 2010

From an engineering perspective, it is relatively easy to identify what makes GPS such a successful navigation system: it is accurate, passive, and global. Its accuracy is unquestionably impressive. Even a basic receiver can provide coordinates within about ten meters; with extended time-averaging or supplemental enhancement, submillimeter accuracy is possible. GPS is “passive” because it only requires one-way communication between satellites and receivers, meaning that the system can support an unlimited number of users at the same time. And the planetary diagram of GPS orbits— shown in figure 6.1— is such an iconic image of (American) global technology that it could easily stand for globalism itself. Put simply, GPS works. It works for anyone with a clear view of the sky, anywhere on earth, and it works even when its intersection with other kinds of geographic information— unreliable maps, for example— might lead to unhappy surprises. From an economic point of view it is also easy to understand how GPS has become such a ubiquitous, infrastructural service. Since the first tests of the system in the 1970s, GPS receivers have become ever smaller and cheaper, enabling the system to be used in ways entirely unanticipated by its original designers. This miniaturization has been driven by self-reinforcing economies of scale and ever-expanding markets for GPS-enabled products and “location-based services” well outside the specialist world of military navigation. The basic numbers are again incredibly impressive: more than one billion GPS receivers in use around the world by 2010, the latest chips smaller than a fingernail. Here the iconic image might be something like figure 6.2: a GPS collar used for tracking endangered wildlife in southwest Africa. A technology developed to make the battlefield clean, precise, and knowable is modified— within just a decade or two— to bring a similar legibility to the rest of the world.1 253

Figure 6.1: The GPS satellite constellation design as of the mid-1980s: twenty satellites orbiting about twenty thousand kilometers above the earth, completing one orbit roughly every twelve hours. (The exact number and configuration of satellites has varied over the life of the system.) Note that the globe is drawn to highlight North America, making this illustration from a US military report into a rather pat summary of American technological globalism in general. From R. L. Beard, J. Murray, and J. D. White, “GPS Clock Technology and the Navy PTTI Programs at the U.S. Naval Research Laboratory,” in Proceedings of the Eighteenth Annual Precise Time and Time Interval (PTTI) Applications and Planning Meeting (1986), 50 (available from dtic.mil; high-resolution image courtesy Ronald Beard).

Two other aspects of GPS, however, are much less self-evident. One is why the system exists at all. Today, it is tempting to see GPS’s passive, global design as the obvious way— perhaps even the only way— that the gridding of geographic space would reach its fullest expression. But in the 1960s, the reigning engineering sensibility, both within and outside the US military, was that each navigational task and user group deserved its own technological solution with its own geography, and there was little enthusiasm for creating a single navigation system for all purposes. Viewed in this light, the sponsorship of GPS in 1973 by the US Department of Defense should be seen as a somewhat unexpected event. The other, broader difficulty is interpreting the system’s popularity in the decades since its much-publicized debut during the Gulf War. Most evaluations of GPS rarely venture beyond excited superlatives or 254

Chapter Six

Figure 6.2: A GPS collar on a brown hyena— nicknamed “Floggy”—in Namibia in 2006. The miniaturization of GPS receivers transformed the system from a specialist navigation and targeting device not just into a ubiquitous part of everyday life, but also into a new kind of research tool. Starting in 2004, German researchers used GPS to track threatened brown hyenas through the coastal Namib Desert to gauge the impact of diamond mining on their movement and activity patterns. Photo courtesy Ingrid Wiesel, from “Predicting the Influence of Land Development on Brown Hyena (Parahyaena brunnea) Movement and Activity” (poster at the Society of Conservation Biology Conference, Port Elizabeth, South Africa, July 2007).

breathless lists of its many applications. Political judgments, on the other hand, have primarily been of two (equally problematic) varieties: one is the common assumption that GPS is an inescapably military system; the other is the countervailing idea that GPS is a neutral technology with no inherent politics at all. These various reactions span such a wide range— from technophilic enthusiasm to pessimistic critique— that it can easily seem like GPS is simply acting as a metaphor for technology as a whole, with no essential identity of its own.2 Both of these problems— the design and the impact of GPS— are about universalism. How did a universal navigation system come into being, and what has happened as a result? Framed in this way, the virtues of accuracy, globalism, and miniaturization are not themselves explanations— they are what needs to be explained. My analysis in this chapter thus follows directly from my interest in the universalism of the International Map of the World and the UTM grid. The crucial lesson of this comparison is that universalisms are always specific: they construct a specific geographic subjectivity and enable specific strategies of geographic management. The god’s-eye view of the IMW reinforced the clean partition of the world into bounded territorial areas. GPS, much like UTM before it, was instead deployed to enable (and sometimes enforce) coordination across political boundaries, and its greatest impact has The Politics of Global Coverage

255

less to do with the wizardry of precision than the more fundamental shift from the territoriality of bounded areas to the territoriality of disconnected points. The origins of GPS are therefore not simply about militarism; they must also be about geographic coordination more broadly. And although GPS has evolved to become something quite expansive indeed, even the most expansive systems are still not neutral. This final chapter advances two historical arguments in support of a broader interpretive conclusion. The first is simply a friendly reminder that GPS is not the final solution to all geographic problems. As early as 1985, an article in the New York Times Magazine declared that “G.P.S. . . . will change everything” and that “the problem [of navigation] is just about to end.”3 This millennial interpretation is only reinforced by the genuinely multipurpose nature of GPS— the same satellites, after all, provide not only global location to a few meters, but also universal time to a few nanoseconds. What more could anyone want? But GPS is certainly not the first global spatial technology, and it will not be the last. There is ample reminder even just within the realm of radionavigation. Proposals for covering the earth with a single electronic grid were published as early as 1950, and the working systems of the US Navy— one in space, one on land— were available to civilians by the mid-1970s. And in the last few decades other countries have sponsored their own alternatives: the Soviet system, GLONASS, was also developed during the 1970s, while those of the European Union and China— Galileo and Compass, respectively— are both scheduled to be completed in the late 2010s. GPS has been the most popular and prominent of all these global grids, but its dominance is no less historical than the once-universal appeal of the International Map. Second, and more important, the universalism of GPS has never been— and can never be— as totalizing as it may appear. In particular, I want to distinguish between four flavors of universalism. The universalism of GPS was originally administrative: the goal was to create one system for the entire US military, and it was initially resisted by everyone it was meant to benefit. Its geographic universalism, I argue, along with its remarkable accuracy, was only a byproduct of this larger goal, and although global coverage was indeed important to the navy, it was not prioritized at all by the air force— or by civilian agencies. In the decades since its design, its functional and political universalism— its ability to serve all purposes in all countries— have likewise been hotly contested, and as a result the system that exists today is noticeably different from what the military originally wanted. In other words, although the universalism of GPS now appears rather monolithic, with each flavor reinforcing all the others, it did not begin this way, and not all forms of universalism are compatible. Moreover, what separates the GPS of the 1970s from the GPS of the 2010s is not several decades of military pressure, but rather several decades of GPS being wrested from full military control and appropriated for new purposes. GPS was thus unquestionably an important break from earlier radionavigation systems, but it was not the break one might expect. It did not immedi256

Chapter Six

ately solve the problem of geographic knowledge once and for all, and its wide adoption resulted neither from any self-evident technological superiority nor from any sustained military push. The shift represented by GPS is instead best understood as a process of discovery, as different groups, one by one, found that it offered new forms of spatial (and temporal) stability. High precision, global coverage, and the availability of cheap, easy-to-use equipment were all vitally important, but the more fundamental appeal was that GPS could offer an alternate version of space itself, one that was less historical and more purely geometric. Epistemologically, GPS coordinates need not be translated into some other form— a representational map, for example, or a specific coordinate system— in order to be useful, and for many purposes they can entirely supersede the familiar geographies of land, water, and territorial states. Understanding the universalism of GPS thus means understanding the appeal of this profound substitution of one space for another. This chapter is divided into three sections. I begin by looking at the various planetary-scaled navigation systems pursued in the United States in the 1950s and 1960s, in particular the Omega and Transit systems of the US Navy and the handful of proposals for civilian satellite navigation sponsored by NASA. During this time there was a strong inverse relationship between global coverage and functional utility: systems that were designed for coordination (in the expansive sense) were not global, and vice versa. I then turn to GPS itself and ask how it came to be designed to satisfy navy, air force, and civilian requirements alike. This was a precarious game, since none of these constituencies actually supported the program. These first two sections are where I address the different flavors of universalism. The navy, NASA, and the Department of Defense each prioritized a different alternative— geographic, functional, and administrative, respectively— and none sought the kind of broad political buy-in that supported the IMW, or even UTM. Finally, in the last section I analyze the emergence of GPS as a ubiquitous technology: a multipurpose information system widely described as a utility or an infrastructure that, along with the Internet, has come to define late twentieth-century globalism. Analyzing the ways that GPS is actually used— including the divorce of GPS space from physical space, the symmetry of mapmaking and navigation, and the adaptation of GPS to nonmilitary goals— shows how its geo-epistemology implies a specific way of understanding and inhabiting geographic space. Ultimately, GPS does indeed enable certain kinds of interventions and not others—despite its universality, it is not neutral— but its politics are defined less by the militarycivilian divide than by a certain approach to local knowledge.

UNIVERSALISM BEFORE GPS: THE US NAVY AND NASA The idea of a universal radionavigation system— that is, one system that would serve all air and marine needs, anywhere in the world— emerged soon The Politics of Global Coverage

257

after World War II, and throughout the 1950s and 1960s engineers talked liberally of an “ultimate system” that would replace a hodgepodge of conflicting technologies with one global grid. And in the 1950s, many possible solutions were in fact proposed by national agencies and private companies in several countries. Both Decca (in the UK) and ITT (in the US) designed global systems that would use antennas on land, and ideas for satellite systems were published even before the launch of Sputnik. The adoption of a single global system was also one of the main goals of the standardization debates at the International Civil Aviation Organization.4 The wide appeal of this dream, however, must be qualified in two important ways. First, only a very small number of organizations were actually able to put serious money into these ideas: the US Navy, NASA, and similar agencies in the USSR. The US Navy created two operational systems— its terrestrial Omega system was simple and low-accuracy; the satellite-based Transit was complex and precise— while NASA’s efforts, which focused exclusively on satellites, attracted wide notice but nevertheless did not advance beyond experimental trials. (The USSR, in turn, seems to have all but copied the US Navy’s systems.)5 Universalism, in other words, was an international goal but largely an American project. Second, the universal systems pursued by the navy and NASA were not universal in the same way, and their universalism also differed significantly from what would come later with GPS. The navy focused on geography alone, while NASA pursued a new functional synthesis between navigation, communications, and traffic control. Politically, both followed a pragmatic approach and sought international partners only when expedient. Both global coverage and satellites were also entirely compatible with the “integrated navigation” approach of the 1960s; they added something new but did not render other technologies obsolete. So even though the first global systems were indeed the immediate predecessors of GPS, both technologically and institutionally, they did not threaten any broader political or operational attitudes about radionavigation or geographic space. Of the two navy systems, Omega was the first. Much like other early proposals for worldwide coverage (especially from Decca), it was in fact a direct outgrowth of the hyperbolic systems of World War II; the main difference was that it used very-low-frequency radio waves that could span vast distances. Following the usual trade-off between range and accuracy, this meant that a small handful of transmitters could provide coverage nearly everywhere in the world, but the resulting coordinates could only be accurate to about two kilometers, or about the same as traditional celestial techniques. Its main designer— a Harvard physicist-engineer named John Pierce, who had been one of the leaders of Loran during the war— began experimenting with low frequencies as early as 1944,6 but engineering difficulties ended up stretching development over several decades, and a workable proposal was only ready for approval in the early 1960s. Actually building the system took almost another twenty years, and (nearly) global coverage was finally available in 1982.7 258

Chapter Six

Figure 6.3: The “birdcage” of Transit satellites. Each satellite circled the earth at an altitude of roughly one thousand kilometers, completing one orbit every 107 minutes. There was no coordination between separate satellites, and obtaining a position fix only required one satellite to be visible. Coverage was global, but discontinuous. Image from T. A. Stansell Jr., “The Many Faces of Transit,” Navigation (US) 25 (Spring 1978): 56.

Transit, on the other hand, had no wartime precedents, and its development proceeded very quickly. The possible use of satellites for navigation had been widely discussed in the 1950s, but nearly all early pundits predicted that satellites would be useful mostly as stable “radio stars”: artificial, all-weather replacements for traditional heavenly bodies.8 Transit, however, worked instead by tracking the Doppler shift of a low-orbit satellite passing relatively quickly overhead. With prior knowledge of the satellite’s precise orbit, just a single pass was enough to give a remarkably accurate position— to within about two hundred meters in the early 1960s, and better than thirty meters by 1980.9 This technique was a major and unexpected departure— one engineer later described it as “less an invention than a discovery”—that followed immediately from the tracking of Sputnik in 1957, when two young physicists at the Johns Hopkins Applied Physics Laboratory (William Guier and George Weiffenbach) found that it was possible to analyze very subtle interactions between the satellite and the earth’s gravity field.10 This discovery led to two symmetric applications. The first was in geodesy, since analyzing gravity could give very precise information about the earth’s size and shape; the technique quickly attracted the attention of John O’Keefe, designer of the UTM grid.11 The other was in navigation. The navy was actively searching for a positioning system that could be used on its ballistic-missile Polaris submarines, and the global coverage and simple antennas of satellite tracking offered obvious advantages. The first test satellite was launched in April 1960— just nine months after initial funding— and the system was declared fully operational in 1964, with four satellites orbiting around the poles (later increased to six; see figure 6.3).12 The Politics of Global Coverage

259

Figure 6.4: Coverage of Omega lattice charts: gray shapes show areas where maps— and therefore Omegaenabled navigation— were not available. Omega radio coverage was global, but throughout its development the US Navy treated it as a single-purpose system for surface ships only. Map based on information from US Naval Oceanographic Office, “Proposed Omega Navigational Charts,” appendix to “Second Omega Mission Report,” Apr 1971 (PRO, AVIA 65/2100).

Both Omega and Transit were fully global, but both were designed quite narrowly for specific needs, and initially the navy had little interest in accommodating alternate uses, even within the military. Omega was seen as only a lowaccuracy guide for oceangoing ships and, eventually, submarines— basically an all-weather alternative to traditional navigational methods— with virtually no thought given to other possible applications. Pierce admitted in his memoirs that his team “did not really consider” anything but oceanic navigation, even though Omega had obvious implications for aviation, search-and-rescue, coastal navigation, and time synchronization. (Omega relied on some of the earliest workable atomic clocks.) In the mid-1960s, British engineers even saw it as a direct challenge to their own civil-aviation systems. Yet as shown in figure 6.4, the first Omega lattice charts did not even include the interiors of continents.13 Similarly, Transit was very good at providing precise location for Polaris submarines, but it was not useful for most other purposes. With only four satellites in orbit, a fix was possible only about once every ninety minutes, and it took roughly fifteen minutes to take a reading. Reliance on the Doppler effect also meant that the ground speed of the receiver had to be taken into account, which was especially problematic in aircraft. So even though Transit was quite accurate, it barely qualified as a navigation system at all.14 260

Chapter Six

Likewise, although both systems were eventually opened to civilians— Transit in 1967, Omega in 1968— the navy showed little interest in encouraging any new uses or cultivating a larger user base. This was especially true for Transit, which required a powerful (and therefore very large and very expensive) computer for onboard calculations; in the mid-1960s, a Transit set cost roughly $100,000. Primarily for this reason, a review article from 1966 pronounced Transit “highly unsuitable for commercial users.”15 The navy did little to make receivers smaller or cheaper, and miniaturization was pursued almost exclusively by private companies— figures 6.5 and 6.6, for example, show the original military receiver in comparison to the civilian equipment made by Magnavox. In 1980, a typical set still cost about $12,000.16 In the eyes of the navy, Omega was for large oceangoing ships and Transit was for nuclear submarines, and that was that. But despite being treated quite similarly during their early development, there was an important geographic difference between the two systems that ultimately led to remarkably different paths of diplomatic and civilian engagement. Although both Omega and Transit required ground installations, Omega was much more dependent on local conditions, especially outside the United States. Perhaps most noticeably, Omega transmitters were enormous: the preferred installation was a three-kilometer-long antenna strung lengthwise across a valley or fjord, while the alternative was a freestanding tower fifteen hundred feet tall and twelve feet in diameter. As shown in figure 6.7, finding suitable— and politically stable— locations for these antennas turned out to be a rather difficult task.17 Likewise, because the propagation of verylow-frequency signals is highly sensitive to local ground and atmospheric conditions, accurate use of the system required extensive mapping, monitoring, and publication of correction tables around the world. Eventually, raw Omega signals were also supplemented by locally specific “differential” broadcasts from yet another set of monitoring stations; these additional signals were especially important for civilian applications near harbors and airports. (Figures 6.8 and 6.9 show proposals for some of these installations as of the mid-1970s.)18 In contrast, Transit only required extensive groundwork during its initial development, when experimental satellites were tracked by stations around the world in order to construct an accurate gravity model. As shown in figure 6.10, once the system was operational only four stations were required. These installations were crucial for monitoring and updating each satellite’s broadcast data, but they were small and could all be located on US territory.19 These geographic differences had serious political implications; the end result was that Omega became an international civilian system subject to multilateral governance, while Transit remained firmly in the control of the US Navy alone. As a military system in the 1960s, Omega prompted ongoing political headaches with transmitter sites, both abroad and at home. For example, in Norway and New Zealand, Omega faced stiff local opposition from antinuclear activists, especially Quakers, who thought (incorrectly) that it The Politics of Global Coverage

261

Figure 6.5: The US Navy’s AN/UYK-1 computer for calculating position from Transit signals, circa 1961. It was custom built to fit through the hatch of a Polaris submarine and sealed to be waterproof. Described by its manufacturer as “small, inexpensive, and highly adaptable,” it was nevertheless far more expensive and cumbersome than the commercial market would bear. Image from a pamphlet by the manufacturer, RamoWoolridge (a division of TRW), “AN/UYK-1: A Multiple Purpose Digital Computer,” 21 Apr 1961 (available at the US Merchant Marine Academy Library, Kingston, NY).

Figure 6.6: Shrinking Transit equipment from Magnavox, 1968 to 1976. The top image shows two early models; note that the equipment to the left of the woman is the same as the equipment shown in the leftmost column of the bottom image. By 1975 there were about six hundred Transit receivers in civilian use; by 1980 there were ten thousand. Images from Thomas A. Stansell Jr., “Transit, the Navy Navigation Satellite System,” Navigation (US) 18 (Spring 1971): 104; and T. A. Stansell Jr., “The Many Faces of Transit,” Navigation (US) 25 (Spring 1978): 64.

Figure 6.7: Omega transmitter locations, from the first experimental stations of the early 1950s to the final configuration of the late 1970s. Finding suitable sites for Omega’s huge antennas was difficult both geographically and politically, and some antennas had to be located far from where the engineers had hoped. For GIS layers, see www.afterthemap.info. Map based on information in Peter B. Morris et al., Omega Navigation System Course Book, vol. 1 (Alexandria: US Department of Transportation, 1994; available at dtic.mil), pp. 2-11 to 2-21.

was part of the US nuclear program and that Omega transmitters would aid nuclear-missile guidance and become prime targets for the Soviets. The antenna originally installed in Trinidad likewise had to be moved in the wake of the island’s Black Power revolution.20 In addition, the US State Department voiced constant concern about the diplomatic implications of Omega installations on foreign soil (most of the engineers’ site suggestions were rejected out of hand), and final congressional approval in 1968 was contingent on the system being opened not just to civilians but also to international cooperation, both political and financial. The idea was that foreign governments would finance their own Omega stations (roughly $10 million each) in exchange for a seat on a new governing board, thus simultaneously giving the system international legitimacy and lowering its cost.21 This approach was a stark departure from the muscular unilateralism of earlier American systems, and it called for a healthy dose of backdoor diplomacy. In West Africa, for example, the US initially sought the help of the British government when trying to find 264

Chapter Six

Figure 6.8: Proposed network of Omega monitoring stations as of 1978. Because of local variations in signal quality, increasing the accuracy and reliability of Omega required extensive ground installations in dozens of countries; the actual network from the 1980s included roughly forty-five stations. Map from E. R. Vass, “OMEGA Navigation System: Present Status and Plans 1977– 1980,” Navigation (US) 25 (Spring 1978): 44; shading added.

Figure 6.9: A 1975 proposal for “differential” Omega stations in the contiguous United States. Unlike the simple monitoring stations of figure 6.8, which were designed to refine models of Omega accuracy, differential Omega stations would broadcast real-time corrections to any suitably equipped user. (Since the monitoring stations were stationary, the fluctuating local errors in Omega signals could be determined quite easily.) Again, the actual network from the 1980s was not this extensive, but differential services did become available from a number of countries on several continents, especially for coastal navigation. Map from H. G. Miller, “Differential Omega in the Domestic Air Traffic Control Environment,” Navigation (US) 22 (Summer 1975): 68; shading added.

Figure 6.10: Transit ground installations. These stations were critical for ensuring the reliability of Transit coordinates. The temporary, extensive network— dubbed “Tranet”—was used to create a model of the earth’s gravity field that could be used for orbit prediction. The four permanent stations then monitored the satellites in order to compute corrections to these predictions. Since the satellites used these predicted orbits when broadcasting their location, new coefficients had to be periodically uploaded to ensure accuracy. Map based on locations in Thomas A. Stansell Jr., “Transit, the Navy Navigation Satellite System,” Navigation (US) 18 (Spring 1971): 100– 102.

a new home for the ex-Trinidad antenna; the hope was that the UK would take the lead in negotiations with its former colonies and thus eliminate any appearance of American military imperialism. (The UK, however, declined, and the new antenna— the tallest structure in Africa— was eventually built in Liberia.)22 Ultimately, the only way that the navy could build a global terrestrial system was to see it transform into a project shared with six other countries, administered by civilian agencies, and held together by a complex series of individually negotiated treaties.23 This contrasted greatly with how easily the navy kept control of Transit. In 1977 a Magnavox engineer declared that Transit was “suffering from a split personality”: it was by then widely used outside the military for a variety of purposes, but it was still essentially a “closed system . . . made available only as a courtesy,” and civilians were left guessing about the navy’s long-term plans. More civilian money was invested in Transit equipment than in any other system, and it was especially important for offshore oil and gas surveying, both in 266

Chapter Six

the US and abroad. It was even used in the British gas fields of the North Sea, where it superseded high-accuracy Decca in the late 1970s. But despite calls for organizing a civilian user group or establishing an ombuds office within the US government, the navy continued to treat the system as if it had only one purpose: nuclear-submarine navigation.24 In other words, the only feature that unified the navy’s approach to radionavigation— and the only feature that set navy systems apart from previous ones— was the need for global coverage, and global coverage alone did not dictate either a specific technological solution or a particular political approach. At most, globalism simply opened a series of dilemmas: the trade-offs between international legitimacy and military control, ubiquity and cost, and global uniformity and local specificity. And for most users, neither Omega nor Transit were final solutions; they were simply additional inputs for integrated navigation computers. Indeed, despite their global reach, both systems were eventually rendered obsolete by GPS and suffered the same fate as nearly every other system from the same era. Transit was shut off in 1996, and Omega antennas began to be destroyed within months of the last transmission in 1997. The type of universalism pursued by NASA was almost perfectly orthogonal to the navy’s globalism. Rather than worldwide coverage narrowly tailored to specific needs, civilians wanted a regional focus and expansive functional synergy. Specific proposals for satellite systems came from a variety of sources, but NASA was the central hub of this effort from the early 1960s through the late 1970s. In addition to its own research group (the Electronics Research Center, located near MIT), NASA also sponsored dozens of research reports from private companies (including General Electric, RCA, Boeing, and many others) and collaborated with the National Academy of Sciences, the National Research Council, and researchers at MIT, Johns Hopkins, Stanford, and the University of Michigan.25 The centrality of NASA was largely a question of money; many other agencies took an interest in satellites, both in the US and around the world, but only NASA managed to get funding to launch experimental equipment for general-purpose systems. (Just in the US, for example, the Federal Aviation Administration— the FAA— instigated several projects of its own, and Congress also solicited advice from the Departments of Defense, Commerce, the Treasury, and the Interior.26 Studies were also published in other countries— including France, Germany, Japan, the UK, and the USSR— and France in particular launched several satellites with limited capabilities for position finding and tracking. Satellites were also discussed many times at ICAO.27) The approach taken at NASA was very similar to ideas that appeared elsewhere, and nearly all foreign work was also considered by NASA. There was great technological diversity among the various systems that were proposed— one report from 1971 cataloged sixteen different approaches— but interest in multipurpose design and regional coverage unified nearly all civilian work.28 From a functional point of view, the main goal was to combine navigation, traffic surveillance, and communications into one system. This The Politics of Global Coverage

267

Figure 6.11: A civilian satellite system for navigation and traffic coordination. The basic idea was that a satellite would act as a relay point for two-way communication between traffic control stations (shown here in the eastern US and western Europe) and a limited number of ships and aircraft. These signals could automatically measure distance, and therefore location, although navigation onboard would almost always rely on combining inputs from a variety of sources. Image from Leo M. Keane et al., “Global Navigation and Traffic Control Using Satellites,” NASA Technical Report R-342, July 1970 (available from ntrs.nasa.gov), 12.

was largely an outgrowth of the conclusion that the world did not need yet another stand-alone navigation system. What was needed instead was to aggregate and coordinate what already existed, and this meant everything from centralized tracking of aircraft and ships to midflight sharing of information on turbulence and radiation.29 Technologically, this immediately implied that civilian systems would be active rather than passive— that is, civilian systems would use two-way communication between satellites and receivers rather than the one-way communication required by the military. (If military receivers sent out signals of their own, they could easily be tracked by the enemy.)30 In a typical arrangement, such as the one shown in figure 6.11, a central ground station would send a signal to one or more satellites, which would then rebroadcast this signal to ships and aircraft below. The receivers would then automatically respond, and the user’s location could be monitored without any input from the navigator.31 Another approach simply added a communications channel to passive systems like Loran or Decca. Notably, 268

Chapter Six

Figure 6.12: Coverage of a geostationary satellite orbiting thirty-six thousand kilometers above the equator. Almost all civilian proposals for combining communications, navigation, and surveillance called for similar megaregional coverage; this particular proposal was prepared for NASA by Westinghouse. Note as well the four ground stations needed for monitoring. From David J. Sheftel, “Some Considerations in Determining the Role of Satellites in Air Navigation,” Navigation (US) 13 (Summer 1966): 170.

one of the first systems tested by NASA was a rebroadcasting satellite known as OPLE— Omega Platform Location Equipment— that could automatically collect data from Omega-equipped buoys and weather balloons, thus putting Omega to use as part of an active two-way system and subverting exactly the feature most prioritized by the military.32 This fundamental divide between civilian and military requirements also had bureaucratic payoff, since it helped NASA deflect budget competition from the Department of Defense.33 The overwhelming civilian preference for regional rather than global coverage was also given concrete technological form. Nearly all civilian proposals called for satellites in high-altitude geosynchronous orbits that would exactly match the rotation of the earth, which meant that they would provide consistent coverage over slightly less than one full hemisphere, as shown in figure 6.12; a 1970 review article even suggested that it was “generally accepted” that such orbits “must be used.”34 When designing these orbits, most of the focus was on two regions in particular: the contiguous United States and the North Atlantic. Coverage over the ocean was especially enticing, since satellite-based traffic control had the potential to dramatically increase the capacity of transatlantic airways and reduce the costs of commercial shipping.35 This specific regional focus meant that many systems did not prioritize geographic coverage at all. Most offered no service near the poles, but many were also unusable The Politics of Global Coverage

269

Figure 6.13: Near-global civilian coverage provided by duplicating regional systems around the equator. This proposal required signals from three satellites at once; each triplet would be controlled by a separate ground station. The heavy line enclosing most of the Americas shows the coverage of one such triplet. (Satellites 2, 4, and 6 would be part of two triplets at once). From Roy E. Anderson (of GE), “Satellite Navigation and Communication for Merchant Ships,” Navigation (US) 14 (Summer 1967): 130.

near the equator; some even ignored the Southern Hemisphere altogether (for a stark example, see again figure 6.11). For those proposals that did call for wider reach, expanded coverage almost always resulted from simply stringing together several regional systems, as in figure 6.13. This kind of regionalism was not entirely incompatible with military requirements— indeed, a similar geography was favored by the air force— but geosynchronous civilian proposals were often explicitly framed as a solution to the problems encountered with low-altitude Transit orbits. They also happened to be much cheaper: while some multiregional schemes called for as many as eighteen satellites, most systems required no more than one or two.36 The political implications of these systems, like their navy counterparts, largely followed ground-station requirements, but high-minded diplomacy was never at issue. Systems designed only for US coverage would only require domestic support, while systems that would span the Atlantic were instead presented as inherently international. But this internationalism was again largely of the pragmatic, American variety, and reports by NASA (and others) contained at most only fleeting invocations of the kind of political univer270

Chapter Six

salism that animated ICAO standards or the International Map of the World. A 1969 strategy report, for example, argued (somewhat tendentiously) that “any satellite system . . . will, of necessity, be international in character” before going on to explain that this necessity was in fact only financial. In particular, the United States, with its relatively small share of global merchant-marine traffic, would stand to benefit relatively less than other countries from improved shipping management, and Congress would be unlikely to subsidize other countries’ transport industries. And indeed, the only systems that showed any promise of success were those that ended up including provisions for international cost sharing.37 This anemic internationalism followed directly from civilian requirements: the universalism of regional systems for traffic management was only applicable to a rather small set of users, and it was simply not possible to expand into new areas, either geographically or otherwise. Comparing the different approaches of the US Navy and NASA— one passive and global, the other active and regional— the most obvious takeaway is that there was no tight one-to-one match between the new technologies of the space age and any particular political-geographic project. Being global does not require using satellites, and using satellites does not necessarily imply global reach. Thus even though NASA’s systems were never actually deployed, they were still important as a clear and compelling alternative to military navigation, and they remained appealing for civilian purposes even well after the launch of GPS. Yet at the same time, it is also worth pointing out one major similarity between the two agencies: during the 1960s and the 1970s, neither challenged the prevailing interest in combining multiple navigation technologies into a larger whole. Not only did engineers continue to stress the benefits of redundancy, especially in civilian applications, but many of the new systems actually required close integration with other techniques in order to work at all. This was true both for civilians (with airborne uses of Transit and Omega, as well as the NASA projects) and for the military, which used Transit mostly to update submarines’ inertial guidance systems. And although navy and NASA systems certainly had political implications, they did not pose any major challenge to the international agreements reached at ICAO.38 They pushed the boundaries of where and how navigation would be possible, but unlike GPS in the decades to come, the kinds of universalism they prioritized— geographic and functional— were meant to provide more options, not fewer.

ADMINISTRATIVE UNIVERSALISM: THE DESIGN AND LAUNCH OF GPS When looking back at the initial design of GPS in the early 1970s, it is tempting to see a straightforward story of technological ingenuity responding to pressing needs— and to see GPS as a success simply because it was better than what came before. And the technical features of GPS are indeed nothing less than The Politics of Global Coverage

271

dazzling. The basic idea is that each orbiting satellite continually broadcasts a signal giving its location and the time when the signal was sent. Since the signal travels at roughly the speed of light, calculating the precise distance between the satellite and a receiver just requires knowing how long the signal took to reach the earth. But because GPS only uses one-way communication, this is only possible if all GPS clocks, on the satellites and in the receiver, are synchronized within only a few nanoseconds, since a time error of just one millisecond would mean a coordinate error of nearly three hundred kilometers. The solution— technically complex, remarkably expensive, but conceptually quite elegant— was to equip each GPS satellite with an atomic clock (accurate to about three seconds over a million years) and to always have at least four satellites in view from any point on earth. Because the clocks in most receivers are not nearly as accurate as those in space, these four satellites are used to solve for four unknown values: three for distance and one to synchronize receiver time with satellite time. Thus unlike earlier passive systems (Gee, Loran, Omega, etc.), GPS relies on direct, and therefore very precise, measurements of distance. And unlike earlier direct-measurement systems (Oboe, Shoran, etc.), it does so without the need for two-way communication. Precise timekeeping is so central to this accomplishment that GPS as a whole has been summarized quite simply as “clocks in space.”39 These technological feats, however, were only possible as part of a broader story about why GPS was proposed, how its scope grew over time, and how it came to enroll ever more constituencies. In particular, what the technological story misses is just how much opposition GPS faced at the time it was designed, even from users that had the most to gain. Put simply, the existence of GPS has less to do with its technological superiority than with the internal administrative machinery of the US Department of Defense (the DOD), and many of its universalist qualities— geographic, functional, at times political— should not be seen as obvious military features fulfilling obvious military needs, but rather as the by-product of an administrative universalism of a rather more local and fragile variety. For the DOD, the main selling point of GPS was that it promised to eliminate the competition and incompatibilities that resulted from having every branch (and subbranch) of the US military using its own custom-designed solution. Earlier systems had been just as accurate, just as global, and just as suitable for military purposes, but only GPS was able to combine all these features at once and reverse the decades-long pattern of technological fragmentation. The result was the creation of one system, one grid, for all users— even, as it turned out, users that the military had not originally anticipated. The basic timeline of the project is relatively straightforward. DOD administrators were already growing weary of the multiplication of navigation systems by the mid-1960s. In 1968, the Joint Chiefs of Staff issued comprehensive requirements for positioning and navigation, and a satellite coordinating committee was set up in an attempt to facilitate communication between different 272

Chapter Six

Figure 6.14: Launch and operational status of GPS satellites, 1978– 2010. Ten experimental satellites— known as Block I— were launched in the late 1970s and early 1980s; the first fully operational Block II satellites were launched in early 1989. Because GPS satellites do not last forever, new launches are required to maintain the constellation. During the 1990s and early 2000s, the original Block II satellites were replaced by more advanced designs, and the last Block II satellite was decommissioned in 2007. For raw data, see www .afterthemap.info. Data from United States Naval Observatory, “Block I Satellite Information,” 12 Apr 1996, and, “Block II Satellite Information,” 16 Jun 2014 (available as gpsb1.txt and gpsb2.txt from ftp://tycho.usno .navy.mil/pub/gps/).

service branches.40 On their own, however, these efforts were largely ineffectual, and within a few years the idea of funding a shared project emerged as a more forceful solution. GPS was thus approved in late 1973 as the inaugural project of the DOD’s Joint Program Office. It combined two satellite systems already under development: a navy system known as Timation (short for “time navigation”) and an air force system known simply as Project 621B, both of which were begun in the mid-1960s. It was also clear that GPS had the potential to render all other existing systems obsolete as well.41 Design and development proceeded over the course of the 1970s, but funding and support remained precarious into the early 1980s. As shown in figure 6.14, functional deployment finally began in 1989, just before the Gulf War, and the system was declared fully operational in 1995. Perhaps inevitably, the double origin of GPS in both the navy and the air force has led to a rather heated priority dispute, and most existing accounts of the history of GPS have stressed the early competition between Timation and Project 621B. In the last few decades, the leaders of each project have also made strong, almost legalistic claims for the importance of their own contribution, and both were awarded medals— separately— as the “inventor” or The Politics of Global Coverage

273

“father” of GPS. In particular, the leader of the Timation project at the Naval Research Laboratory, the lifelong radio engineer Roger Easton, has argued that “the GPS invention” was using atomic clocks to measure distance, and that this was a navy innovation. (Test satellites for Timation were first launched in 1967; precise time was not as far advanced in the air force.)42 In contrast, the director of Project 621B, Bradford Parkinson— an air force colonel with degrees in engineering and aeronautics, who went on to lead GPS development in the 1970s as well— has instead identified the specifics of the GPS signal structure as the “keystone technology” that allowed multiple satellites to broadcast on the same frequency, thus keeping complexity within reasonable limits. Not surprisingly, this signal was a main focus of research at the Aerospace Corporation, where much of the design of 621B took place.43 On its own this priority dispute is somewhat unproductive, but it does point to two important features of the initial GPS design. First, the intractability of the dispute— and its sometimes less-than-civil tone44—shows that the DOD in fact succeeded quite well in merging multiple competing systems into one unified whole. At the time, this strategy was explicit. The initial proposals for GPS came from the air force, which was chosen as the lead agency to propose a single navigation system in April 1973 and which submitted 621B for approval four months later. The DOD, however, did not accept this first proposal, and Parkinson was told (informally) that the stumbling block was as much bureaucratic as technological: what was needed was not just a good design, but one that would enroll both the air force and the navy alike. In response, Parkinson immediately assembled an alternative that lifted key features from Timation (and from Transit) while still preserving many of the engineering details of 621B, and it was this new proposal that was approved in December.45 Administratively, the ability for both Parkinson and Easton to make robust claims to GPS should thus not be seen as problem at all— it was instead exactly the goal. More important, the priority dispute also shows how this administrative strategy created a positive feedback loop of ever-increasing technological, financial, and political-geographic ambition. The essential dilemma of GPS was that it had the potential to be useful for everyone, but it was required by no one. Just before its initial approval, for example, a DOD official admitted that an ultra-accurate global system would have “hundreds of diverse applications” for the military, but the relevant question was not whether more accurate navigation would be useful— of course it would. The question was instead “whether we want to”: whether building an advanced system, one that could not be justified by existing requirements, would be worth the cost. His own conclusion was that support for GPS would primarily depend on its contribution to a “class of ‘smart’ weapons” that were as yet unbuilt.46 GPS, in other words, needed to be all things to all people and also solve problems that did not yet exist. Parkinson has described this as a “go for broke” approach: GPS was successful only because it was pushed to an extreme, under the continual threat of failing altogether for lack of a clear purpose.47 274

Chapter Six

Even the basic global diagram of GPS only makes sense as part of this logic. From the user’s point of view, both Timation and Project 621B would have seemed almost identical. Both were fully passive, and the most advanced versions of each proposal called for distance measurements using atomic clocks, with four satellites in view at all times. What most distinguished the two systems was their geographic footprint. Timation was an irreducibly global system, and development plans from 1971 called for a constellation of twentyseven satellites circling the globe in eight-hour orbits. Project 621B, in contrast, would have been a fundamentally regional system, with global coverage (except for the poles) provided by four separate clusters of four geosynchronous satellites each, spaced equally around the equator much like some NASA projects.48 Not surprisingly, this difference has led some of Easton’s partisans to conclude that GPS followed directly from Timation, since the constellation in place today is substantially the same as what was included in early navy reports.49 The Timation-like design of the GPS constellation, however, was only viable as part of the go-for-broke approach. On their own, both systems were problematic. The obvious problem with the satellite-intensive Timation system was its price. Indeed, the regionalism of 621B was specifically intended to reduce start-up costs, since launching four satellites is much cheaper than launching twenty-seven.50 But the cheaper, regional approach presented difficult political problems. Like all satellite systems, it would have required ground stations to track each satellite and keep the broadcasts accurate. And like all regional systems, it would be difficult to site these stations on US territory. For the global 621B scheme, at least two would have been outside US control, which Parkinson noted was “not acceptable from a survivability standpoint.” 621B design plans also located one of the regional clusters over the USSR, which could hardly have been interpreted as a peaceful gesture.51 Initial proposals for Timation instead allowed ground stations to be on US territory only, since nongeosynchronous orbits allowed more flexibility on the ground. In other words, before GPS it appeared that a satellite system could either be financially feasible or in US control, but not both. The maximalist approach of GPS overcame this fundamental impasse; it was the most expensive option, but it had a greater chance of success than either Timation or 621B separately.52 The development of GPS after its initial approval continued much of this same trajectory of expanding ambition and a widening (but contested) network of constituents. Within the Department of Defense, the go-for-broke gamble continued through the early 1980s, as GPS leadership sought out opportunities to make the system ever more administratively universal. In the mid-1970s, for example, satellite designs were modified to add equipment for detecting nuclear detonations (in order to enforce test-ban treaties) and also to provide support for the navy’s submarine-launched Trident missiles (thereby preventing a major upgrade to Transit).53 Even so, GPS faced a series of existential budget battles in the late 1970s and early 1980s, and Parkinson has told of The Politics of Global Coverage

275

the need to cultivate a “GPS Mafia” within the DOD that could defend it from cuts.54 The air force in particular grew increasingly skeptical that the system would prove cost effective. The budget was first cut in 1979 by 30 percent (about $500 million), and initial plans for a twenty-four-satellite constellation had to be quickly reduced to an eighteen-satellite configuration— the bare minimum to ensure four satellites in view.55 The air force then proposed to eliminate funding altogether from 1980 to 1982, and the program only survived by the intervention of DOD administrators who remained optimistic about ultraaccurate weapons guidance.56 GPS also faced ongoing competition from the navy, which continued to push Transit upgrades throughout the 1970s and was, off the record, committed to retaining control of a non-GPS system.57 A similar dynamic governed the evolution of GPS for civilian needs, with technological capabilities again determined more by administrative strategy than by navigational requirements or the enthusiasm of potential users. Both Timation and 621B had made allowance for civilian uses— mostly civil aviation and shipping— and the goal with GPS was to allow civilian applications while still being able to prevent abuse by unfriendly forces.58 This initial accommodation of civilian receivers, however, had nothing to do with highminded universalism or military-civilian cooperation; it was instead mostly a bureaucratic and financial strategy pursued by the military alone, and DOD administrators even admitted that neither they nor their nonmilitary counterparts saw any immediate civilian need for high-accuracy navigation.59 The payoff was instead that the promise of civilian use made the initial GPS proposal more appealing to Congress; the military also saw the creation of a civilian market as an easy way to bring down the cost of their own equipment. Civilian money was implicated in a rather more concrete way during the budget battles of the early 1980s, when the DOD proposed an annual usage fee for all civilian receivers— a minimum of $370 a year per user.60 And yet the military made it clear that it intended to retain full control of the system. For example, after discovering that prototype low-cost receivers were far more accurate than expected, it decided that the civilian signal would be intentionally degraded to give accuracy only to about one hundred meters. This policy— known as Selective Availability— was seen by civilians as a most ungenerous gesture.61 The civilian potential of the system was also not especially well advertised, especially early on, and the press gave conflicting reports of how— or even whether— GPS would be made available.62 In return, GPS faced an openly hostile response from civilian agencies whose satellite interests still focused on multipurpose traffic-control systems. In the late 1960s NASA had quickly dismissed Project 621B as “unsuitable” because of its passive design, and when Parkinson invited the FAA to early reviews of GPS progress he received a rather unfriendly reply: “We don’t want GPS, we don’t need GPS, and if it is ever deployed, we will never use it.” 63 Even once development was much further along and reports of coming civilian wonders began to appear in the press, official civilian skepticism remained. In 1978, for 276

Chapter Six

example, the Department of Transportation’s National Plan for Navigation took pains to counter rumors that GPS could fulfill all civilian requirements and concluded that using GPS as “a single universal system . . . does not appear feasible.” Congressional testimony the next year, however, revealed that this recommendation did not actually involve testing the system or even having a full-time employee assigned to evaluate it.64 Yet even after acceptance tests were undertaken in the first years of the 1980s, the FAA continued to voice “significant concerns” with nearly every aspect of the system, including “denial of accuracy, coverage, reliability, integrity, and cost.”65 Regardless of the merit of these conclusions— and they were not entirely without merit— they made it clear that there was still very little overlap between civilian and military goals. Ultimately, an unexpected calamity intervened to upstage military strategy and civilian disinterest alike, and the GPS user base was decisively widened well beyond what the military initially had in mind. In September 1983, the USSR shot down a Korean Airlines flight that had wandered into Soviet airspace, killing all 269 passengers aboard, including a US congressman. In response, President Reagan, along with Illinois senator Charles Percy, quickly offered GPS as a possible way to avoid similar disasters in the future and promised to make it freely available once it was operational. As a result, GPS funding was put on much more stable ground, the military’s user-fee policy was eliminated, and the FAA was instructed to speed up its work adapting GPS to civil requirements.66 This ensured the future of GPS as an open-access, dual-use system for a remarkably wide array of users, but in a way that left none of the main players especially happy. Indeed, even though the military retained final responsibility and national security still took precedence, over the next decade the distinction between military and civilian control was substantially blurred. In 1990, the DOD and the Department of Transportation officially agreed to promote GPS for use in international aviation, and the next year the head of the FAA promised his counterparts at ICAO that GPS would be available as an open system for at least ten years. This promise was renewed in subsequent years, and the military further promised that no changes would be made to the system during peacetime without FAA approval. In 1993, an interagency task force recommended a joint governance structure, and a 1995 study by RAND found “no compelling historical or legal argument for preferring civil or military federal control.” The next year, President Clinton, following a recommendation from the National Research Council, formed a new GPS executive board chaired jointly by the Departments of Defense and Transportation and announced that Selective Availability would be turned off within a decade. (It was actually discontinued in 2000.)67 In other words, the military’s go-for-broke strategy clearly succeeded— all too well. The main lesson here from the first two decades of GPS design and deployment is that calling GPS a military system, while certainly true, is incomplete—in two ways. First, and most obviously, the system’s nonmilitary potential was always an important selling point, and it was a high-profile civilian disaster— one The Politics of Global Coverage

277

directly precipitated by Cold War brinkmanship— that finally ensured the program’s budgetary survival. From the point of view of overall American strategy, it is simply impossible to make any clean separation between military and civilian goals. But second, there is no direct alignment between the features of GPS and any specifically military requirements. Yes, GPS is passive, global, and accurate enough to guide precision missiles, but its immediate goals were administrative: reducing overall navigation costs and stopping the multiplication of systems. These goals would not have arisen outside the US military bureaucracy, but they are not particularly militaristic. Instead, what most drove the design of GPS was an expansive strategy of accommodating a diverse set of users. Designing for global coverage and multipurpose flexibility followed as a result of this basic administrative commitment. GPS, in other words, was a managerial reaction to the 1960s preference for redundant systems and custom solutions. It was a rationalization of radionavigation pushed by administrators rather than users, and it superseded existing systems as much by bureaucratic force as any practical appeal. And even though GPS today is functionally and politically quite universalist— it is used for a wide variety of tasks, all around the world—these features are largely the result of its all-or-nothing agenda migrating from military administration up the American political hierarchy. The universalism of GPS grew slowly over time, and it did not emerge by consensus.

THE USES OF GPS: A NEW SUBJECTIVITY, A NEW TERRITORIALITY Between its initial approval and its launch in the late 1980s, the history of GPS was largely governed by strategic design decisions and bureaucratic strategy, and the major players were all US government agencies. But almost immediately after its debut, the history of GPS instead becomes a history of how it has been used. The major trends are relatively clear: civilian applications quickly outnumbered military uses, and GPS became tightly integrated into other systems of communication and geographic management. By the late 1990s, it was becoming far more ubiquitous than early forecasts had predicted, and new and unanticipated uses began to emerge that challenged the idea that GPS was simply a system for providing geographic location and time synchronization. The mid-1990s were also when GPS began to be described as a new kind of public utility— one available everywhere, free of charge, for any purpose. Policy and strategy still continued to matter, but it was this great expansion of the applications and understanding of GPS that ultimately did the most to reorganize geographic knowledge and territorial space, both at the level of international strategy and at the level of day-to-day experience. Again, the immediate cause of the increasing ubiquity of GPS is easy to identify: during the 1990s, economies of scale and new markets led to rapid miniaturization and falling prices for receiver equipment. In contrast, the sat278

Chapter Six

Figure 6.15: The shrinking size of GPS equipment. In 1973 the designers of GPS had hoped to eventually produce a portable military receiver weighing less than twelve pounds; the Manpack of 1978 (top) weighed more than thirty pounds, while the Defense Advanced GPS Receiver (“DAGR”) of 2004 (bottom) weighed just under one. Images from Steven Lazar, “Modernization and the Move to GPS III,” Crosslink: The Aerospace Corporation Magazine of Advances in Aerospace Technology 3 (Summer 2002): 45; and Rockwell Collins, Inc. (http://www3.rockwellcollins.com/ news/page10585.html).

ellites, ground systems, and overall system design did not change much at all. The satellite constellation was expanded as more money became available— twenty-seven satellites were in orbit as of 1994— and the satellites themselves became more robust, but otherwise, a review of GPS in 1995 pronounced the operational system “virtually identical” to the original proposal.68 Changes in receivers, however, were radical. Figure 6.15, for example, shows the shrinking size of portable military receivers between 1978 and 2004; not only did they become smaller and lighter, but the later equipment also began displaying electronic maps rather than just raw coordinates. Civilian receivers likewise transformed from specialist instruments to mass-market commodities: the cost of an entry-level set fell from $1,000 in 1992 to $100 in 1997. The smallest receiver in the early 2000s was the size of a wristwatch.69 The Politics of Global Coverage

279

These straightforward changes had profound effects. Although some early descriptions of GPS were quite enthusiastic— one article from 1983 predicted (correctly) that GPS “will change our perception of reality, of what is possible, of space and time”70—there is little indication that GPS in the 1970s and 1980s was seen as anything other than an advanced positioning and radionavigation system, and there was relatively little meditation on its geographic or epistemic implications. Even today, most accounts of GPS simply present a variety of unusual applications and a few impressive statistics, without addressing broader changes. And here is where the relationship between the micropolitical and the macropolitical is especially helpful; two effects deserve emphasis in particular. First, at a practical level, GPS participated liberally in one of the headline themes of globalization: simultaneous local intensification and global interconnection. Even the most isolated and idiosyncratic uses of GPS are still, by necessity— and sometimes unintentionally— participating in a project of global coordination. But second, there has also been a more conceptual effect at the level of geo-epistemology. One of the most appealing features of GPS— one stressed by users themselves, although not in these terms— has been that the knowledge it creates is nonrepresentational. In keeping with the metaphor of infrastructure, it is available directly at the site of consumption, and its main appeal is its resolute geometric stability. In short, GPS has created a parallel reality: an intangible knowledge space of electronic points that shares space with the physical world but does not refer to it. In many cases it can even take precedence. The two most prominent early uses of GPS were in cartography and war, and together these applications give a good sense of its initial geographic impact. GPS began to be used for surveying even before it was available for military operations, since surveying did not require a full set of satellites for uninterrupted coverage. The first specifications for civilian survey receivers were issued in 1984; equipment began to be sold soon thereafter, and GPS quickly rendered other radiosurveying systems obsolete. The immediate appeal was that using GPS allowed routine surveys to be completed much more quickly, and costs were reduced by 80 to 90 percent.71 Not only did surveys in remote areas no longer need to begin at existing monuments, but since GPS provided fully three-dimensional coordinates— unlike the two-dimensional coordinates of both traditional surveying and most radio systems— separate methods did not need to be used for mapping locations and elevation, and almost no postsurvey calculations were required. Instead of creating a complex network of angles or distances (as with triangulation or trilateration), and then a separate system for measuring height above sea level, a surveyor using GPS would simply position the receiver and wait, and the resulting coordinates would be accurate to a few centimeters.72 The large-scale geographic result, however, was that the dream of creating a single global coordinate system almost immediately became a practical reality. This was a task that had first begun in earnest in the 1940s when 280

Chapter Six

Figure 1.1: Sheet of the International Map of the World, Hudson River, published in 1927 by the US Geological Survey. Detail of the Albany area is shown at actual size. For high-resolution versions of all images, see www.afterthemap.info.

Figure 1.8: Elevation colors approved in 1909 (left) and 1913 (right). The 1909 scale was criticized for being insufficiently scientific, with the transition from red to purple at high altitudes being seen as incompatible with prevailing ideas about the optics of the human eye. The two scales also used slightly different cutoffs for various elevation levels, especially in the oceans. These scales are from the same sources as figures 1.6 and 1.7, corrected for paper yellowing.

Figure 1.13: A French aeronautical chart from 1911, scale 1:200,000; detail is shown at actual size. Unlike the IMW, this was described as a “picture of the ground.” This chart predates the ICAN standards but uses much of the same logic: railroads are shown as prominent heavy lines, potential obstructions are highlighted, color is used to distinguish forest from open land, and relief is shown using naturalistic shading. In charts from the 1920s or 1930s roads— which were often difficult to see from the air— were usually shown much less prominently, with light gray or pale teal, for example. This map from P. Pollacchi, “La carte aéronautique du service géographique de l’armée,” Annales de Géographie 20, no. 112 (1911): plate 18.

Figure 1.14: An ICAN aeronautical chart from 1934 that used the IMW as a base map. Much new information has been added that was not included on the IMW itself, and the green-to-red color scale has been modified so that green could be used for forests. Red was likewise used to show urban areas where pilots should not expect safe landings. Although many features were redrawn or added for the benefit of pilots— especially railroads, forests, and obstructions— the use of layered elevation colors and the lack of naturalistic hill shading meant that this map was not the “picture of the ground” that pilots and aviation cartographers preferred. The rigid IMW gridiron also did not align with standard air routes, and almost all commercial pilots would have found the edges of this map to be rather inconvenient. Published by the British War Office.

Figure 2.6: Fortune’s “One World, One War”: a polar view showing the centrality of the US to the global conflict, connected by supply lines stretching around the world. The Allies were colored a strong red, which made Africa, Asia, and North America into one large bloc facing attack on two fronts. Map by Richard Edes Harrison, from Fortune, Mar 1942; foldout map originally 21 × 27 inches.

Figure 2.11: Two maps of southern Mexico. At top is a 1938 map by the American Geographical Society, following IMW standards; at bottom is a World Aeronautical Chart from 1946. The scale in both cases is 1:1,000,000, and the main map areas are each about 17 × 25 inches. Both are primarily topographic maps; their differences are in matters of detail, not conception. (The WAC, for example, shows distances and elevations in imperial rather than metric units, does not show ocean depths, and includes information about radio beacons, airways, and magnetic declination. The IMW sheet shows more small towns and gives greater graphic prominence to railroads. They are also cropped slightly differently.)

Figure 2.15: World Land Use Survey maps from central Italy (top) and western Japan (bottom), both reproduced at actual size. The maps use different scales— 1:200,000 and 1:50,000, respectively— and different classification and color schemes. (The Japanese map, for example, uses eight categories for urban land, while the Italian map uses only one.) The Italian map is part of a series covering the entire country; the Japanese map was a prefectural map for “a specific development project.” As with population mapping, comparing land-use patterns between the two areas is tedious at best. From Hans H. Boesch, “The World Land Use Survey,” Internationales Jahrbuch für Kartographie 8 (1968): facing 141.

Figure 3.1: French map of the western front near Amiens, edition of 5 August 1918. Allied trenches (north and west) are in red; German trenches (in the southeast) are blue. Shown at actual size, scale 1:20,000, with lines spaced every kilometer. (Sheet Moreuil, Service Géographique de l’Armée, 1918.)

Figure 3.7: Detail from a German trench map showing a junction between two grids: the Netz von Paris and the Netz von Lille. Notice the discontinuity both in the reference numbers on the left margin and in the grid pattern overlaid on the map (the grids are also rotated roughly 0.7 degrees from each other). It would be impossible for artillery to aim across this break; it also led to general confusion when reporting coordinates. Other German maps, instead of showing a sharp discontinuity, would sometimes provide overlap between adjacent grids and show them in different colors— but this also caused confusion. This detail of sheet Bourlon, edition of 18 Sept 1918, is shown at actual size (scale 1:25,000). Facsimile from Oskar Albrecht, Das Kriegsvermessungswesen während des Weltkrieges 1914– 18 (Munich: Bayerische Akademie der Wissenschaften, 1969).

Figure 4.3: The British and American grids of World War II, as of April 1943. Each colored area is a separate British-designed coordinate system; the idea was to match these systems to the probable theaters of the war as much as possible. The US system covered the rest of the world— the white areas of the map— with a series of narrow north-south strips, the boundaries of which are shown here with purple vertical lines. Map from Army Map Service, Grids and Magnetic Declinations, Memorandum 425, 2nd ed. (Washington DC, 1943); gray shading added.

Figure 4.4: The jigsaw puzzle of British grids in Europe. The “Nord de Guerre” zone was an expansion of the original système Lambert.

Figure 5.2: Lattice chart showing central Great Britain and eastern Ireland. The two numbers from the black box in figure 5.1 correspond to two of the colored hyperbolic lines crisscrossing the map. The intersection of 49.1 (purple) and 4.8 (red) is in the upper right, just north of Leeds. This is a postwar reprint of a British map dated February 1944 (NARA Cartographic Records, RG 77, box 54 of 215, folder “907.2: Raydist, Decca”).

Figure 5.14: A Loran lattice chart of the area around Nantucket. Grid lines are only shown in the ocean, but some limited coverage was also available on (and over) land. More than two million of these Loran charts were published during the war. This detail is shown at actual size; map is Loran Chart, Atlantic Coast: Cape Sable to Cape Hatteras, chart 1000-L (Washington DC: US Coast and Geodetic Survey, 1948).

the US Army Map Service recalculated all European triangulation, and it had continued with high-precision radio surveys in the 1950s and 1960s. Yet this effort had remained largely a project of regionalization, not globalization, and unbridgeable discontinuities still persisted between neighboring survey systems. By the 1980s many of these regional systems spanned entire continents, but there were also many smaller systems, especially in Africa and Southeast Asia, and coordinate consolidation remained a pressing concern.73 Any measurements taken with GPS, however, were automatically integrated into a single global survey system, with no discontinuities at national borders and no gaps between land and ocean. The reason is relatively simple. Like all other satellites (and long-range missiles as well), GPS orbits are constantly interacting with the particularities of the earth’s gravity field, which means that precise knowledge of the earth’s size, shape, and gravity is required when computing coordinates. GPS uses a mathematical model of the earth known as WGS 84, the fourth iteration of the US military’s World Geodetic System. Thus GPS coordinates are not generic global coordinates; they are tied to particular values for the size and shape of the earth, just as was the case with earlier regional systems, and GPS measurements can only use WGS 84.74 GPS was not the first surveying tool that gave inherently global coordinates— that distinction belongs mostly to Transit— but before GPS, global coordinates were only available for specialized purposes. They were used for offshore exploration, isolated islands, and missile targeting, but they would not have been used to lay out a highway.75 GPS thus integrated the day-to-day surveying of property boundaries or engineering work with intercontinental war, without any special effort. By extension, WGS 84 has also become the unofficial standard for most computerized mapping data, especially data collected in the field. Even for very local projects, coordinates are now global by default. This convergence of GPS surveying and electronic mapping also put new emphasis on the cartographic virtues of points. Writing in 1993, the executive director of the Association of American Geographers, Ronald Abler (who would later become president of the IGU), argued that GPS should provoke geographers to “embrace a pointillistic cartography,” since “points are basic and robust,” while lines, areas, and volumes are derivative. As a result, statistical maps might be able to avoid aggregate data subdivided by political boundaries or tabulation areas; instead, maps could show “individual instances of phenomena” in their exact location, regardless of scale. At least in theory, a map of a forest could be replaced by a database of individual trees— both literally and proverbially.76 The potential, in other words, was to bypass most of the traditional infrastructure of mapping— not only cumbersome equipment, but even national surveying and statistics agencies— in order to make representation more intuitive and reliable. In the early twentieth century, this same goal had led geographers to embrace the state as the guarantor of trustworthy knowledge; by the end of the century, the logic was reversed. The Politics of Global Coverage

281

The military use of GPS reveals a similar relationship between local precision, global reach, and the power of points. Although GPS had been used for limited operations in the 1980s, its first major deployment was during the Gulf War in 1991— Parkinson has described the conflict as “almost a boutique war to demonstrate the effectiveness of GPS.”77 It was used in two important ways. The most visible application was in the ever-hoped-for “smart bombs,” especially cruise missiles that could be launched hundreds of miles from their targets. These were not the first precision-guided bombs, but they were cheaper and more weather-safe than their laser-guided predecessors. And while their overall effectiveness is debated— GPS-guided bombs are not as accurate as laser-guided weapons and did not lead to any obvious decrease in civilian deaths per bomb drop— automated high-tech weaponry was almost uniformly embraced as the future of warfare, both in the military and in the popular imagination.78 But no less important than these long-distance operations were new techniques for local engagement. On the ground, GPS gave the US the unprecedented ability to move troops without any need for expert wayfinding or local geographic knowledge. The most famous such maneuver was the decisive four-day advance known as the “Left Hook”: a massive coordinated movement of roughly two hundred thousand American troops across the roadless, sandswept desert of southeastern Iraq and Kuwait that caught the Iraqis completely unprepared and effectively ended the war. Here, what made GPS useful was not its globalism, but its ability to replace a local system of (nonexistent) physical landmarks with a new local system of electronic coordinates.79 This new reliance on GPS has been widely interpreted as evidence of a major historical shift in military grand strategy. The most common response has been to see GPS as the central technology of an expansive “precision revolution” that could replace large-scale offensives with small-scale, largely automated interventions.80 But more profoundly, GPS has also led to a shifting sense of the importance of national territory. The political scientist Barry Posen, for example, has located GPS as part of an emerging “command of the commons” strategy for the American monopolization of sea, air, and outer space, everywhere in the world. By maintaining functional supremacy in areas beyond any formal territorial borders, this strategy all but negates centuriesold assumptions about the separation between a controlled domestic homeland and unclaimable international space.81 A similar but more radical analysis was offered in 1992 by the Russian military scientist Vladimir Slipchenko, who concluded that advanced-technology weapons would replace wars of “front lines and flanks” with a battlefield divided simply “into targets and non-targets”—an open field of points. As a result, “there will be no need to occupy enemy territory.”82 The implication here is not just that enemy territory is now porous and fragmented, but that it is in fact always already occupied, at least in the sense relevant to conventional military strategy. So it is not just that GPS gives the United States and its allies the upper hand in battles for ter282

Chapter Six

ritorial control; rather, international boundaries themselves lose much of the physical importance they once had. The question is not whether boundaries can be crossed, but whether they should be. The civilian adoption of GPS in the 1990s followed quickly— and quite directly— from the war in Iraq. On television and in newspapers, the war acted as a powerful advertisement for precision navigation; thousands of troops also gained firsthand experience with portable receivers. But the war had more material effects as well. Because war had broken out before GPS was fully operational, the military found itself woefully short of receivers, and emergency orders had to be placed with commercial firms for more than ten thousand handheld sets. As a result, almost 90 percent of all receivers used in Iraq were commercial units; some troops even asked their families at home to send receivers in the mail. In the span of just a few months, the capacity of civilian firms was thus greatly expanded, and the price of civilian equipment substantially reduced.83 After the war, the trajectory of civilian adoption was similar in many domains, and equipment price was indeed determinant. Most of the best-known uses of GPS had been under development since the early 1980s, but they only saw mass-market adoption in the mid-1990s before finally becoming commonplace in the early 2000s. This pattern is perhaps most obvious in the case of transportation guidance. For example, the first GPS-enabled car navigation systems— complete with voice prompts— began to be developed in the mid1980s, and the first working systems were available around 1990, with wider installation in production cars finally starting in 1995.84 For aircraft, both the FAA and ICAO began investigating GPS in earnest in the mid-1980s as well; the FAA issued its first standards in 1992, and ICAO followed in 2001.85 Other transportation applications followed a similar path: the use of GPS for tracking railway trains was still quite experimental in the late 1980s, but by the early 2000s specialized installations were available for everything from close control of farm equipment in “precision agriculture” to low-visibility snow plowing.86 This same trend toward diversification and ubiquity is also evident in stationary applications, especially in science and engineering. GPS was used for the direct measurement of tectonic-plate drift as early as 1984, and by the late 1980s GPS was used for time synchronization between farflung laboratories and observatories. But fifteen years later, it was possible to use GPS for a remarkably wide variety of data-collection projects— including lightning surveys, wildlife tracking, and atmospheric monitoring— and GPSsynchronized time was integrated into cell-phone networks, power grids, Internet routing equipment, ATMs, and even municipal traffic signals.87 This proliferation of uses led to an important change in the way that GPS was understood. Formally, GPS is described as a PNT system— positioning, navigation, and timing. But many of the most visible applications of the late 1990s and early 2000s— especially tracking and monitoring— did not fit easily within this rubric, and the more that GPS was hybridized with other techThe Politics of Global Coverage

283

nologies, the more its potential uses came to seem increasingly difficult to predict. In popular discourse, “GPS” also began to refer not just to a system of satellites, but to a wide variety of allied technologies, including electronic maps, tracking devices, and eventually other positioning systems as well— everything from rival radionavigation systems to the location information collected by wireless phone or data networks.88 It was also at this time that GPS, either alone or in combination, started to be described as a new form of infrastructure. These new descriptions of GPS as a public utility or infrastructure contained a provocative ambiguity. When Parkinson described GPS as a “navigation utility” in 1994, the implication was relatively clear: a particular service— navigation— was becoming invisible and taken for granted. (Parkinson has long stressed the system’s clandestine qualities. In 1980 it was a “quiet revolution”; by 2010 it had become a “stealth utility.”)89 But most other infrastructural analogies have been rather more vague. For example, when GPS was described in 2003 as simply the “fifth utility” alongside water, gas, electricity, and communication, it was not terribly clear what service it would deliver. Or when it was described in the same year as the “first global utility,” it was almost as if GPS would deliver globalism itself.90 One of the most common comparisons has been between GPS and the Internet. Not only were both systems developed by the military and subsequently turned into open platforms, but, as a legal scholar put it in 1995, “GPS is like the Internet in that many of its future uses have not yet been conceived.”91 At times the analogy is even more radical, with GPS not simply being compared to the Internet, but actually forming part of the same system. In 1994, the architect and designer Laura Kurgan described GPS as a “cyberspace . . . a scaleless information zone” that blurred the distinction between physical space and social or informational spaces. Two years later, President Clinton referred to GPS as “an integral component of the emerging Global Information Infrastructure”—a similarly all-encompassing totality. Similar claims have been made ever since: GPS is not just one utility among many, but one component of the— singular— global infrastructure.92 These analogies reveal a great deal about the experience and epistemology of GPS. In a limited sense, describing GPS as a utility is simply a way of saying that it is ubiquitous and black boxed, much like other public utilities. While certainly important, such a claim could also be made for other forms of geographic knowledge— paper maps, for example, are public utilities in a similar way. The more radical analogies, however, go further, suggesting that GPS is somehow synonymous with geographic information in general, or even geographic space itself. This is a profound claim that presents a serious challenge both to the everyday idea of mapping and to our familiar geographic subjectivity. But experientially, it makes sense. GPS coordinates do not refer to the world the way a map does— they do not stand at arm’s length and describe. Instead, they create their own world. (Or, to take the popular analogy one step further, if the Internet is a cyberspace without any geographic presence at all, 284

Chapter Six

GPS is a cyberspace that perfectly overlaps and even competes with ordinary space-time.) And indeed, navigating by GPS often means ignoring the physical landscape altogether and placing trust only in the coordinates themselves. Two seemingly disconnected examples make this clear. In reports of its adoption for new uses, GPS is usually not presented as simply a more precise version of earlier techniques, as if it were only a more accurate kind of map. Instead, the emphasis is on reliability, permanence, and the ability to use it as a replacement for other forms of local knowledge. In 1994, an interdisciplinary team conducting research on coral reefs in the Dominican Republic wrote glowingly of GPS as a “Rosetta Stone” that could integrate anthropological, remote-sensing, and marine-ecological data. An anthropologist could accompany a fishing boat to traditional fishing sites far from land, and a Scuba team could return later to take a fish inventory.93 A few years later, research on Inuit use of GPS in northern Canada found, not surprisingly, that it tended to promote the use of straight-line paths, since hunters would simply set a digital waypoint before heading out on the ice and then later use the same GPS unit to return directly to their starting point. This was quite different from the more circuitous routes that hunters would follow when navigating by wind, ice patterns, and other environmental cues.94 In both cases, what made GPS appealing was not centimeter-level accuracy, but the ability to create a durable geographic marker that could remain the same across time, space, and culture. No maps were required, no landmarks, no prior knowledge at all. Despite obvious differences in equipment, purpose, and time period, the geo-epistemology of these later uses is thus quite similar to earlier applications like surveying and missile guidance: in all cases, geographic knowledge consists of placing, reading, and connecting points within a virtual electronic grid. For example, note how much GPS mapping has in common with GPS wayfinding, where both the mapmaker and the hunter are using (roughly) the same equipment to record coordinates. As a 1995 study of GPS put it, “Applications by surveyors and hikers are mirrors of each other”—the surveyor is essentially navigating a route and recording the result, while the hiker is making a map that may never be saved.95 The GPS smart bomb and the GPSguided snowplow are doing something similar as well. Even though the points they connect have already been indexed through target reconnaissance or turn-by-turn map software, the virtual landscape still takes precedence over the physical. Or consider something like an international boundary. For centuries it was a formidable task to make a boundary a real and stable part of the physical world. But as chains of electronic coordinates, boundaries can be just as stable in the open ocean as along a mountain range; they can also be multiple, shifting, and transient. The knowledge space of the electronic grid is self-contained and self-sufficient. Compare this kind of knowledge to the knowledge of representational maps. With representation, mapping must come prior to navigation, and the truth of the map is always at issue, in part because there is such a clear The Politics of Global Coverage

285

distinction between the complex reality of the world and the abstracted information presented on the map. The map and the territory are in tension, and knowledge is always contingent on access. But with GPS there are none of these divisions: anyone with the right equipment can access the same coordinates as anyone else, anywhere in the world. GPS thus challenges both the temporality and the governmentality of traditional surveying. Making GPS coordinates useful does not require collecting geographic knowledge at a central archive, integrating it with prior knowledge, or maintaining an active presence on the ground. GPS coordinates are perfectly real— they are not imaginary abstractions— but they have no history and ignore all boundaries. They also cannot be true or false. The recorded locations of buildings and roads can be correct or incorrect, but the coordinates themselves have no truth value— they simply exist. To be sure, GPS has not eliminated the need for representational mapping, and the two can (and do) coexist quite happily. Tasks like in-car navigation, after all, are completely dependent on maps, and the use of GPS for surveying has only led to the acceleration and proliferation of cartographic information. But as forms of knowledge, the two are quite different. Maps are entwined with the world; GPS stands on its own. Following Lewis Carroll or Jorge Luis Borges, one could easily conclude that with GPS the map and the territory have merged: the world becomes a map of itself at a scale of 1:1. But perhaps it would be more appropriate to say that with GPS, traditional paper (or digital) maps take on a role much like the lattice charts used for bombing during World War II. Electronic coordinates are not a way of locating oneself on a map; maps are instead “interpretation systems” for navigating a field of coordinates. Maps are helpful— at times indispensable— but hold no particular epistemological priority. What matters is the coordinates, local and global at the same time, available in situ and nowhere else.

CONCLUSION: NEUTRALITY AND INTERVENTION The transformation of GPS from a specialized system into a ubiquitous part of everyday life has provoked two parallel developments. One is the popularity of uses that had been altogether unanticipated in the 1980s and early 1990s. Foremost among these is the use of GPS for tracking (of wildlife, criminals, children, or cargo) and for amateur mapping by artists and activists.96 The other is an upswell of both popular and scholarly concern with the system’s US-military provenance, often focusing on issues of civil liberties, privacy, and the democratization of cartography. This concern has taken a variety of forms, but on the whole the debate has focused less on what GPS can and cannot do and more on differing assumptions about the politics of technology, in particular the degree to which technologies are forever marked by their origins. Both of these developments raise similar questions about the politics of GPS 286

Chapter Six

at the beginning of the twenty-first century. Do its unexpected new uses confirm or challenge its military personality? Is GPS inevitably an instrument of war, or is it instead a multipurpose tool of civil society? Yet as pressing as these questions may be, they are somewhat badly formed, since they posit GPS as a somewhat singular (and static) technological device rather than as part of a continuum of technological change. And if anything is certain, it is that there can be no definitive statement about radionavigation that is true for past, present, and future alike. But this does not mean that nothing can be said at all. GPS most definitely does have a politics. Understanding this politics, however, requires seeing the military as only one part of a broader story about new ways of creating and using geographic knowledge. In the last fifteen years, there have been two common responses to GPS. The first is that it is an inherently military technology and that its widespread use represents the militarization of civil society. The strongest versions of this argument claim that GPS— especially when combined with computerized mapping— has created a cultural obsession with precision so pervasive that techniques of military targeting end up blending seamlessly into practices like targeted marketing. Not only has GPS turned American consumers into “militarized subjects,” but the integration of GPS into everything from cell phones to traditional hunting practices will “deliver American militarized realities” abroad as well.97 A less forceful version of this interpretation has also driven much of the debate about competition between GPS and the European Union’s planned civilian (and partly commercial) Galileo system. Many observers, from American journalists to foreign heads of state, have been inherently distrustful that a system maintained by the US military would remain reliably accessible, despite all assurances from the American government. In 2001, for example, French president Jacques Chirac suggested that without a satellite system of their own, European countries could become “vassals” of the United States.98 The second interpretation— often explicitly opposed to the first— instead posits GPS, and technology in general, as an inherently neutral tool that can be used either for good or for evil, regardless of its origins. Optimistic scholars tend to emphasize the usefulness of GPS for things like tracking endangered species, clearing land mines, or the rapid mapping of Haiti after the 2010 Port-au-Prince earthquake (which was indeed impressive— see figure 6.16). Optimists also stress that although GPS can be used for top-down surveillance by police or employers, it can also be used for bottom-up “sousveillance” to hold governments accountable, such as when marginalized citizens use GPS for reporting broken street lights in New Jersey or for mapping informal settlements in Kenya. Even advanced missile guidance has its good side, since surgical strikes on infrastructure— like the one in figure 6.17— could mitigate the senseless killing of area bombing.99 Belief in technological neutrality is also embraced by certain pessimists as well. For example, the geographers Jerome Dobson and Peter Fisher have issued strong warnings about the coming The Politics of Global Coverage

287

Figure 6.16: Rapid humanitarian response using GPS: coverage of Port-au-Prince, Haiti, by the collaborative OpenStreetMap project before (top) and two days after (bottom) the 2010 earthquake. Maps modified from http:// www.flickr.com/photos/mikel _maron/tags/quake2010.

mass-surveillance society and the potential for a new “geoslavery” enabled by coercive GPS tracking. For them, the worry is not the military, or even GPS itself, but its exploitation by unscrupulous corporations and individuals; arguing that technology is neutral is important rhetorically for defending GPS against these abuses.100 Both of these views collapse past, present, and future into a single political essentialism: either technology is entirely determined by its origins, or it is entirely free from them. Teasing apart these three temporalities, however, allows a rather more specific and nuanced interpretation of GPS. First, looking historically, it is clear that the assumption that military-sponsored technology can only further militarist goals is simply not true.101 Take the example of GPS tracking. Today this is seen as one of the main threats of ubiquitous GPS, and it is seen as evidence both of militarism and of technological neutrality. But tracking is neither military nor neutral. From the 1960s to the 1980s, it was civilian agencies like NASA and the FAA that wanted to make surveillance an integral part of satellite navigation, not the military. In fact the explicit lack 288

Chapter Six

Figure 6.17: An airfield in Afghanistan is surgically disabled: white arrows show the craters left by GPS-guided bombs dropped by an American B-2 after flying nonstop from an air force base in Missouri. US Department of Defense Bomb Damage Assessment photo, released 8 Oct 2001 (available at http://www.defense.gov/news/ briefingslide.aspx?briefingslideid=184).

of surveillance in GPS’s passive design— its lack of technological neutrality, in other words— is what made the system suitable for military needs; it is also what made it possible for the system to be open to anonymous, nonmilitary users. The main purveyors of GPS tracking in the early 2000s have instead mostly been law enforcement agencies and private corporations who couple GPS with consumer technologies like cell phones. Similar conclusions can be drawn from other parts of GPS as well; even something as simple as the low cost of receivers shows the importance of nonmilitary influence.102 Zooming out, however, there is also a broader question here about what counts as an origin in the first place. Should we trace the origins of GPS only back to the Department of Defense in the early 1970s? Or should we start instead with the global coverage of the 1960s, or even the first grid-like systems of World War II? What about the civilian systems of the 1920s, or the explicit similarities with nonradio technologies like the UTM grid? Seen in this light, the overall development of coordinate-based navigation and surveying can The Politics of Global Coverage

289

only make sense as a story about ongoing fluidity between military intelligence and civilian interest. If GPS is neither inherently militaristic nor inherently neutral, what is it? Understanding the present-day politics of GPS is largely about analyzing the ways that it is used. And from a practical point of view, the key feature of GPS is that it replaces lumpy, historical, human space with a globally uniform mathematical system. By extension, the central political fact about GPS is that it substitutes a locally available grid of geographic coordinates for other kinds of local knowledge and encourages intervention without long-term commitment. This intervention can be initiated from afar— precision bombing, humanitarian relief, GPS tracking— or it can be projected outward, as with activist mapping. In all cases, however, the goal is to encourage action and to bridge the political divide between center and periphery, in Haiti as much as in Afghanistan. The relevant political distinction is therefore not between military and civilian, state and nonstate, or even good and bad, but between local and nonlocal decision making. In other words, the politics of GPS in the last twenty years are not aligned with particular institutions, but with a particular approach to governance— one that is spatially intensive but fundamentally temporary. Finally, looking to the future, it is clear that the politics of GPS are changing. Since the mid-1990s, the evolution of GPS has been increasingly dominated by civilian and internationalist concerns, and the result has been that GPS is now becoming universalist in ways that were never part of the original military design. Perhaps the most obvious changes have been modifications to the GPS satellites themselves. As part of a modernization campaign first announced in 1998, satellites scheduled for launch in the late 2010s will include not just an upgraded military signal, but two new civilian signals as well, one of which is designed specifically for civil aviation.103 Even more important is the ongoing civilian development of local and regional augmentation systems for increasing accuracy and reliability; figure 6.18 shows the current state of these systems. The technology here is essentially the same as with differential Omega, and the first systems for GPS began to be developed immediately after Selective Availability was announced. (One of the leaders of this effort, ironically enough, was Brad Parkinson, who left the military in 1978 and worked on civilian applications at Stanford throughout the 1980s.)104 The success of these systems in the 1990s effectively rendered Selective Availability obsolete even before it was officially discontinued in 2000, and because they support life-critical applications like harbor and air navigation they have now severely reduced the military’s ability to disable the civilian signal in wartime. These systems are also changing the basic spatial logic of GPS, since they are regionalizing the otherwise homogeneous global grid and creating new hierarchies of access and accuracy. They have likewise enabled some of the most Orwellian GPS applications, including tracking in places where GPS signals themselves cannot penetrate— such as inside hospitals or office buildings.105 290

Chapter Six

Figure 6.18: The increasing regionalization of GPS coverage. In response to the military’s intentional degradation of civilian GPS signals, various government agencies and companies (especially the US Coast Guard, Federal Aviation Administration, and John Deere) began providing Differential GPS (DGPS) and SatelliteBased Augmentation System (SBAS) services in the 1990s; these systems increase accuracy by monitoring raw GPS signals and broadcasting real-time corrections. The satellite-based systems use geosynchronous orbits. For GIS layers, see www.afterthemap.info. DGPS stations from a database maintained by Alan Gale (version 10.3, 11 Jan 2015, http://www.ndblist.info/datamodes.htm).

A similar story can be told for the migration of GPS away from full US control. The governance of GPS has been internationalized from two directions at once. First, in order to prevent the construction of competitor systems, the US has put considerable effort into encouraging the use of GPS by other countries, often quite formally. In addition to the promises made at ICAO since the early 1990s, since 2005 even the ground-station network has been expanded to include civilian stations outside US territory, as shown in figure 6.19.106 At the same time, European and Chinese plans for their own systems— and the Russian rehabilitation of GLONASS, which fell into neglect for many years after the breakup of the USSR— also promise to significantly de-Americanize global coordinates. Although the creation of redundant systems might seem to foreshadow impending incompatibilities and fragmentation, these new systems are usually seen as a stepping-stone to a single, irreducibly internaThe Politics of Global Coverage

291

Figure 6.19: The gradual internationalization of GPS ground stations. In 1973, the original GPS proposal called for stations on US territory only. Two of the stations eventually built in the early 1980s were sited on US military bases on British territory (Ascension and Diego Garcia). The expansion of the ground network after 2005 included stations in several additional countries. Map based on information in C. H. Yinger et al., “GPS Accuracy versus Number of NIMA Stations,” in ION GPS/GNSS 2003: Proceedings of the 16th International Technical Meeting of the Satellite Division of the Institute of Navigation, September 9– 12, 2003, Portland, Oregon (Fairfax, VA: Institute of Navigation, 2003).

tional system. For example, many kinds of user equipment— even the lowly cell phone— are already being designed to receive signals from all systems at once. Many GPS specialists, from ICAO officials to international legal scholars, likewise prefer to distinguish the current configuration of GPS from what they call the Global Navigation Satellite System (GNSS)— the singular megasystem made up of all national systems together.107 All these efforts are essentially attempts to retain the functionality of GPS while freeing it from American control: that is, to separate the politics of GPS even more fully from the influence of any particular institution. These changes may well push GPS toward greater functional and political universalism; this certainly seems to be the goal. But even if this is not in fact the result, it is nevertheless true that GPS is becoming something different from the homogeneous, passive system designed in the 1970s. As a result, both its spatial and political logic will also change. Many of these changes are dif292

Chapter Six

ficult to predict. Military and civilian applications may become increasingly blurred, or increasingly distinct. It may construct new forms of transnational territory, or it may expand the scope of national space— or, as has been true thus far, it may do both at once. But one thing seems clear: GPS and its various competitors are in no danger of being shut down. Millions of people use GPS every day, and countless auxiliary systems— both military and civilian— rely on its uninterrupted operation.108 The choice is not between keeping or discarding GPS as a monolithic whole, but between different future versions of a rather heterogeneous blend of military, civilian, and commercial systems. The ongoing evolution of GPS is also a strong reminder that although the shift from representational knowledge to full-scale grid systems was an important change of political-geographic technology, it was more than that as well. It was also a change in how political-geographic practices are embedded within geographic space in the first place. With the representational mapping of the early twentieth century, the concern was to establish geographic truth. The idea was that there is a single geographic reality, and total knowledge of that reality is not only possible but self-evidently useful for decision making and administration. The politics of representational mapping thus depend on how truth is created, certified, and naturalized. The abandonment of this ideal— that is, the shift from representation to raw presentation— was tantamount to a realization that geographic space could instead be manipulated quite directly. Just as paper maps should be designed for certain specific uses, so too could geographic space be designed. There are many ways that this can be done: with cartographic grids, point-to-point radio beams, hyperbolic grids, the atomic clocks of GPS, or whatever successor systems lie ahead. The specifics certainly matter, but they are perhaps not as important as the overall approach. With these new geo-epistemic practices, the politics of knowledge are not just embedded in our view of the world— in our representations— but in the world itself.

The Politics of Global Coverage

293

CO NCLUSI O N

The Politics in My Pocket

Together the IMW, UTM, and GPS tell a relatively coherent story about globalism and the mapping sciences in the twentieth century. They show a gradual but decisive shift from paper to electronic signals, from the logic of representation to the logic of the grid, from a focus on contiguous areas of space to a framework of points, and from meditations on truth to an interest in practical results. They likewise paint a single picture of the changing relationship between sovereignty, territory, and geographic knowledge. Before World War II, geographic knowledge was tightly linked to territorial control, and mapping was split between national, “practical” projects and international, “scientific” ones. After the war, this pervasive national/international duality receded in the face of a new focus on regional coherence and global totality. Representational mapping continued to be widely employed for national ends, but there was no longer a one-to-one relationship between sovereignty, jurisdiction, and geographic knowledge. International mapmaking became seen as largely pointless, and new kinds of coordinates— both mathematical and electronic— enabled quick intervention across political borders by military and civilian users alike. This new spatiality of knowledge and power was not a retreat from territory; instead, it was an intensification and multiplication of territory beyond the cleanly delimited borders of the territorial state. At a more straightforward level, this has also been a story about a transition from international scientific collaboration to American military technology. The IMW— the preeminent project of global legibility in the early twentieth century— was structured as a multilateral exchange, with its roots in academic geography. The UTM grid and GPS instead began as US military systems that enrolled other countries on terms that were rarely symmetric; US military mapmakers were also key players in the postwar retreat from authoritative representation. Even keeping in mind the centuries-long history of national 295

and military mapping, it still seems fair to say that since World War II global legibility in particular has become more clearly a project of state power.1 And although I have highlighted the many ways that the boundaries between military and civilian, scientific and political, or even American and European can never be cleanly drawn, the United States’ active hand in the construction of a new kind of global infrastructure and its role in the redefinition of territory— including its own territory— are still the overall headlines here. About halfway through the writing of this book, I had lunch with a friend and explained this historical trajectory. In response, my friend asked a simple question: “Is this a good thing, or not?” His concern was not just with American hegemony or the international political system, but also with the GPSenabled cell phone in his jacket pocket— could such a useful technology be implicating him in a larger project of power and coercion? At the time, I responded with a healthy pause and some well-intentioned thoughts about the distinction between historical analysis and moral judgment. As I argued in chapter 6, GPS (and the larger strategy it represents) isn’t inherently bad or inherently good— or even inherently neutral. Like all technologies, it makes some things easier and others more difficult, but there are no simple relationships between technological change (or new forms of power) and easy moral categories.2 Rather than choosing between good, bad, and neutral, we should focus instead on how GPS facilitates intervention from afar and reinforces an embedded, borderless experience of geographic space— not just for the US government, but for countless others as well. This is not a moral response; the goal is simply to understand the changing politics of geographic knowledge. And yet when I finally finished writing, I realized that this response is incomplete. Understanding the relationships between geographic knowledge, power, and territory is not just a task for historians of science or cartography. The changes I have described implicate a much wider array of practices and assumptions— everywhere from political theory and military strategy to the day-to-day operation of surveillance, immigration, or international trade— and my friend’s question is only partly about new spatial technology. In addition, it is also about shifts in geographic power relations more broadly, at all scales, nearly everywhere in the world. This is what is at stake with GPS as a global infrastructure. And while it may be possible to give a good response about specific technological systems, the simple truth is that the larger question is still out of reach. I have tried to provide a framework for analysis, but this can only be the first step. Thus rather than ending this book by pretending to offer a final statement about space and power in the twentieth century, I want to end instead by raising a series of new questions that take my analysis as a starting point. These are questions that have not been widely asked; indeed, they are perhaps questions that it is only now possible to ask for the first time. There are four in particular. 296

Conclusion

1. How did the new spatial technologies and new territoriality of the late twentieth century change what it means to be a state? Or rather, what does it mean for a state to be territorial— and states are indeed still territorial— when territory itself is potentially porous, pointillist, and unbounded? These questions present serious dilemmas for political theory, but they also demand close empirical attention to the day-to-day operation of actual state apparatus. How has the gridding of space shifted the way that resources, information, and people are in fact managed? This is clearly important for understanding the place of the United States and other major powers, all of which created new systems to extend their geographic capacity across traditional borders. But it is no less important for that much larger number of states that have made use of these infrastructures without having any significant say in their design or availability. From the early modern period to the mid-twentieth century, states created their own maps (and censuses, and statistics) to suit their particular purposes, and it made good sense to put each such state at the center of its own story. But a global system like GPS decenters this kind of analysis, since most states— or even within the US, most state agencies— cannot claim exclusive authorship of the knowledge they rely on. What Karl Marx said of individual people could now be modified to apply just as well to governments: they may make their own knowledge, but they do not make it under circumstances of their own choosing.3 2. How have new spatial technologies shifted power relations on the ground? Have alternatives to state-produced geographic knowledge become less possible, or more possible? These are again inherently empirical questions. GPS in particular has often been described as a “force multiplier”: by making space more accessible, it amplifies existing resources.4 But are the resources of states and other large organizations multiplied to the same extent as those of small organizations and individuals? As a working hypothesis, I might suggest that they have not, since even though coordinate technologies tend to favor large organizations, they also make it possible to bypass them altogether. For example, before the advent of GPS surveying (or, to a lesser extent, radiosurveying in general), nearly all geographic knowledge was at some point filtered through at least one large organization— a government mapping agency, a private map publisher, or a large company engaged in transportation or resource extraction. The chance of alternative maps being produced by individuals or small groups was essentially zero. But with electronic positioning, geographic knowledge can now be assembled by anyone with a receiver.5 In short, the changes I have described in this book— the creation of new spatial tools, new geographic subjectivities, and new forms of territory— have tended to amplify existing power asymmetries, but they have also created a genuine opportunity for new alternatives. Indeed, this is exactly what it means for the grid to displace the map as a universal form of knowledge. The power here is generative, not repressive. 3. How have technologies like GPS shifted the patterns of everyday life? These changes can be quite profound, but they are all too commonly forced The Politics in My Pocket

297

into simple, transcultural narratives of progress or decline. For example, instead of unfolding a large piece of paper, millions of people now get from A to B by following a blue dot on a screen. This can easily be seen as a straightforward improvement of the human condition, yet it also conjures up fears that new generations will never learn their local geography or enjoy the unexpected benefits of getting thoroughly lost.6 Rather than choosing between these two mythologies, my approach has instead been to describe a change in priorities: an awareness of a large contiguous area is exchanged for the ability to connect an arbitrary constellation of points. Likewise, even though the embedded experience of point-to-point navigation can indeed be a wonderful thing, small-scale maps still guide our understanding of everything from weather to agriculture to urban segregation.7 But this is a case where adding this kind of analytic nuance may not be enough. My understanding of the popular adoption of GPS is largely drawn from journals and news reports written in languages that I can read, but there are almost certainly local cultures of GPS that have yet to be discovered— or even created. Again, it is worth emphasizing that the benefits, dangers, and politics of electronic coordinates do not follow immediately from US military goals. The “worldwide common grid” mingles promiscuously with other technologies (both new and old), and it can intersect with a wide range of legal and cultural assumptions about privacy, policing, and social order. It is also open to innovation and experimentation by artists, activists, and designers of all kinds.8 4. Overall, how does the history of the twentieth century (or the reality of the present day) look different once we accept that there is no simple dichotomy between the global and the territorial? My previous questions about states, power, and everyday life all require setting aside any easy historical trajectories— toward the irrelevance of the state, toward dystopian domination, or toward a life of geographic detachment. All of these easy narratives are different versions of a single idea: the global world is recent, inevitable, and nonspatial. But it again bears repeating that every project analyzed in this book has been fully worldwide, and also fully engaged with practical problems of geographic knowledge. Territory has not disappeared; it has simply been renegotiated. Yet even in serious academic discussion, the rhetoric of globalism provides a comforting sense that we are somehow always at the cusp of finally transcending geography once and for all. By insisting instead that the global can itself be territorial, my hope is to deflate this expectation and firmly locate our spatiality in specific techniques, specific political assumptions, and a specific historical time. These four questions are meant to be expansive and open ended, but they are also a microcosm of my basic argument. The history of territory is not just a history of war, international treaties, or political theory. It is also, and quite profoundly, a history of geographic tools—of geo-epistemology—and tracing the development and use of these tools immediately connects the trajectories of states and globalization to the everyday experience of measuring and navi298

Conclusion

gating. In many respects, geographic power consists of precisely these kinds of connections and unexamined coordinations between vastly different scales, goals, and intentions. Territory is not just historical at the grand scale; it is also something made and remade every day, by individuals as well as institutions, as a form of knowledge and as a way of inhabiting space.

The Politics in My Pocket

299

ACKNOWL E D G ME NTS

A map of my gratitude would strain the typical conventions of cartography— it would reveal a multidimensional space of everything from serious scholarly debate to casual dinner conversation to off-the-record help navigating archival bureaucracy. It would show not only locations in space and time, but also changes in my own approach and assumptions, my dead ends and near misses, and all those hundreds of minor revelations that eventually led to a larger argument. It would show a well-worn path between my living room and the nearest library, but also a network of dozens of special collections, archives, and generous audiences in four countries. Overlaid on top would be a digital realm of thousands of photographs, maps, PDFs, text files, and easy-tofind-but-impossible-to-find-again websites. In many respects it would mirror the analysis in this book: a boundary-crossing constellation of intellectual points rather than a coherent block of disciplinary space. The research for this book began at Harvard under the combined interdisciplinary umbrella of the history of science and architecture. My greatest thanks goes to Peter Galison, Antoine Picon, Mario Biagioli, and Sven Beckert, who not only offered feedback on my work but also supported my dual PhD from its early days. I also want to thank Nana Last, my mentor at the Rice architecture school, who first pointed me toward the history of science many years ago. Since I arrived at Yale, this project has benefited from a new interdisciplinary mix, and I am grateful for my colleagues in the History of Science and Medicine Program, the History Department, and the wider network of sciencestudies and mapping scholars in New Haven and beyond. I am especially grateful for Yale’s sponsorship of a prepublication colloquium to workshop the final manuscript, and for the close reading and very helpful comments of

301

Dave Kaiser, Dan Kevles, Paul Kramer, and Adam Tooze. Special thanks as well to Luke Bulman for feedback on my illustrations. Countless other colleagues have provided help, feedback, and moral support at various stages. At Yale I am thankful for comments and encouragement from Joanna Radin, Naomi Rogers, John Warner, Paola Bertucci, Henry Cowles, Bill Summers, Jenifer Van Vleck, Paul Sabin, Frank Snowden, Rachel Rothschild, Peter Perdue, and the participants of the Holmes, international history, and environmental history working groups. It has also been wonderful to be part of a wider network of careful thinkers, critical readers, and generous interlocutors, whom I can only list alphabetically and hope someday to repay in kind: Samer Alatout, Ken Alder, Lucia Allais, Babak Ashrafi, Graham Burnett, Deborah Coen, Alex Csiszar, Kenny Cupers, Stephanie Dick, Matthew Edney, Michael Gordin, Jeremy Greene, Katja Guenther, Louis Hyman, Daniel Immerwahr, Caren Kaplan, Chris Kelty, Dániel Margócsy, Erika Milam, Michael Osman, Chitra Ramalingam, Robin Schuldenfrei, Susan Schulten, Hanna Shell, Alistair Sponsel, Matt Stanley, Sara Stevens, Heidi Voskhul, Alex Wellerstein, Kären Wigen, and Ann Wilson. There is a more nebulous constellation of anonymous and invisible support as well. In the last few years I have presented portions of this work at several universities and conferences, and I thank everyone who offered their feedback and hospitality at Columbia, Princeton, NYU, Johns Hopkins, MIT, PACHS, Vanderbilt, Northwestern, UCLA, and UC Davis, as well as the annual meetings of the History of Science Society, the Society for the History of Technology, the American History Association, and the Social Science History Association. The reviewers for the University of Chicago Press and Technology and Culture likewise helped hone my argument and offered excellent suggestions for literature that I would have otherwise missed. I have also been lucky to cross paths with many expert librarians and archivists, who together helped make this project manageable— both by opening new doors and by convincing me that others will remain closed. For their willingness to field endless requests for maps and reproductions, I thank Abe Parrish, Margit Kaye, and Susan Powell at the Yale Map Library, George Billy and Don Gill at the US Merchant Marine Academy Library, Sarah Dickinson at Harvard, and Robin Baird at the US Army Corps of Engineers Library. And for their generosity and professionalism, I thank several archivists and researchers: Gordana Milinic, who made the ICAO archives in Montreal amazingly accessible; Daryl Bottoms, who searched extensively in the records of the Army Map Service and the Aeronautical Chart and Information Service at the US National Archives (NARA); Nathaniel Patch, also at NARA, who searched for material relating to satellite navigation; and historian Leo Slater, who continued the hunt at the Naval Research Laboratory. Researching and writing this book would not have been possible without substantial financial support from several sources. The initial research was made possible by a Jacob K. Javits fellowship from the US Department 302

Acknowledgments

of Education and a Graduate Research Fellowship from the National Science Foundation. I was able to properly finish the research thanks to the Hiebert Fund of the Harvard History of Science Department and a fellowship from the American Council of Learned Societies. The final writing was then funded by a year-long Morse Fellowship from Yale, and the images in the book are subsidized by the Frederick W. Hilles Publication Fund of Yale’s Whitney Humanities Center. Any opinions, findings, conclusions, or recommendations expressed in this book are mine alone and do not reflect the views of these organizations. Finally, big thanks and hugs to my parents, my siblings, my extended family near and far, my friends old and new, and especially Meredith and Quill. Without you, I would have finished this book far too quickly.

Acknowledgments

303

ACRO NYMS A ND CO D E NA ME S

I have tried to use as few acronyms as possible, but since the twentieth century was the most heavily acronymed period in all of human history, this has been difficult. International Organizations ICAN: International Commission for Air Navigation. The only permanent intergovernmental organization for regulating civilian aviation between the two world wars, with headquarters in Paris. It was only loosely affiliated with the League of Nations, and the United States was not a member. (For a map of all members, see chapter 1, note 71.) ICAO: International Civil Aviation Organization. The international regulatory body for civil aviation since its founding in 1944, with headquarters in Montreal. It is part of the “United Nations System,” which comprises the General Assembly in New York City (often just called “the UN”) and roughly a dozen “specialized agencies” like the Food and Agriculture Organization (FAO) and the Universal Postal Union (UPU). The USSR was not a member of ICAO until 1970. Today it includes nearly every country in the world. IGU: International Geographical Union. An international scholarly association founded in 1922. Geographers began holding international conferences in 1871 (known at the time as the International Geographical Congress), but only with the creation of the IGU did these meetings begin to be overseen by a permanent organization. IUGG: International Union of Geodesy and Geophysics. An international scholarly association founded in 1919. Its general assemblies, held every three or four years, have been a major site for scholarly debate and for comparison of national surveying activities. It replaced the International Geodetic Association, which dated to 1862. Initially, Germany and Austria were excluded; unofficial delegates from both countries were invited in the 1930s, but neither became a full member until after World War II. The USSR first asked to join in the late 1930s and became a full member in 1954. 305

Mapping Projects and Coordinate Systems IMW: International Map of the World. A project for a uniform series of roughly one thousand maps covering all land areas of the world at a scale of 1:1,000,000; first proposed in 1891. UTM: Universal Transverse Mercator. A global alternative to latitude and longitude designed by the US Army and in use around the world since the late 1940s. WAC: World Aeronautical Chart. A global map series at 1:1,000,000 created by the US Army Air Force during World War II; the primary postwar competitor to the IMW. Navigation Systems Decca: Not an acronym. A radionavigation system developed at the Decca Company (known for its gramophones and records) for the British navy during World War II; it was later commercialized by a new stand-alone company, Decca Navigator. Gee: A codename, not an acronym. A radionavigation system developed at the Telecommunications Research Establishment for the British air force during World War II. GPS: Global Positioning System. The well-known satellite navigation system sponsored by the US Department of Defense in the early 1970s and fully operational by the early 1990s. Its official name is the Navstar Global Positioning System. (Navstar is sometimes written NAVSTAR, for NAVigation Satellite Timing And Ranging, but officially Navstar is not an acronym.) In common language, “GPS” can also refer to location-aware mapping and navigation software, but the satellites themselves only provide coordinates, not maps. Not to be confused with GIS (Geographic Information System), which refers to computerized mapping of various kinds. Loran: LOng RAnge Navigation. A radionavigation system developed at the MIT Radiation Laboratory during World War II, used by the US Navy, and later opened to civilian users. Oboe: A codename, not an acronym. A radionavigation system developed at the Telecommunications Research Establishment for the British air force during World War II. Omega: Sometimes written OMEGA, but not actually an acronym. A worldwide navigation system developed for the US Navy during the 1960s and internationalized in 1968; it was operational (and available for civilian use) until 1997. Shoran: SHOrt RAnge Navigation. A radionavigation system developed by an engineer at the Radio Corporation of America (RCA) for the US Army Air Force during World War II. Timation: TIMe navigATION. A satellite navigation system developed at the US Naval Research Laboratory in the late 1960s; its principles were incorporated into GPS. Transit: Sometimes written TRANSIT, but not actually an acronym. A satellite positioning system developed by the US Navy in the late 1950s; it was operational (and available for civilian use) until 1996. It is also known as NNSS and NAVSAT, both of which stand for Navy Navigation Satellite System. (NAVSAT is also a generic acronym for NAVigational SATellite.) VOR: VHF [Very High Frequency] Omnidirectional Range. Usually shorthand for VOR/DME, when paired with Distance Measuring Equipment. An American civilian radionavigation system designed at RCA in the late 1930s that gives

306

Acronyms and Codenames

direction to a beacon; DME, a product of World War II with British, Australian, and American roots, gives distance. Both systems were first deployed in the late 1940s and are still widely used in civil aviation. Mapping Institutions (in Citations and Notes) ACS/ACIC: Aeronautical Chart Service/Aeronautical Chart and Information Center. The names of the mapping office of the US Air Force from 1944 to 1952 and from 1952 to 1972, respectively. AGS: American Geographical Society. The oldest geographical society in the United States, founded in 1851 and headquartered in New York City. In addition to publishing journals and organizing scholarly meetings, the AGS has also sponsored many expeditions. AMS: Army Map Service. The name of the mapping office of the US Army from 1942 to 1968. It was also responsible for high-precision surveying, geodesy, and studies of the shape of the earth. GSGS: Geographical Section, General Staff. The name of the mapping office of the British army (the War Office) from 1907 to the 1960s; the British counterpart to the US Army Map Service. OS: Ordnance Survey. The primary domestic mapping agency of the United Kingdom since the late eighteenth century. Despite its name and military origin, it has been a civilian agency since 1870, though with some military staff until 1946. RGS: Royal Geographical Society. The primary geographical society of the United Kingdom, founded in 1830 and located in London. Its activities are similar to those of the AGS. SGA: Service Géographique de l’Armée. The name of the primary mapping agency of France from the late nineteenth century to World War II. Like the British Ordnance Survey, it also published maps for civilian use. USGS: United States Geological Survey. The primary civilian mapping agency of the United States. Originally organized in 1879 to inventory the country’s mineral resources, its major output has always consisted of general topographic maps. Archival Sources (in Citations and Notes) AN: Archives Nationales— the French national archives in Paris F-17: Records of the Ministry of Public Instruction HUARC: Harvard University Archives Papers of John Pierce ICAO: International Civil Aviation Organization— archive in Montreal ICAN: Records of the International Commission for Air Navigation PICAO: Records of the Provisional International Civil Aviation Organization ICAO: Records of the International Civil Aviation Organization Mudd: Princeton University Archives (Mudd Library) Papers of Philip Kissam NARA: National Archives and Records Administration II— the US national archives in College Park, Maryland RG 43: Records of International Conferences, Commissions, and Expositions RG 59: General Records of the Department of State RG 77: Records of the Office of the Chief of Engineers, Army Acronyms and Codenames

307

RG 120: Records of the American Expeditionary Forces (World War I) RG 165: Records of the War Department General and Special Staffs RG 237: Records of the Federal Aviation Administration RG 319: Records of the Army Staff RG 331: Records of Allied Operational and Occupation Headquarters, World War II RG 337: Records of Headquarters Army Ground Forces NASM: National Air and Space Museum— archives in Suitland, Maryland Papers of John O’Keefe NOAA: National Oceanographic and Atmospheric Administration— library in Silver Spring, Maryland PRO: Public Record Office— the British national archives in Kew AIR: Records of the Air Ministry AVIA: Records of the Ministry of Aviation BT: Records of the Board of Trade DR: Records of the Civil Aviation Authority HO: Records of the Home Office FCO: Records of the Foreign and Commonwealth Office MT: Records of the Transport Ministries OS: Records of the Ordnance Survey WO: Records of the War Office UN: United Nations— archive in New York City

308

Acronyms and Codenames

NOT E S

For a full bibliography sortable by chapter, author, and date, see www.afterthemap.info.

Introduction 1. For totals, see Peter Chasseaud, “German Maps and Survey on the Western Front, 1914– 1918,” Cartographic Journal 38 (Dec 2001): 120; Ralph Ehrenberg, “Mapping Land, Sea, and Sky: Federal Cartography from 1775 to 1950,” Government Publications Review 10 (1983): 369; John K. Wright, “Highlights in American Cartography, 1939– 1949,” Comptes rendus du Congrès International de Géographie, Lisbonne 1949 (Lisbon: IGU, 1950), 300; and A. B. Clough, Maps and Survey (London: War Office, 1952), 554. For press coverage, see Gregory Mason, “The Story of the War,” Outlook, 14 Apr 1915, 862; “Arms and the Map,” Print 4 (Spring 1946); or John H. Donoghue, “Maps Must Be Made by the Millions,” Military Engineer 34 (Sept 1942): 427– 430. For quotes, J. S. Dodds, “The Government Mapping Program in a Map-Minded Age,” Science 71, no. 1845 (9 May 1930): 474. See also William Bowie, “The National Mapping Plan of the National Resources Board,” Science 83, no. 2144 (31 Jan 1936). 2. For example, Phil Stanford, “The Automated Battlefield,” New York Times, 23 Feb 1975, SM1; or Andrew Pollack, “War Spurs Navigation by Satellite,” New York Times, 6 Feb 1991, D1. During the Gulf War, US and British forces used roughly 8 million maps— about eight maps per solider. Given the much shorter length and geographic scope compared to World War II, it seems clear that interest in new technologies should not suggest that mapmaking is obsolete. See Michael Russell Rip and James M. Hasik, The Precision Revolution: GPS and the Future of Aerial Warfare (Annapolis, MD: Naval Institute Press, 2002), 138. 3. See, for example, Rip and Hasik, Precision Revolution. Even when historians and sociologists of science have successfully unpacked the technological inevitability of accuracy, this has not shifted the central focus of attention. For a canonical example, see Donald MacKenzie, Inventing Accuracy: A Historical Sociology of Nuclear Missile Guidance (Cambridge, MA: MIT Press, 1990). 4. The connection between legibility and centralization— with the map as an important metaphor— is made explicit in James Scott, Seeing like a State (New Haven: Yale University Press, 1998). 5. For historical definitions of map as representation, see the entry “Map” by Charles Close and Alexander Ross Clarke in the 1911 Encyclopedia Britannica; or J. H. Andrews, “What Was a

309

Map? The Lexicographers Reply,” Cartographica 33 (Winter 1996), 1– 11. For the truthfulness of maps, see Cyrus Adams, “Maps and Map-Making,” Bulletin of the American Geographical Society 44, no. 3 (1912): 199. 6. For a compelling examination of the virtuality of GPS space, see Laura Kurgan, “You Are Here: Information Drift,” Assemblage 25 (Dec 1994): 14– 43. 7. For “utility,” see Bradford Parkinson, Thomas Stansell, Ronald Beard, and Konstantine Gromov, “A History of Satellite Navigation,” Navigation 42, no. 1 (1995): 110; cf. Mark Monmonier, Spying with Maps (Chicago: University of Chicago Press, 2002), 175. 8. For the “territorially bounded nation-state” as “power-container,” see Anthony Giddens, A Contemporary Critique of Historical Materialism, vol. 2, The Nation-State and Violence (Cambridge: Polity, 1985). For an elaboration of the “container” idea, see Peter Taylor, “The State as Container: Territoriality in the Modern World-System,” Progress in Human Geography 18 (1994). 9. See, for example, the rather expansive take on territory in Robert David Sack, Human Territoriality: Its Theory and History (Cambridge: Cambridge University Press, 1986); or David Delaney, Territory: A Short Introduction (Malden: Blackwell, 2005), which define territoriality as geographic control, at any scale for any purpose. The classic comparison of human and animal territoriality is Edward W. Soja, The Political Organization of Space (Washington DC: Association of American Geographers Commission on College Geography, 1971). 10. Weber’s term was Gebiet; see “Politik als Beruf,” in Gesammelte Politische Schriften (Munich: Drei Masken, 1921 [orig. 1919]), 397. 11. Stuart Elden, The Birth of Territory (Chicago: University of Chicago Press, 2013); Peter Sahlins, Boundaries: The Making of France and Spain in the Pyrenees (Berkeley: University of California Press, 1989); John Gerard Ruggie, “Territoriality and Beyond: Problematizing Modernity in International Relations,” International Organization 47 (Winter 1993): 139– 174; Benno Teschke, “Geopolitical Relations in the European Middle Ages: History and Theory,” International Organization 52 (Spring 1998): 325– 358; Silviya Lechner, “Sovereignty and Territoriality: An Essay in Medieval Political Theory” (paper presented at the 6 SGIR Pan-European Conference on International Relations, Turin, 12– 15 Sept 2007). For the long history of sharp boundaries, see John Stilgoe, “Jack-o’-Lanterns to Surveyors: The Secularization of Landscape Boundaries,” Environmental Review 1 (1976): 14– 30. 12. James Sheehan, “The Problem of Sovereignty in European History,” American Historical Review 111 (Feb 2006), 1– 15. For helpful arguments against taking any single date— especially 1648— as the wholesale invention of territorial sovereignty, see Ruggie, “Territoriality and Beyond”; Alexander B. Murphy, “The Sovereign State System as a Political-Territorial Ideal: Historical and Contemporary Considerations,” in State Sovereignty as a Social Construct, ed. Thomas J. Biersteker and Cynthia Weber (Cambridge: Cambridge University Press, 1996); Derek Croxton, “The Peace of Westphalia of 1648 and the Origins of Sovereignty,” International History Review 21 (Sept 1999). On the slippages and fictions of borders, governance, and methodological nationalism in the early twentieth century, see Friedrich Kratochwil, “Of Systems, Boundaries, and Territoriality: An Inquiry into the Formation of the State System,” World Politics 39 (Oct 1986): 27– 52; John Agnew, “The Territorial Trap: The Geographical Assumptions of International Relations Theory,” Review of International Political Economy 1 (Spring 1994); Andreas Osiander, “Sovereignty, International Relations, and the Westphalian Myth,” International Organization 55 (Spring 2001): 251– 287. 13. See, for example, the great effort required to make the Domesday Book legible to a twentieth-century audience, first in H. C. Darby’s twenty-five-year, seven-volume Domesday Geography (Cambridge: Cambridge University Press, 1952– 1977), and more recently online (http://domesdaymap.co.uk). 14. The basic argument is succinctly put in Denis Wood and John Krygier, “Maps,” in The Encyclopedia of Human Geography (New York: Elsevier, 2009). The historical outlines of each individual technology are well known, but they have generally not been subject to monographic treatment. See James R. Akerman, “The Structuring of Political Territory in Early Printed Atlases, Imago Mundi 47 (1995): 138– 154; Roger Kain and Elizabeth Baigent, The Cadastral Map in

310

Notes to Pages 3–7

the Service of the State (Chicago: University of Chicago Press, 1992); Michael Biggs, “Putting the State on the Map: Cartography, Territory, and European State Formation,” Comparative Studies in Society and History 41 (Apr 1999): 374– 405; Ricardo Padrón, “Mapping Plus Ultra: Cartography, Space, and Hispanic Modernity,” Representations 79 (Summer 2002): 28– 60; Monica L. Smith, “Networks, Territories, and the Cartography of Ancient States,” Annals of the Association of American Geographers 95, no. 4 (2005): 832– 849; Jeremy Black, “Government, State, and Cartography: Mapping, Power, and Politics in Europe, 1650– 1800,” Cartographica 43, no. 2 (2008): 95– 105; Jeppe Strandsbjerg, “The Cartographic Production of Territorial Space: Mapping and State Formation in Early Modern Denmark,” Geopolitics 13, no. 2 (2008): 335– 358. For a similar dynamic in the nineteenth century, see Susan Schulten, Mapping the Nation (Chicago: University of Chicago Press, 2012). It is important to note, however, that cartography did not make space nearly as legible and manipulable as its boosters have sometimes claimed. See Matthew Edney, Mapping an Empire: The Geographical Construction of British India, 1765– 1843 (Chicago: University of Chicago Press, 1997); D. Graham Burnett, Masters of All They Surveyed: Exploration, Geography, and a British El Dorado (Chicago: University of Chicago Press, 2000); Raymond B. Craib, “Cartography and Power in the Conquest and Creation of New Spain,” Latin American Research Review 35, no. 1 (2000): 7– 36. 15. Mary Berry, Japan in Print: Information and Nation in the Early Modern Period (Berkeley: University of California Press, 2006); Marcia Yonemoto, Mapping Early Modern Japan: Space, Place, and Culture in the Tokugawa Period (1603– 1868) (Berkeley: University of California Press, 2003); Laura Hostetler, “Contending Cartographic Claims? The Qing Empire in Manchu, Chinese, and European Maps,” in The Imperial Map, ed. James Akerman (Chicago: University of Chicago Press, 2009); Thongchai Winichakul, Siam Mapped: A History of the Geo-Body of a Nation (Honolulu: University of Hawai‘i Press, 1994); Sahlins, Boundaries; Peter Sahlins, “Natural Frontiers Revisited: France’s Boundaries since the Seventeenth Century,” American Historical Review 95 (Dec 1990): 1423– 1451; Susan Schulten, The Geographical Imagination in America, 1880– 1950 (Chicago: University of Chicago Press, 2001); Sumathi Ramaswamy, The Goddess and the Nation: Mapping Mother India (Durham: Duke University Press, 2010). Broader critiques of “nation-state”—or, more forcefully, “the nation-state”—focus on the fact that nationhood is a constructed collective identity that generally follows from statehood, rather than the reverse. See, for example, the classic work of Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism (London: Verso, 1983); and Eric Hobsbawm, Nations and Nationalism since 1780: Programme, Myth, Reality (Cambridge: Cambridge University Press, 1990). 16. The vast majority of research in the history of cartography has focused on the period before 1900, and its implicit object of study has been the transition between a premodern and a modern approach to mapping, territory, and nationalism. This neglect of the twentieth century is commonly noted and stems in part from the field’s historical roots in map collecting. See J. B. Harley, “The Map and the Development of the History of Cartography,” in The History of Cartography, vol. 1, ed. J. B. Harley and David Woodward (Chicago: University of Chicago Press, 1987), 1– 42; Jeremy W. Crampton, “Exploring the History of Cartography in the Twentieth Century,” Imago Mundi 56, pt. 2 (2004): 200– 206; Jeremy W. Crampton, “Cartographic Calculations of Territory,” Progress in Human Geography 35 (Feb 2011): 92–103. 17. For example, compare the borders in figure 1 to the borders shown in Hammond’s New World Atlas (Garden City: Garden City Pub., 1948). Not all partition has happened along existing internal boundaries, but the exceptions— such as the fraught divisions of Punjab or Bengal— only tend to confirm the seriousness of postwar boundary work. Also note the lack of international recognition of unilateral annexations, such as Kuwait, Western Sahara, or the Crimea (all along existing borders). 18. Thomas D. Grant, “Defining Statehood: The Montevideo Convention and Its Discontents,” Columbia Journal of Transnational Law 37, no. 403 (1999); Jennifer Frakes, “The Common Heritage of Mankind Principle and the Deep Seabed, Outer Space, and Antarctica: Will Developed and Developing Nations Reach a Compromise?,” Wisconsin International Law

Notes to Pages 7–8

311

Journal 21, no. 2 (2003); Peter Radan, “Post-secession International Borders: A Critical Analysis of the Opinions of the Badinter Arbitration Commission,” Melbourne University Law Review 24, no. 1 (Apr 2000). 19. For air sovereignty, see John C. Cooper, “Air Transport and World Organization,” Yale Law Journal 55 (Aug 1946); for its upper limit, see “Legal Aspects of Reconnaissance in Airspace and Outer Space,” Columbia Law Review 61 (June 1961): 1083– 1085; or Andrew G. Haley, Space Law and Government (New York: Appleton-Century-Crofts, 1963), 96– 97. Upper boundaries had been discussed before the space age as well; see, for example, W. Carlton Harris, “Real Estate and Real Estate Problems,” Annals of the American Academy of Political and Social Science 148 (Mar 1930): 1– 6. For territorial water before the twentieth century, see Thorsten Kalijarvi, “Scandinavian Claims to Jurisdiction over Territorial Waters,” American Journal of International Law 26 (Jan 1932); for its expansion, see S. N. Nandan, “The Exclusive Economic Zone: A Historical Perspective,” in FAO Essays in Memory of Jean Carroz: The Law and the Sea (Rome: FAO, 1987); Lewis M. Alexander, “The Expanding Territorial Sea,” Professional Geographer 11, no. 4 (1959): 6– 8; Legislation Branch, FAO, “Limits and Status of the Territorial Sea, Exclusive Fishing Zones, Fishery Conservation Zones and the Continental Shelf,” International Legal Materials 8 (1969): 516– 546. For definitions of the continental shelf and mapping, see Luke W. Finlay, “The Outer Limit of the Continental Shelf: A Rejoinder to Professor Louis Henkin,” American Journal of International Law 64 (Jan 1970): 42– 61; Louis Henkin, “The Outer Limit of the Continental Shelf: A Reply to Mr. Finlay,” American Journal of International Law 64 (Jan 1970): 62– 72; Clive R. Symmons, “The Rockall Dispute Deepens: An Analysis of Recent Danish and Icelandic Actions,” International and Comparative Law Quarterly 35 (Apr. 1986): 344– 373; A. D. Couper, “The Marine Boundaries of the United Kingdom and the Law of the Sea,” Geographical Journal 151 (July 1985): 228– 236. There have been three UN Conferences on the Law of the Sea, but the second (in the 1960s) did not result in any new treaties. 20. As elsewhere, much of the interest in the Arctic has been driven by the search for oil and gas deposits, especially now that the ice cap is quickly disappearing. Clear lines of national regulatory jurisdiction are seen as a crucial prerequisite to any eventual drilling. See Clifford Krauss et al., “As Polar Ice Turns to Water, Dreams of Treasure Abound,” New York Times, 10 Oct 2005; “Drawing Lines in Melting Ice,” Economist, 16 Apr 2007; and “Arctic Land Grab,” National Geographic, May 2009 (for which I produced the maps in fig. 3). 21. For an overview, see Michael Lang, “Globalization and Its History,” Journal of Modern History 78 (Dec 2006): 899– 931; Kevin Cox, “Territory, Scale, and Why Capitalism Matters,” Territory, Politics, Governance 1 (2013). The most consistently cited example of overexuberant globalization theory is Manuel Castells, The Rise of the Network Society (Cambridge: Blackwell, 1996); see also Saskia Sassen, The Global City: New York, London, Tokyo (Princeton: Princeton University Press, 1991). For a capsule summary of globalization theory, see Gearóid Ó Tuathail, “The Postmodern Geopolitical Condition: States, Statecraft, and Security at the Millennium,” Annals of the Association of American Geographers 90 (Mar 2000): 166– 178. For a capsule summary of the historical shift, see Charles Maier, “The End of Empire and the Transformations of the International System,” in History of Humanity: Scientific and Cultural Development, vol. 7, The Twentieth Century, ed. Sarvepalli Gopal and Sergei Tikhvinsky (London: Routledge; Paris: UNESCO, 2008), 53. 22. For the function-based “sovereignty” of UN specialized agencies, see Walter Sharp, “The Study of International Administration: Retrospect and Prospect,” World Politics 11 (Oct 1958): 111. For international law, see Anne Orford, “Jurisdiction without Territory: From the Holy Roman Empire to the Responsibility to Protect,” Michigan Journal of International Law 30 (2009). 23. Charles Maier, “Consigning the Twentieth Century to History: Alternative Narratives for the Modern Era,” American Historical Review 105 (June 2000): 807– 831. For a critique avant la lettre of Maier’s argument about territory and engineering, see Antoine Picon, “Naissance du territoire moderne: Génies civil et militaire à la fin du XVIIIe siècle,” Urbi 11 (1989): 100– 114; Marc Desportes and Antoine Picon, De l’espace au territoire: L’aménagement en France, XVIe–XXe siècles (Paris: ENPC, 1997).

312

Notes to Pages 9–12

24. Charles Bright and Michael Geyer, “Late Twentieth-Century Economic Development: What’s New? What’s Different?” (paper presented at the Political Economy of Modern Capitalism workshop, Harvard University, 2 Apr 2007). An exclusively economic understanding of globalism is not uncommon; see, for example, Miles Kahler and Barbara F. Walter, eds., Territoriality and Conflict in an Era of Globalization (Cambridge: Cambridge University Press, 2006). For empirical work that complicates the idea of capitalism as nonpolitical or globalization as antinational, see Sven Beckert, “Das Reich der Baumwolle: Eine globale Geschichte,” in Das Kaiserreich transnational: Deutschland in der Welt 1871– 1914, ed. Sebastian Conrad and Jürgen Osterhammel (Göttingen: Vandenhoeck & Ruprecht, 2004), 280– 301; Sebastian Conrad, Globalisation and the Nation in Imperial Germany, trans. Sorcha O’Hagan (Cambridge: Cambridge University Press, 2010 [orig. 2006]). 25. Neil Smith, American Empire: Roosevelt’s Geographer and the Prelude to Globalization (Berkeley: University of California Press, 2003), 19; Neil Smith, The Endgame of Globalization (New York: Routledge, 2005). Compare to similar arguments in Charles Maier, Among Empires: American Ascendancy and Its Predecessors (Cambridge, MA: Harvard University Press, 2006); Charles Bright and Michael Geyer, “Regimes of World Order: Global Integration and the Production of Difference in Twentieth-Century World History,” in Interactions: Transregional Perspectives on World History, ed. Jerry H. Bentley, Renate Bridenthal, and Anand A. Yang (Honolulu: University of Hawai‘i Press, 2005), esp. 227. For an analysis of American imperial history, see Paul Kramer, “Power and Connection: Imperial Histories of the United States in the World,” American Historical Review (Dec 2011). 26. John Agnew, Globalization and Sovereignty (Lanham, MD: Rowman and Littlefield, 2009), which builds on John Agnew, “Sovereignty Regimes: Territoriality and State Authority in Contemporary World Politics,” Annals of the Association of American Geographers 95, no. 2 (2005): 437– 461. For his call for a pluralist approach to territory in particular, see “Territory, Politics, Governance,” Territory, Politics, Governance 1 (2013): 1. 27. Saskia Sassen, “When Territory Deborders Territoriality,” Territory, Politics, Governance 1 (2013): 38. The larger argument is Saskia Sassen, Territory, Authority, Rights: From Medieval to Global Assemblages (Princeton: Princeton University Press, 2006), which builds on Saskia Sassen, Losing Control? Sovereignty in an Age of Globalization (New York: Columbia University Press, 1996). 28. See in particular the work of Neil Brenner and Joe Painter: Neil Brenner, “Beyond State-Centrism? Space, Territoriality, and Geographical Scale in Globalization Studies,” Theory and Society 28 (Feb 1999); Neil Brenner, New State Spaces: Urban Governance and the Rescaling of Statehood (Oxford: Oxford University Press, 2004); Joe Painter, “Territory-Network” (paper presented at the Association of American Geographers Annual Meeting, March 2006); Joe Painter, “Rethinking Territory,” Antipode 42 (2010). 29. Denis Wood with John Fels, The Power of Maps (New York: Guilford Press, 1992); J. B. Harley, The New Nature of Maps, ed. Paul Laxton (Baltimore: Johns Hopkins University Press, 2001); Christian Jacob, L’empire des cartes (Paris: A. Michel, 1992); Matthew Edney, “Cartography without Progress: Reinterpreting the Nature and Historical Development of Mapmaking,” Cartographica 30 (Summer/Autumn 1993): 54– 68; Matthew Edney, “The Origins and Development of J. B. Harley’s Cartographic Theories,” Cartographica 40, no. 1– 2 (2005), entire issue. For a similar argument outside cartography, see Timothy Mitchell, Rule of Experts: Egypt, Techno-politics, Modernity (Berkeley: University of California Press, 2002). 30. Rob Kitchin and Martin Dodge, “Rethinking Maps,” Progress in Human Geography 31, no. 3 (2007): 335. See also Vincent J. Del Casino Jr. and Stephen P. Hanna, “Beyond the ‘Binaries’: A Methodological Intervention for Interrogating Maps as Representational Practices,” ACME 4, no. 1 (2006): 34– 56. 31. Broadly, my stance here echoes the argument in Bruno Latour, “Why Has Critique Run Out of Steam? From Matters of Fact to Matters of Concern,” Critical Inquiry 30 (Winter 2004): 225– 248. 32. On the false dichotomy of “state” and “nonstate” actors, see Kramer, “Power and Connection.” For zero-sum views of national and global, see Charles Tilly, “Globalization

Notes to Pages 12–14

313

Threatens Labor’s Rights,” International Labor and Working-Class History 47 (Spring 1995): 1– 23; Kevin O’Rourke and Jeffrey Williamson, Globalization and History: The Evolution of a NineteenthCentury Atlantic Economy (Cambridge, MA: MIT Press, 1999); Brian Blouet, Geopolitics and Globalization in the Twentieth Century (London: Reaktion, 2001). 33. In 1938, the International Geodetic Association was described as “without doubt the oldest and most powerful” international scientific organization. It was first proposed in 1861 as a collaboration in central Europe alone; it became European in 1867 and International in 1887. See W. Torge, “The International Association of Geodesy, 1862 to 1922: From a Regional Project to an International Organization,” Journal of Geodesy 78 (2005). For earlier global imaginaries, see Denis Cosgrove, Apollo’s Eye: A Cartographic Genealogy of the Earth in the Western Imagination (Baltimore: Johns Hopkins University Press, 2001). 34. See, for example, word-use data from the Corpus of Historical American English, http://corpus.byu.edu/coha. Although global is definitely a word of the late twentieth century, it first appears with regularity in the 1940s. The growth curve for international is similar but shifted about eighty-five years earlier. It is also instructive to compare the nouns that global has most commonly modified: in the 1940s the top two were war and strategy; in the first decade of the twenty-first century they were warming and economy. (Positioning was ninth.) 35. For a similar megahistorical argument about the relationship between maps, points, and territory, see Stuart Elden, “Missing the Point: Globalization, Deterritorialization and the Space of the World,” Transactions of the Institute of British Geographers, n.s., 30 (2005): 8– 19. 36. The rhetoric of “traditional” territory is widespread, and perfectly understandable, but it tends to falsely oversimplify the past in service of a complex present. 37. This taxonomy is my own, and there are certainly other ways to organize mapping practices. My goal here is not to propose a rigorous classification scheme, but simply to displace cartography as a general term for all practices of geographic knowledge. If nothing else, most of the historical actors I follow would not have described themselves as cartographers. For the invention of cartography as a practice distinct from other fields, see J. B. Harley, “The Map and the Development of the History of Cartography”; Robert McMaster and Susanna McMaster, “A History of Twentieth-Century American Academic Cartography,” Cartography and Geographic Information Science 29, no. 3 (2002): 305– 321; Matthew Edney, “Putting ‘Cartography’ into the History of Cartography: Arthur H. Robinson, David Woodward, and the Creation of a Discipline,” Cartographic Perspectives 51 (Spring 2005): 14– 29. 38. The classic problem of geodesy is the measurement of the deviation of the shape of the earth from a perfect sphere; the first such investigation was initiated in France in the eighteenth century. But for several centuries before this, the word geodesy simply meant surveying, and modern geodesy has been defined as any surveying that takes into account the curvature of the earth. For the history of the shape of the earth, see Irene Fischer, “The Figure of the Earth: Changes in Concepts,” Geophysical Surveys 2 (1975): 3– 54; Ken Alder, The Measure of All Things (New York: Free Press, 2002); John Greenberg, The Problem of the Earth’s Shape from Newton to Clairaut (Cambridge: Cambridge University Press, 1995); M. R. Hoare, The Quest for the True Figure of the Earth (Burlington, VT: Ashgate, 2004). 39. On cases and context, see Peter Galison, Image and Logic (Chicago: University of Chicago Press, 1997), section 1.6; Kristin Asdal and Ingunn Moser, “Experiments in Context and Contexting,” Science Technology & Human Values 37 (2012). 40. For example, Charles Close and Harold Winterbotham (both directors of the Ordnance Survey) were equally involved with the IMW and grids. V. E. H. Sanceau (interwar secretary of the IMW Central Bureau), Phillip Kissam (major promoter of grid systems in the US), and John O’Keefe (designer of UTM) all went on to do work with radiosurveying. 41. The politics of satellite mapping (with Landsat, for example) are similar to those of the global mapping projects discussed in chapter 2. Global projects also rely on the geodetic structures of chapter 4, especially the WGS 84 datum— a major geodetic project fundamental to GPS. For work on aerial and satellite photography, see Jeanne Haffner, The View from Above: The Science of Social Space (Cambridge, MA: MIT Press, 2013); John Cloud, “Hidden in Plain

314

Notes to Pages 14–18

Sight: CORONA and the Clandestine Geography of the Cold War” (PhD diss., University of California, Santa Barbara, 2000); Mark Monmonier, “Aerial Photography at the Agricultural Adjustment Administration: Acreage Controls, Conservation Benefits, and Overhead Surveillance in the 1930s,” Photogrammetric Engineering & Remote Sensing 68 (Dec 2002): 1257– 1261. For GIS, see, for example, John Pickles, ed., Ground Truth: The Social Implications of Geographic Information Systems (New York: Guilford, 1994). 42. Although international collaboration has certainly been important in aerial photography, specific projects have been short lived, and the aerial and satellite mapping projects initiated by the US military did not involve the kind of multilateral arrangements found in geodesy and navigation. As I discuss in chapter 2, collaborations like the international “Global Mapping” project for GIS are explicitly positioned as younger cousins of the IMW. 43. For example, see Hannah Landecker, “Cellular Features: Microcinematography and Film Theory,” Critical Inquiry 31 (Summer 2005): 903– 937; Peter Galison, Image and Logic; David Kaiser, Drawing Theories Apart: The Dispersion of Feynman Diagrams in Postwar Physics (Chicago: University of Chicago Press, 2005); Ursula Klein, Experiments, Models, Paper Tools: Cultures of Organic Chemistry in the Nineteenth Century (Stanford: Stanford University Press, 2003); Joanna Radin, “Latent Life: Concepts and Practices of Human Tissue Preservation in the International Biological Program,” Social Studies of Science 43 (2013). See also the discussions of “epistemic things” or “knowledge objects” in Hans-Jörg Rheinberger, Toward a History of Epistemic Things: Synthesizing Proteins in the Test Tube (Palo Alto: Stanford University Press, 1997); or Karin Knorr Cetina, “Objectual Practice,” in The Practice Turn in Contemporary Theory, ed. T. R. Schatzki, K. Knorr Cetina, and E. von Savigny (London: Routledge, 2001). For earlier interest in tools, see Adele Clarke and Joan Fujimura, eds., The Right Tools for the Job: At Work in Twentieth-Century Life Sciences (Princeton: Princeton University Press, 1992). 44. Cited as “personal communication” in Hubert L. Dreyfus and Paul Rabinow, Michel Foucault: Beyond Structuralism and Hermeneutics (Chicago: University of Chicago Press, 1982), 187. 45. The agency of users and the importance of analyzing use rather than just design is a persistent theme in the history of technology. In much of this literature, however, users are seen as important when they modify a technology, while I am more comfortable seeing enthusiastic uptake as itself a form of agency. See David Edgerton, Shock of the Old: Technology and Global History since 1900 (Oxford: Oxford University Press, 2007); or David Edgerton. “Innovation, Technology, or History: What Is the Historiography of Technology About?,” Technology and Culture 51 (July 2010): 680– 697.

Chapter 1 1. The official title was “Carte du monde au millionième.” The French was chosen because carte was more suitable for bibliographic alphabetization than international; see “Carte Internationale du Monde au Millionème,” Geographical Journal 43 (Feb 1914): 180. The acronym began to be used in the 1960s. 2. The map shows countries that sent delegates to one of the official IMW meetings or sent an official letter of acceptance; see Resolutions and Proceedings of the International Map Committee Assembled in London, November, 1909 (London: HMSO, 1910); Carte du Monde au Millionième: Comptes rendus des séances de la deuxième conférence internationale, Paris, décembre 1913 (Paris: Service Géographique, 1914); appendices to Charles Close, “International Map of the World,” in Atti del X Congresso Internazionale di Geografia, Roma MCMXII (Rome: Reale Società Geografica, 1915), 9– 48; M. N. MacLeod, “The International Map,” Geographical Journal 66 (Nov 1925). 3. A. Penck, “Die Herstellung einer einheitlichen Erdkarte im Massstabe von 1:1 000 000,” Compte rendu du Vme Congrès International des Sciences Géographiques tenu à Berne du 10 au 14 août 1891 (Berne: Schmid, Francke, 1892), 191– 192. Unless otherwise noted, all translations are my own. 4. Arthur Hinks, “The International One-in-a-Million Map of the World,” Royal Engineers Journal 17 (Feb 1913): 78.

Notes to Pages 18–26

315

5. Norman Thrower, Maps and Man (Englewood Cliffs, NJ: Prentice-Hall, 1972), 163, 165. 6. Naturally, the extent and success of these national mapping projects varied widely. For the nineteenth century, see George M. Wheeler, Report upon the Third International Geographical Congress and Exhibition at Venice, Italy, 1881, Accompanied by Data Concerning the Principal Government Land and Marine Surveys of the World (Washington DC: USGPO, 1885). 7. See The International 1:1,000,000 Map: Report for 1938 (Southampton: Ordnance Survey, 1939), 9; International Map of the World on the Millionth Scale: Report for 1957 (New York: UN, 1958), 8– 9. I discuss a few of the most notable cases in chapter 2. The USSR began using IMW sheet lines for domestic mapping as early as 1919. 8. For indications that the IMW was not very well known outside specialist circles, see The International 1:1,000,000 Map: Report for 1932 (Southampton: Ordnance Survey, 1933), 6; “The International Map of the World on the Millionth Scale and the International Cooperation in the Field of Cartography,” World Cartography 3 (New York: UN, 1953): 2; Freeman W. Adams, “Tabula Imperii Romani,” American Journal of Archaeology 58 (Jan 1954): 50. For easily accessible maps at Paris, see Sidney Edward Mezes, “Preparations for Peace,” in What Really Happened at Paris, ed. Edward House and Charles Seymour (New York: Charles Scribner’s Sons, 1921), 5. One other interesting exception proves the rule: in 1968, a legal scholar proposed using the IMW to record authoritative international boundaries, but this plan would have been unpracticable for a number of reasons. See Daniel Wilkes, “Conflict Avoidance in International Law: The Sparsely Peopled Areas and the Sino-Indian Dispute,” William and Mary Law Review 9 (1968): 744– 745. 9. For notable assessments, see Max Eckert, Die Kartenwissenschaft, vol. 1 (Berlin: Walter de Gruyter, 1921), 108– 112; Lloyd Arnold Brown, The Story of Maps (Boston: Little, Brown, 1949), 296– 307; Thrower, Maps and Man; Simon Winchester, “Taking the World’s Measure,” Civilization 2 (Nov/Dec 1995); David Rhind, “Current Shortcomings of Global Mapping and the Creation of a New Geographical Framework for the World,” Geographical Journal 166 (Dec 2000); Alastair Pearson, D. R. Fraser Taylor, Karen D. Kline, and Michael Heffernan, “Cartographic Ideals and Geopolitical Realities: International Maps of the World from the 1890s to the Present,” Canadian Geographer 50, no. 2 (2006); Mark Monmonier, Coast Lines (Chicago: University of Chicago Press, 2008), 86– 95. 10. For a sense of the project’s reach, see the two-hundred-page bibliography compiled by Emil Meynen, International Bibliography of the “Carte internationale du monde au millionième” (International Map of the World on the Millionth Scale) (Bad Godesberg: Bundesanstalt für Landeskunde und Raumforschung, 1962). 11. Michael Heffernan, “Geography, Cartography and Military Intelligence: The Royal Geographical Society and the First World War,” Transactions of the Institute of British Geographers 21, no. 3 (1996): 510. 12. Penck, “Die Herstellung einer einheitlichen Erdkarte,” 191– 198. Perhaps the clearest argument Penck gave for the finality, completeness, and authority of the map was his suggestion that printing costs could be trimmed by using (cheap) lithography for sheets that would likely undergo revision and (expensive) copper engraving for maps of well-surveyed areas. The hope was that eventually all sheets would be worthy of high-quality printing. See E. G. Ravenstein, “A Proposed International Map of the World,” Proceedings of the Royal Geographical Society and Monthly Record of Geography 14 (Oct 1892): 717. 13. These quotes from Eduard Brückner, “Rapport du président de la commission pour l’établissement d’une carte de la terre à l’échelle de 1:1,000,000,” in Report of the Sixth International Geographical Congress, Held in London, 1895 (London: John Murray, 1896), 381, 371. For comparisons with geology, see Albrecht Penck, “Über die Herstellung einer Erdkarte im Maassstab 1:1000000,” in Verhandlungen des Siebenten Internationalen Geographen-Kongresses, Berlin 1899, vol. 2 (Berlin: W. H. Kühl, 1901), 65. For comparisons with the Carte du ciel, see comments of Léotard in discussion of J. V. Barbier, “Projet de carte de la Terre à l’échelle de 1/1.000.000e,” in Compte rendu de la 23me session [Association française pour l’avancement des sciences] (Paris, 1894), 289; A. Penck, “The Construction of a Map of the World on a Scale

316

Notes to Pages 26–29

of 1:1,000,000,” Geographical Journal 1 (Mar 1893): 259; comments of C. E. Stromeyer in Verhandlungen des Siebenten Internationalen Geographen-Kongresses, Berlin 1899, vol. 1 (Berlin: W. H. Kühl, 1901), 215. Note that this comparison became a liability as the Carte du ciel languished: see Hinks, “The International One-in-a-Million Map of the World,” 80. Only later was the project seen as a continuation of past ambitions in cartography; for an early example see Eckert, Die Kartenwissenschaft, vol. 1. 14. That is, distortions due to the projection are less than distortions due to the natural warping of paper. At 1:1,000,000 most projections will satisfy this requirement, as long as the map is not too large. The projection chosen for the IMW was a modified polyconic (sometimes called polyhedric or tronconique). Similar projections had earlier been used for topographic maps in India, the United States, and parts of Germany. See Ravenstein, “A Proposed International Map of the World,” 717; Charles Lallemand, “Sur les déformations résultant du mode de construction de la Carte internationale du monde au millionième,” Comptes rendus de l’Académie des Sciences 153 (1911): 559– 567. 15. The change was made in 1895, both to reduce distortion and for aesthetic reasons. See Brückner, “Rapport du président,” 373– 374; Penck, “Über die Herstellung,” 68– 69. 16. Peter Galison and Lorraine Daston, “Scientific Coordination as Ethos and Epistemology,” in Instruments in Art and Science, ed. Helmar Schramm, Ludger Schwarte, and Jan Lazardzig (Berlin: Walter de Gruyter, 2008). 17. For initial endorsement, see Report of the Sixth International Geographical Congress, Held in London, 1895 (London: John Murray, 1896), 781. For a summary of debates, see Brückner, “Rapport du président,” 367– 369. For specific discussions in the Sociétés de Géographie of Marseille, Paris, and Nancy, see J. V. Barbier, “Le projet de carte de la Terre à l’échelle du 1/1,000,000e,” Bulletin, Société de Géographie de l’Est 16 (1894): 263– 308. 18. “Discussion on the Projected Map of the World,” in Report of the Sixth International Geographical Congress, 379– 380. 19. Ibid., 382. Other vocal opponents of the map included the German geographers Hermann Habenicht and Richard Lüddecke, who had just published a map of Africa at 1:4,000,000. 20. In discussion in Verhandlungen des Siebenten Internationalen Geographen-Kongresses, vol. 1, 210. 21. See “Discussion on the Projected Map of the World,” 382– 386. There was also discussion of an international cartographic association at later conferences, but support waned after Tillo’s death in 1900. See, for example, Tillo’s remarks in Verhandlungen des Siebenten Internationalen Geographen-Kongresses, vol. 1, 221– 227; Franz Schrader, “Note sur la commission préparatoire d’Union cartographique internationale établie par le Congrès géographique de Berlin,” in Report of the Eighth International Geographic Congress, Held in the United States, 1904 (Washington DC: GPO, 1905), 95– 102; Jules de Schokalsky, “Proposition de la création d’une Association cartographique internationale,” in Report of the Eighth International Geographic Congress, 104– 106; “Résolution relative à la préparation d’une Association cartographique internationale et à la publication d’un Répertoire graphique,” in Neuvième Congrès International de Géographie, Genève, 27 juillet–6 août 1908: Compte rendu des travaux du congrès, vol. 1 (Geneva, 1909), 134f. A free-standing International Cartographic Association would not be formed until the late 1950s. 22. Penck, “Über die Herstellung,” 71. Penck remained agnostic about who would actually publish the first maps and offered only vague ideas for securing financial support. His main strategy was to urge his fellow geographers to issue still more resolutions; he also suggested that the congress could make a map template that might save mapping agencies some mathematical work. The trade-off between the meter and Greenwich was indeed an issue, but it was not the roadblock that later scholars have suggested. The French had come to the 1895 meeting ready to adopt the Greenwich meridian in exchange for British use of the meter; the British refused, arguing that a metric map would never sell well in the UK. See Brückner, “Rapport du président,” 372– 375; “Sixth Geographical Congress,” New York Times, 25 July

Notes to Pages 30–32

317

1895, 5. Ultimately, the transition in each case only occurred gradually and piecemeal. France adopted the Greenwich meridian for timekeeping in 1911 and for navigation in 1914, but the Paris meridian (and metric angles) still continued to appear on some French maps. The British adopted the meter for elevations long before they did so for distances. For later overemphasis, see the references in n. 9 (esp. Pearson et al.); Dean Rugg, “The International Map of the World,” Scientific Monthly 72 (Apr 1951): 234; Alastair W. Pearson and Michael Heffernan, “The American Geographical Society’s Map of Hispanic America: Million-Scale Mapping between the Wars,” Imago Mundi 61 (2009): 216. 23. Specifically, the Spanish-American War, Greco-Turkish War, and Boxer Rebellion, and later the Russo-Japanese War and uprisings in Persia. Berthaut had also contributed an in-depth study of possible map projections for the project; see his La Carte de France, 1750– 1898: Étude historique (Paris: Service Géographique, 1899), 337f. For the French Army’s maps, see Albrecht Penck, “Plan of a Map of the World,” in Report of the Eighth International Geographic Congress, 553. The French also started a series of the Austro-Russian frontier that was never published. In the academic press, these maps were presented as a response to past International Geographical Congress resolutions; see “La carte au millionième du Service Géographique de l’Armée,” Annales de Géographie 9 (1900): 176– 177. 24. For British mapping of Ethiopia, Nigeria, Equatorial Guinea, and Morocco, see E. H. Hills, “Note on the Map of Africa on the Scale of 1:1,000,000, Published by the Intelligence Division of the War Office, London,” in Report of the Eighth International Geographic Congress, 569– 570. For larger military scheme, E. H. Hills, “The Present and Future Work of the Geographer,” Geographical Journal 32 (Oct 1908): 393. For the United States, see “Map of the United States on a Scale of 1:1,000,000,” Bulletin of the American Geographical Society 37, no. 12 (1905): 730– 732. For Roosevelt’s reaction and Russian plans, see Albrecht Penck, “Die Erdkarte im Masstabe 1:1000000,” in Neuvième Congrès International de Géographie, 332, 335. 25. At dinner were Franz Schrader (head cartographer at the French publisher Hachette), William Morris Davis (geomorphologist and cofounder of the American Association of Geographers), John Scott Keltie (secretary of the Royal Geographical Society of London), John George Bartholomew (a major Scottish map publisher, known as “Cartographer to the King”), Charles Moore Watson, and Herbert Leland Crosthwait. A subsequent meeting took place on a boat; this included Penck, Close, Davis, Schrader, and Yuly Shokalsky. See Charles Close, “The International Map of the World,” in Geographical By-Ways (London: Edward Arnold, 1947), 111– 112. 26. Close, “The International Map of the World,” 112. 27. Maps were sent by the US, UK, France, Spain, Hungary, Italy, Portugal, Sweden, Argentina, Chile, and Japan— only the first five of which had participated in the first conference. See Carte du Monde au Millionième: Comptes rendus 1913 (see n. 2 above), 71. 28. Emmanuel de Margerie, “La carte internationale du monde au millionième et la conférence de Paris,” Annales de Géographie 23 (1914): 105. 29. Carte du Monde au Millionième: Comptes rendus 1913, 131. The specifications were revised slightly in 1928, but without the same intergovernmental pomp. 30. For the functions of the Central Bureau, see Carte du Monde au Millionième: Comptes rendus 1913, 130– 131. For praise and discussion of its usefulness, see Margerie, “La carte internationale,” 106– 107. For comparisons between the Central Bureau and a Tillo-esque international association, see “The Carte du Monde au Millionième,” Geographical Journal 55 (Jan 1920): 45– 47, or J. E. Robert Bourgeois’s comments in Carte du Monde au Millionième: Comptes rendus 1913, 61. 31. Quotes from Arthur Hinks, “The International One-in-a-Million Map of the World” (see n. 4 above), 80; and H. G. Lyons, “Relief in Cartography,” Geographical Journal 43 (Apr 1914): 398, 403. For an overview of the problem, see Ulla Ehrensvärd, “Color in Cartography: A Historical Survey,” in Art and Cartography, ed. David Woodward (Chicago: University of Chicago Press, 1987). 32. For various competing schemes, see H. G. Lyons, “Relief in Cartography,” pts. 1 and 2, Geographical Journal 43 (Mar and Apr 1914); Eduard Imhof, Cartographic Relief Presentation (Red-

318

Notes to Pages 32–35

lands, CA: ESRI Press, 2007 [orig. 1965]); Ingrid Kretschmer, “The First and Second Austrian School of Layered Relief Maps in the Nineteenth and Early Twentieth Centuries,” Imago Mundi 40 (1988). 33. Max Eckert, “On the Nature of Maps and Map Logic,” trans. W. Joerg, Bulletin of the American Geographical Society 40, no. 6 (1908): 346, 348, 347. 34. C. F. Close, “The Ideal Topographic Map,” Geographical Journal 25 (June 1905): 633– 647. 35. Cyrus Adams, “Maps and Map-Making,” Bulletin of the American Geographical Society 44, no. 3 (1912): 194– 201. 36. Lyons, “Relief in Cartography,” pt. 1, 398– 399. 37. In particular Eduard Brückner, who was a member of the altimetry committee in 1913 and had published, with Arthur Brückner, “Zur Frage der Farbenplastik in der Kartographie,” in Mitteilungen der Kaiserlich-Königlichen Geographischen Gesellschaft 52 (1909). For criticism of the 1909 scheme, see Lyons, “Relief in Cartography,” pt. 2, 243; C. F. Close, “Relief in Cartography,” Geographical Journal 43 (Mar 1914): 347. 38. The original treatise is Karl Peucker, Schattenplastik und Farbenplastik (Vienna: Artaria, 1898). Einthoven’s theory had been published in 1885 as Stereoscopie door Kleurverschil. See Imhof, Cartographic Relief Presentation, 59, 304– 305. Imhof was one of the first to attack Peucker’s ideas, in 1925. 39. For the later influence of the IMW scale, see Joseph E. Williams, “Use of Colors on Relief Maps,” Professional Geographer 3, no. 4 (1951): 38; Erhardt A. Siebert and John E. Dornbach, “Chart Altitude as a Function of Hypsometric Layer Tints,” Navigation (US) 3 (June 1953). 40. Penck, “Plan of a Map of the World” (see n. 23 above), 555. 41. The United States, Canada, Australia, and Russia would all map their own territories. Germany would publish maps of China, France would publish the rest of Asia, and the UK would publish Africa. This left certain parts of the world— notably South America— unclaimed. See Resolutions and Proceedings, 1909 (n. 2 above), 23. 42. For example, Albrecht Penck, “Fortschritte der Herstellung der Einheitliche Erdkarte 1:1.000.000,” in Atti del X Congresso Internazionale di Geografia (see n. 2 above), 7– 8. 43. Carte du Monde au Millionième: Comptes rendus 1913, 116– 117. 44. For the Italian claim to Abyssinia, see Carte du Monde au Millionième: Comptes rendus 1913, 73. The hope was that the planned conference in Berlin would see a similar agreement for the entire world. 45. Letter from the British minister at Peking to the secretary of state for foreign affairs, reproduced in Atti del X Congresso Internazionale di Geografia, 16. For claims of Russia and Japan, see A. R. H[inks], “The 1/Million Map of Spanish America,” Geographical Journal 61 (May 1923): 370. 46. See Franz Schrader, “Note sur la participation de la France à la Carte internationale du Monde au 1/1,000,000,” Apr 1911 (AN, F-17-13063, folder “Carte internationale de la Terre au 1.000.000e”). Berthaut had also been criticized by the French army’s budget office for unnecessary spending on the 1:1,000,000 maps of Asia; see “Commission Centrale des Travaux Géographiques, Séance du 23 Avril 1910” (AN, F-17-13063, folder “Carte internationale de la Terre au 1.000.000e”), 5. 47. Report from Paul Vidal de la Blache, 3 Dec 1909; letter from the minister of public instruction, 7 Jan 1910; letter from minister of war, 12 Jan 1910 (all in AN, F-17-13063, folder “Carte internationale de la Terre au 1.000.000e”). 48. “A Map of the World,” Times (London), 19 Dec 1913, 7. 49. “Boston on the International Map,” Independent, 11 Sept 1913, 644. 50. For “géographique,” see Carte du Monde au Millionième: Comptes rendus 1913 (n. 2 above), 39. For “general,” see Bailey Willis, “The International Millionth Map of the World,” National Geographic Magazine 21 (1910): 127. For “authoritative,” see Close’s comments on Arthur Hinks, “The Map on the Scale 1/1,000,000, Compiled at the Royal Geographical Society under the Direction of the General Staff, 1914– 1915,” Geographical Journal 46 (Aug 1915): 143. For others, see Paul Vidal de la Blache, “La carte internationale du monde au millionième,” Annales de Géographie 19 (1910): 4. The lack of explicit purpose was noted at the time; see C. M.

Notes to Pages 35–41

319

Watson, “Progress in the Sudan: The International Map,” Geographical Journal 40 (Oct 1912): 428; Hinks, “The International One-in-a-Million Map” (n. 4 above), 78. 51. In other words, I disagree with the stronger claim by Denis Wood and John Fels that maps always “mask the interests that brought them into being”; see their The Power of Maps (New York: Guilford Press, 1992), 104. This more radical stance suggests that cartographers are nothing but the unselfconscious tools of capital and nationalism, acting uncritically on behalf of these institutions. In turn, map users are also assumed to naively accept maps’ epistemic claims at face value. For more recent examples, see Emanuela Casti, “Towards a Theory of Interpretation: Cartographic Semiosis,” Cartographica 40 (2005): 1– 16; Vincent J. Del Casino Jr. and Stephen P. Hanna, “Beyond the ‘Binaries’: A Methodological Intervention for Interrogating Maps as Representational Practices,” ACME 4, no. 1 (2005): 34– 56. In reality, maps (and mapmakers) do not always perfectly support their sponsors’ interests, which are rarely monolithic in any case, and users (especially specialists) will often point out their maps’ lacunae. My interest is to understand which maps get made and how they are used, not to critique maps for inevitably falling short of the ideals of objectivity. 52. The “view from nowhere” has been a common trope among historians of science since the late 1980s, following Thomas Nagel, The View from Nowhere (New York: Oxford University Press, 1986). 53. A. Pissis, Mapa de la República de Chile (Santiago: P. Cadot, 1884); Richard Andree, The Times Atlas (London: The Times, 1895)— see in particular the map of Scotland; A. H. Byström report on the IMW in Atti del X Congresso Internazionale di Geografia, 35, which specifically references problems of scale. Note that a sign for sand dunes was added in 1913. 54. Hinks, “The Map on the Scale 1/1,000,000, Compiled at the Royal Geographical Society,” 31. 55. Resolutions and Proceedings, 1909, 4. A similar situation held even in some Latin-script countries. When the Hungarian delegation complained in 1913 that many cities in Hungary had multiple official names in different languages, the committee responded that although the rules did allow for printing a “customary” name underneath the official one, the importance of having a single official name was “a question of principle” and the rules could not be changed. See Carte du Monde au Millionième: Comptes rendus 1913, 73– 74. 56. For colors, see Watson, “Progress in the Sudan: The International Map,” 428. In 1914 a British military officer suggested that naturalistic shading should be added to the IMW to make it more useful “for soldiers, travellers, and for the ordinary individual”; see W. C. Hedley comments on Lyons, “Relief in Cartography” (n. 32 above), pt. 2, 401. 57. For discussion of Latin American borders, see Carte du Monde au Millionième: Comptes rendus 1913, 50, 69– 70, 76. On the use of the map for highlighting poorly known areas, see Franz Schrader’s comments, 146– 147. 58. The IMW standards were revisited in 1928, but the changes only reinforced this same ideal. Symbols for international boundaries, for example, were redesigned for clarity, and new symbols were added for wireless telegraphy, aerodromes, airship stations, tunnels, railway depots, and oil wells; the pattern for forests was deprecated as “unsuitable.” The 1928 conference also issued a specific request to Japan to choose one official system of transliteration and rebuffed an Egyptian request to color its low-lying deserts something other than green. Carte du Monde au Millionième: Rapport pour 1928 (Southampton: Ordnance Survey, 1929), 6, 7, 26. The situation in Japan— where different survey offices had published maps using both the Hepburn and “Nippon Romazikwai” (Nihon-shiki) systems— is explained in J. H. Reynolds, “The Official Romanization of Japanese,” Geographical Journal 72 (Oct 1928): 360– 362. 59. There were, of course, different competing visions, some more racist and oppressive than others. My comments here pertain to the IMW in particular. For civilization more generally, see, for example, Eric Hobsbawm, Nations and Nationalism since 1780: Programme, Myth, Reality (Cambridge: Cambridge University Press, 1990); Timothy Mitchell, Rule of Experts: Egypt, Techno-Politics, Modernity (Berkeley: University of California Press, 2002); H. W. Arndt, Economic Development: The History of an Idea (Chicago: University of Chicago

320

Notes to Pages 41–46

Press, 1987); Charles S. Maier, “The Politics of Productivity: Foundations of American International Economic Policy after World War II,” International Organization 31 (Autumn 1977): 607– 633. 60. Schrader, “Note sur la commission préparatoire d’Union cartographique internationale” (see n. 21 above), 100, 98. 61. Carte du Monde au Millionième: Comptes rendus 1913, 20. 62. Penck, “Die Herstellung einer einheitlichen Erdkarte” (see n. 3 above), 191; Charles Close comments on Arthur R. Hinks, “The Map on the Scale 1/1,000,000, Compiled at the Royal Geographical Society,” 143; Georges Arnaud, “L’état des travaux de la Carte internationale du monde au 1 : 1 000 000e,” Annales de Géographie 34 (1925): 83; Eckert, Die Kartenwissenschaft, vol. 1 (see n. 9 above), 102. 63. Carte du Monde au Millionième: Comptes rendus 1913, 76. 64. Resolutions and Proceedings, 1909, 23. See also Willis (of USGS), “The International Millionth Map of the World,” 130; and W. L. G. Joerg, “Development and State of Progress of the United States Portion of the International Map of the World,” Bulletin of the American Geographical Society 44, no. 11 (1912): 839– 840. 65. Ideology has been a persistent theme in studies of cartography since the late 1980s, especially in work following J. B. Harley. In most contexts ideological seems synonymous with interested or biased, rather than hegemonic as articulated by Gramsci or Althusser. For noteworthy recent examples, see Rob Kitchin and Martin Dodge, “Rethinking Maps,” Progress in Human Geography 31, no. 3 (2007); and Denis Wood and John Fels, The Natures of Maps: Cartographic Constructions of the Natural World (Chicago: University of Chicago Press, 2008), chap. 1. 66. Word-use patterns from full-text searching on Google Books, JSTOR, etc. Especially important were the base maps of the Army Signal Office, the Coast and Geodetic Survey, and the US Geological Survey. For an analysis of the relative incoherence of mapping policy in the United States, see Matthew Edney, “Politics, Science, and Government Mapping Policy in the United States, 1800– 1925,” American Cartographer 13, no.4 (1986): 295– 306. 67. Thematic did not become the preferred term in English until the 1950s. Eckert used both Spezialkarte (special map) and angewandte Karte (applied map). See Max Eckert, Die Kartenwissenschaft, vol. 2 (1925), iii. See also Wolfgang Scharfe, “Max Eckert’s ‘Kartenwissenschaft’: The Turning Point in German Cartography,” Imago Mundi 38 (1986): 64– 65; Denis Wood and John Krygier, “Map Types,” in The Encyclopedia of Human Geography (New York: Elsevier, 2009). 68. The proposal was again made by mapmakers at the US Geological Survey. See Otis Smith in Compte rendu de la XIe session du Congrès Géologique International (Stockholm 1910) (Stockholm: P. A. Norstedt & Söner, 1912), 164– 166. Instead of 1:1,000,000, geologists initiated a world map at the scale of 1:5,000,000; in 1926 this was revised to 1:3,000,000, and the first sheets appeared a few years later. See Carlo von Kügelgen, “The International Geological Congress in Pretoria and a Geological Map of the World,” Science 70, no. 1806 (9 Aug 1929): 142; O. Dottin and G. Gabert, “The Commission for the Geological Map of the World (CGMW) and Small-Scale Earth-Science Cartography of Europe,” Engineering Geology 29 (1990): 387– 391. 69. Counting international projects is always a delicate operation, and I defer to cartographers of the time. See, for example, Karl Peucker, “Die drei Weltkartenprojekte,” Petermanns geographische Mitteilungen 60, pt. 1 (1914): 68– 70. (The third project was a collaboration in aerial photography that was almost immediately ended by the war.) Likewise, a report on the International Aeronautical Map described “the two important series of International Maps”; see “Report by the Maps Sub-commission on Item 1/11” (PRO, AVIA 2/149, “International Commission for Air Navigation: Proceedings of Cartographical Sub-cttee”), 3. In 1925 Close described the IMW as the “only . . . international . . . geographical enterprise”: C. F. C[lose], review of Carte du Monde au millionième: Rapport de 1924, in Geographical Journal 65 (Mar 1925): 258. There were certainly other cartographic projects that spanned political borders— the international geological map of Europe, British Admiralty charts, the General Bathymetric Chart of the Oceans— but these made different claims to internationalism.

Notes to Pages 46–49

321

70. The 1911 proposal was adopted by a few countries in 1912 and was added to ICAN when it was first signed in 1919. The treaty came into force in 1922, but the annex regulating aeronautical maps was not finalized for another three years. For the 1911 proposal, see Charles Lallemand, “Sur un projet de Carte internationale et de Repères aéronautiques,” Comptes rendus de l’Académie des sciences 152 (1911): 1439– 1446. It is difficult to point to any final form of ICAN specifications, since they were almost constantly being revised. But see annex F to Convention portant réglementation de la navigation aérienne en date du 13 octobre 1919 (Paris, June 1925), esp. 332, 35; the plates illustrating annex F were first printed in ICAN Official Bulletin 6 (June 1924). For overviews of aeronautical map standardization see Walter Ristow, Aviation Cartography, 2nd ed. (Washington DC: Library of Congress, 1960); E. Blondel la Rougery, “Les cartes et les repères,” in Premier Congrès International de la Navigation Aérienne, Paris 15– 25 Novembre 1921, vol. 2 (Paris, 1921), 119– 122; A. Duval, “Cartes et procédés de navigation,” in Premier Congrès International de la Navigation Aérienne, 150– 155; E. F. W. Lees, “International Aeronautical Maps,” Geographical Journal 57 (May 1921). 71. For an overview of ICAN, see Laurence C. Tombs, International Organization in European Air Transport (New York: Columbia University Press, 1936); or the pamphlet What the I.C.A.N. Is (Sept 1944; a copy is available in PRO, AVIA 2/2722). The US-backed counterpart to the ICAN was the Havana Convention of 1928, which was signed by all independent countries in the Western Hemisphere except Canada; the Havana Convention did not have a permanent secretariat. This map shows ICAN countries as of 1939, following the May 1946 edition of ICAN, Convention Relating to the Regulation of Aerial Navigation Dated 13th October 1919.

72. The small-scale map was drawn using a Mercator projection with a scale at the equator of 1 degree = 3 cm, which is roughly 1:3,700,000 at the equator and 1:2,500,000 at European latitudes. 73. Lees, “International Aeronautical Maps,” 344, 340. 74. There was some debate about whether the small- and large-scale maps were fundamentally similar or different (see Lees, “International Aeronautical Maps,” 343); the enacted specifications, however, called for fundamentally different kinds of maps. There were also calls

322

Notes to Pages 49–50

for having different maps for flying during the day and at night; see Blondel la Rougery, “Les cartes et les repères,” 119– 122. 75. Quote in a letter from Lees, 29 June 1923 (PRO, AVIA 2/149, “International Commission for Air Navigation: Proceedings of Cartographical Sub-cttee”), 1. The official position is given in “Report on the Proposed Amendments to Paragraph 7 of Annex F of the Convention,” 16 Oct 1923 (PRO, AVIA 2/149, “International Commission for Air Navigation: Proceedings of Cartographical Sub-cttee”). 76. Penel, “Note on the Subject of the Aeronautical Maps,” annex O to minutes of 21 Mar 1930 (PRO, AVIA 2/376, “International Commission for Air Navigation, Proceeding of Cartographical Sub-commission”), 5. For problems of expense and impracticality, see “Note by the Secretary General on Item 1/11a,” annex A to minutes of 21 Mar 1930 (PRO, AVIA 2/376, “International Commission for Air Navigation, Proceeding of Cartographical Sub-commission”). 77. The IHB map (the Carte générale bathymetrique, known today by its English acronym GEBCO) was published at 1:10,000,000. For the decision, see Edouard de Martonne, “Le congrès d’aéronautique coloniale,” Annales de Géographie 41 (1932): 80. 78. The International 1:1,000,000 Map: Report for 1931 (Southampton: Ordnance Survey, 1932), 5; The International 1:1,000,000 Map: Report for 1938 (see n. 7 above), 8. The IMW Central Bureau had also administered the earlier ICAN maps, but not much work had been required; see “Note by the Secretary General on the Question of the Maintenance of the Central Bureau for the International General Aeronautical Map,” 24 Nov 1932 (typescript copy available in vitrine of the ICAO Legal Bureau, Montreal). 79. For a guide to modifying the IMW for aviation, see “Resolution No. 726,” June 1934 (PRO, AVIA 2/704, “Decisions of the I.C.A.N. on Map Questions”). For maps vs. charts, see “Maps versus Charts,” Military Engineer 27 (Sept–Oct 1935). 80. For examples of enforcement, see “Report by the Maps Sub-commission on Item 1/11b,” 22 Dec 1930, 8, or “Report by the Maps Sub-commission on Items 1/11a–1/11b,” 8 Dec 1932, 2 (both in PRO, AVIA 2/533, “Decision of the Main Commission on Map Questions— I.C.A.N.”). For the Italian position, see “Note by the Italian Delegation on Item 1/11,” 3 Nov 1936, annex F to “Maps Sub-commission, Sitting of 18th November 1936” (PRO, AVIA 2/1054, “I.C.A.N. Maps Sub-commission”), 1. For discussion of modifications for radionavigation, see ICAN Maps Sub-commission, “Minutes No. 18: Sittings of 22nd November 1938” (typescript copy available in vitrine of the ICAO Legal Bureau, Montreal), 6– 16. 81. The original map was produced in 1924; an informal agreement between three countries was made in 1929; the Central Bureau began coordinating the project in 1931; official regulations were adopted in 1935. See “Tabvla Imperii Romani,” Geographical Journal 86 (Dec 1935): 523– 526. 82. Charles Close, “The Carte du Monde au Millionième,” Geographical Journal 83 (Apr 1934): 324. Theodor Stocks’s oceanographic sheet showed the South Sandwich Islands; see The International 1:1,000,000 Map: Report for 1938, 4. 83. Quotes from Paul Lebel, “La Tabula Imperii romani,” Annales de Bourgogne 11 (Mar 1939): 23– 24. See also R. G. Collingwood’s review of the map in Journal of Roman Studies 22, pt. 2 (1932): 249. For progress through 1938, see The International 1:1,000,000 Map: Report for 1938. A twelfth sheet, of Germany, was published in 1940 and reviewed by Howard Comfort in the American Journal of Archaeology 45 (Oct–Dec 1941): 648– 649. Freeman Adams described the IMW as “fundamental” to the project: Adams, “Tabula Imperii Romani” (see n. 8 above), 48. 84. There was diversity throughout the 1930s in the treatment of river names, sheet titles, and other details by different countries. See “Tabvla Imperii Romani,” 525– 526; R. A. Gardiner, “The International Map of the Roman Empire,” Geographical Journal 139 (Feb 1973): 111. 85. Carte du Monde au Millionième: Comptes rendus 1913 (see n. 2 above), 112. 86. Penck in discussion with other geographers, in Verhandlungen des Siebenten International Geographen-Kongresses (see n. 13 above), vol. 1, 220. The fact that this allowance was only extended to depicting elevation, however, already gave it a subtle political valence, since in the years before mass aerial photography, generally only national survey agencies could sponsor the leveling and plane-table work required for nonprovisional representation. Most other infor-

Notes to Pages 50–56

323

mation could be obtained through explorers’ traverses or corporate surveys. For discussion of various kinds of surveys, see Raye Platt, “The Millionth Map of Hispanic America,” Geographical Review 17 (Apr 1927); Raye Platt, “Surveys in Hispanic America: Notes on a New Map Showing the Extent and Character of Surveys in Hispanic America,” Geographical Review 20 (Jan 1930); Raye Platt, “Catalogue of Maps of Hispanic America,” Geographical Review 23 (Oct 1933). 87. Information from IMW annual reports; Meynen, International Bibliography (see n. 10 above); Emmanuel de Margerie, “L’Atlas de l’Asie centrale de Sven Hedin,” Annales de Géographie 50 (1941): 196– 200. The Soviet Union also made its own IMW-style maps of Europe between 1932 and 1940, but since the USSR was not part of the project, its maps did not reach an international audience. Some are available on Mapster as “Генеральный Штаб РабочеКрестьянской Красной Армии 1:1 000 000”: http://igrek.amzp.pl/maplist.php?cat=USSR1M. 88. The RGS maps, for example, were particularly flagrant in their disregard for international naming conventions. On AGS exceeding other IMW maps, see H. S. L. Winterbotham, “Recent Sheets of the One-in-a-Million Map of the American Geographical Society,” Geographical Journal 71 (Feb 1928): 172. For similar praise of the Brazilian sheets, see the 1924 annual report. For the relation between the Brazilian maps, governmental efforts, and the AGS, see Arthur Hinks, “The 1/M Map of Brazil,” Geographical Journal 62 (July 1923): 38– 40. While also praising the maps, Hinks pointed out that much of the information presented as definitive “must be conjectural.” On the layering of the RGS maps, see “War Work of the Society,” Geographical Journal 53 (May 1919): 336– 339. 89. Heffernan, “Geography, Cartography and Military Intelligence” (see n. 11 above), 509, 519– 520. 90. For official Brazilian plans as of 1911, see Miguel Lisboa, “Note Concerning the Sheets of the World Map on the Scale of 1:1,000,000 in Brazil,” in Atti del X Congresso Internazionale di Geografia (see n. 2 above), 43– 48. The Brazilian government would eventually riposte as well; see “The Map of Hispanic America on the Scale of 1:1,000,000,” Geographical Review 36 (Jan 1946): 28. 91. “The Map of Hispanic America on the Scale of 1:1,000,000,” 7. For source material, see again the articles by Raye Platt (n. 86 above). 92. Eckert, Die Kartenwissenschaft, vol. 1 (see n. 9 above), 112. 93. “The Carte du Monde au Millionième,” Geographical Journal 55 (Jan 1920): 47. This article was published with no byline, but it is credited to Hinks in Meynen, International Bibliography, 90; it is also written with a distinctively Hinksian brusqueness. 94. Hinks, “The 1/Million Map of Spanish America” (see n. 45 above), 369, 370. 95. H. S. L. Winterbotham, “The International Map of the World,” Geographical Journal 62 (July 1923): 60, Winterbotham, “Recent Sheets,” 172. For the Hinks-Bowman feud, see Heffernan, “Geography, Cartography and Military Intelligence,” 521. 96. In the first reports, the Brazilian series (privately published but still domestic) were shown on the index of nonprovisional maps, while the AGS series was shown as provisional, despite the fact that all these maps used similar graphics and had been labeled “provisional” by their publishers. The 1924 report underscored the graphic interpretation by arguing that provisionality called for “more detailed rules for the treatment of areas and regions not yet sufficiently well surveyed.” In 1928, nonconforming official maps included oversized sheets from Finland and a French map of the Pyrenees that did not include any information from Spain. Due to French misinterpretation of the changing rules, uncolored French maps of the Sahara were also included. See Carte du Monde au Millionième: Rapport pour 1929 (Southampton: Ordnance Survey, 1930), 7; and Carte du Monde au Millionième: Rapport pour 1930 (Southampton: Ordnance Survey, 1931), 8. For ongoing concern with internationality, see The International 1:1,000,000 Map: Report for 1931 (n. 78 above), 6, which states that IMW-style maps of the UK and Denmark that do not use the IMW grid “cannot be classed as properly international.” The problem of national versus international only got worse, as national agencies increasingly published maps without regard for the official IMW grid lines. By 1940 such maps were made by the UK, Ireland, Germany, Spain, France, Denmark, Romania, Estonia, Finland, and Yugoslavia. See Meynen, International Bibliography, index map 5.

324

Notes to Pages 57–79

97. First quote from The International 1:1,000,000 Map: Report for 1931, 6. Second from Carte du Monde au Millionième: Rapport pour 1930, 8. Third from MacLeod, “The International Map” (see n. 2 above), 448; “authorised” was also used in the report for 1930. 98. For its AGS provenance, see Isaiah Bowman, “The Millionth Map of Hispanic America,” Science 103, no. 2672 (15 Mar 1946): 322. It was highly regarded by other cartographers; see for example, Emmanuel de Martonne, “La carte au millionième de l’Amérique du Sud,” Annales de Géographie 32 (1923): 466– 468. It was influential outside the IMW as well, being used, for example, on many maps produced by the Army Map Service after World War II; see http://www.lib.utexas.edu/maps/ams. 99. [V. E. H. Sanceau], “The Present State of the Carte Internationale du Monde au Millionième,” in The International 1:1,000,000 Map: Report for 1938, 8 (originally presented at the International Geographical Congress in Amsterdam, July 1938). 100. The Italian sheets were published in the early 1930s. The series did not show fullcolored elevation and disregarded the 1913 agreement about responsibility for Africa. Italy did not send copies of these maps to the Central Bureau, and they were not included in the official index, but they were noted in the text of the report; see The International 1:1,000,000 Map: Report for 1936 (Southampton: Ordnance Survey, 1937), 5. For their “provisional” status, see A. R. H[inks], “The 1/Million Map of Europe,” Geographical Journal 94 (Nov 1939): 406. For aviation and Egyptian responsibility, see ICAN Central Bureau, “Annual Report,” 2 May 1939 (PRO, AVIA 2/1170, “Publication of Maps for Air Navigation”), 3. 101. For American mapping deficiency before World War II, see F. J. Marschner, “Maps and a Mapping Program for the United States,” Annals of the Association of American Geographers 33 (Dec 1943). For a global summary, see index maps in United Nations Department of Social Affairs, Modern Cartography: Base Maps for World Needs (Lake Success, NY: UN, 1949). 102. On boundaries, see “Use of the American Geographical Society’s Millionth Map of Hispanic America in the Chaco and Other Territorial Disputes,” Geographical Review 25 (Jan 1935): 157. On interaction with the US government, see “The Map of Hispanic America on the Scale of 1:1,000,000,” 24. 103. Hinks, “The 1/Million Map of Europe,” 405– 406. 104. Ibid., 408. 105. “Millionth Map of Hispanic America,” Life, 8 Dec 1941, 104. 106. Bowman, “The Millionth Map of Hispanic America,” 320, 323, 322, 321. See also “Dinner to Celebrate the Completion of the Map of Hispanic America,” Geographical Review 36 (Apr 1946): 312. 107. Spruille Braden, “Congratulatory Address,” Science 103, no. 2672 (15 Mar 1946): 324, 323. 108. Neil Smith, American Empire: Roosevelt’s Geographer and the Prelude to Globalization (Berkeley: University of California Press, 2003), 97.

Chapter 2 1. Eduard Brückner, “Rapport du président de la commission pour l’établissement d’une carte de la terre à l’échelle de 1:1,000,000,” in Report of the Sixth International Geographical Congress, Held in London, 1895 (London: John Murray, 1896), 371. 2. “Report of the Meeting of the United Nations Ad Hoc Group of Experts on the International Map of the World on the Millionth Scale, 9– 11 December 1986,” in Fourth United Nations Regional Cartographic Conference for the Americas, New York, 23– 27 January 1989, vol. 2 (New York: UN, 1992), 42. A few countries have continued making IMW-compliant maps, but these have not been well publicized. The US published a map of Antarctica in 1992; notice of Korean maps from 1995 is given in “Geographical New Names of Korean Cities on the International Map of the World,” working paper 11 for the Eighteenth Session of the UN Group of Experts on Geographical Names, 12– 23 August 1996, http://unstats.un.org/unsd/geoinfo/18th-UNGEGN -Documents.htm; Brazil published new maps in 1998 and 2005.

Notes to Pages 59–65

325

3. R. A. Gardiner, “The International Map of the World: Australia’s Contribution,” Geographical Journal 140 (Feb 1974): 162. 4. David Rhind (former director-general of the British Ordnance Survey), “Current Shortcomings of Global Mapping and the Creation of a New Geographical Framework for the World,” Geographical Journal 166 (Dec 2000): 299. See also Alastair Pearson, D. R. Fraser Taylor, Karen D. Kline, and Michael Heffernan, “Cartographic Ideals and Geopolitical Realities: International Maps of the World From the 1890s to the Present,” Canadian Geographer 50, no. 2 (2006); Jerry Brotton, A History of the World in 12 Maps (New York: Viking, 2013), conclusion. 5. Based on full-text searching, the phrase “truly global war” does not seem to have been used in any prominent publications in English before 1941; by 1942 it is not uncommon. However, this status is not self-evident. A search for “the first truly global conflict” on Google Books gives ninety-four results. The four main contenders are World War II, World War I, the Napoleonic Wars, and the Seven Years’ War. Churchill famously called the Seven Years’ War the “first world war”; see David Reynolds, In Command of History: Churchill Fighting and Writing the Second World War (London: Allen Lane, 2004), 215. 6. The transformative nature of World War II has been argued from many points of view. For the simple idea that aviation and nuclear weapons signaled a departure in the politicalgeographic world system, see John H. Herz, “Rise and Demise of the Territorial State,” World Politics 9 (July 1957); John H. Herz “The Territorial State Revisited: Reflections on the Future of the Nation-State,” Polity 1 (Autumn 1968). Gerhard Weinberg distinguishes the two world wars by their geopolitical “intent”; with the second, “a total reordering of the globe was at stake from the very beginning.” See his A World at Arms: A Global History of World War II (Cambridge: Cambridge University Press, 1995), 2. Even Eric Hobsbawm, who cannot be accused of having a twentiethcentury bias, stresses the globalism of the war; see The Age of Extremes (New York: Pantheon, 1994). 7. The literature on regionalism is vast. For a thorough treatment of regional administration at the international level, see Walter Sharp, Field Administration in the United Nations System (New York: Praeger, 1961). For a more politically oriented synopsis, see Louise Fawcett, “Regionalism in Historical Perspective,” in Regionalism in World Politics, ed. Louise Fawcett, Hurrell Fawcett, and Andrew Hurrell (Oxford: Oxford University Press, 1996). 8. For a clear example of this pragmatic stance, see Joseph S. Nye, International Regionalism: Readings (Boston: Little, Brown, 1968), introduction. 9. The earliest retrospective treatment of these maps is Walter Ristow, “Journalistic Cartography,” Surveying and Mapping 17 (Oct–Dec 1957). For popularity and sales records, see Susan Schulten, The Geographical Imagination in America, 1880– 1950 (Chicago: University of Chicago Press, 2001), 139, 209– 214, 222. The natural foil to wartime journalistic maps— a comparison made by all three scholars cited in the next note— was the early twentieth-century geopolitics of Halford Mackinder, who described global strategy in terms of the relatively static nature of “the natural seats of power.” Whoever controlled the “pivot area” of central Asia would have an impregnable fortress from which to conquer the world. 10. Alan Henrikson, “The Map as an ‘Idea’: The Role of Cartographic Imagery during the Second World War,” American Cartographer 2 (Apr 1975); Schulten, Geographical Imagination; Jeffrey Stone, “Mapping the ‘Red Menace’: British and American News Maps in the Early Cold War Period, 1945 to 1955” (Ph.D. diss., University of Texas at Arlington, 2007). 11. Wendell Willkie, One World (New York: Simon and Schuster, 1943), 2, 38. See also Jenfier Van Vleck, Empire of the Air: Aviation and the American Ascendancy (Cambridge, MA: Harvard University Press, 2013). 12. Richard Edes Harrison, Look at the World: The Fortune Atlas of World Strategy (New York: Knopf, 1944), 23. The first wartime polar map had also been published in Fortune, by the same designer. For a detailed chronology of air-age maps, see Walter Ristow, “Air Age Geography: A Critical Appraisal and Bibliography,” Journal of Geography 43 (Dec 1944). 13. The earliest, most vocal opponent of the Mercator, and the one to brand it “evil,” was the University of Chicago geographer J. Paul Goode, who had campaigned for more appropriate teaching maps as early as 1908; his own projection, published in the 1920s as an equal-area alter-

326

Notes to Pages 65–75

native, nevertheless still showed the US isolated from Europe and Asia. “Facts” quote from “Maps: Global War Teaches Global Geography,” Life, 3 Aug 1942, 64. Laguardia quote in “City’s Schools Will Use Flat ‘Globe’ Maps in Fall,” New York Times, 2 Aug 1943, 12. The American Magazine article was by George Renner, a popular geographer not much respected by his scholarly colleagues; see Richard Edes Harrison, “The War of the Maps,” Saturday Review of Literature 26 (7 Aug 1943): 24. 14. On the “missing link,” see Susan Schulten, “Richard Edes Harrison and the Challenge to American Cartography,” Imago Mundi 50 (1998): 177. 15. The first from the famous journalist Walter Lippmann, the second from the eminent geographer Halford Mackinder; both in Henrikson, “Map as an ‘Idea,’” 32. 16. Historical scare quotes cited in Henrikson, “Map as an ‘Idea,’” 30. Maps centered on cities were ubiquitous. For popular examples, see “Maps: Global War Teaches Global Geography”; Harrison, “War of the Maps,” 24; Erwin Raisz, Atlas of Global Geography (New York: Global Press Corp., 1944). Late in the war several maps were issued using a central point in northern France that had been calculated to allow the most compact representation of land of any air-age map; see J. Parker Van Zandt, The Geography of World Air Transport (Washington DC: Brookings Institution, 1944); S. C. Gilfillan, “World Projections for the Air Age,” Surveying and Mapping 6 (Jan–Mar 1946): 15. See also examples in Stone, “Red Menace”; Harrison was especially fond of hemispheres. 17. The Rand McNally globe is described in Ristow, “Air Age Geography,” 336. For “air ocean” see M. Van Rossum Daum, “Global Maps Representing Reorientation of World Concepts,” in Comptes rendus du Congrès International de Géographie, Lisbonne 1949 (Lisbon: IGU, 1950). For “air-hours” see John H. Donoghue, “The Geography of Total War,” Military Engineer 35 (July 1943): 334. See also Henrikson, “Map as an ‘Idea,’” 28; Stone, “Red Menace,” 102. 18. Henrikson, “Map as an ‘Idea,’” 28. 19. The main battle was between British interest in a “limited cartel” model of international aviation and US support for “free enterprise.” Officially the final result was a compromise, but it was widely seen as a “silent victory” for the US. See Howard Osterhout, “A Review of the Recent Chicago International Air Conference,” Virginia Law Review 31 (Mar 1945); Ivor Thomas, “Civil Aviation: International Questions Outstanding,” International Affairs 25 (Jan 1949); Duane W. Freer, “Chicago Conference (1944),” ICAO Bulletin 41 (Sept 1986): 42– 44. 20. From a 1993 narrative by Donal McLaughlin, chief of the Graphics Division of the Presentation Branch of the Office of Strategic Services at the time of the emblem’s design. Photocopy available in a binder on the shelves of the UN Archives reading room, New York City. 21. There were, however, a few small world maps in the margins to help locate some of the unconventional views. Harrison, Look at the World. 22. In addition to Hinks, see A. J. Dilloway, “Graphics of World Aviation,” Journal of the Royal Aeronautical Society 49 (1945); Albert A. Stanley, “Map Projections for Modern Charting,” Military Engineer 40 (Feb 1948): 55– 58. 23. For Harrison’s work process, see “Perspective Maps: Harrison Atlas Gives Fresh New Look to Old World,” Life, 28 Feb 1944, 61. 24. See, for example, the many examples of globe views in Stone, “Red Menace.” 25. Rand McNally Cosmopolitan World Atlas (Chicago, 1949), preface. In prewar Rand McNally atlases, regional groupings showed groups of countries shaded to form a single wellbounded entity. The Cosmopolitan showed regions much as Harrison did: as a zoomed-in view on a larger whole. See also Schulten, Geographical Imagination, 229– 238. 26. Harrison, Look at the World, 10– 12. For others, see “Maps: Global War Teaches Global Geography”; Otis P. Starkey, “Maps Are Liars,” New York Times Magazine, 11 Oct 1942, 16. 27. The mapping activities of Italy, France, and the minor powers were not substantial. Little is available about the history of cartography in the USSR, but the general sense is that the Soviets’ first priority during the war was to shore up their own domestic mapping; maps of foreign areas only became a priority once the USSR began advancing west. The Soviets had compiled maps of Europe at 1:1,000,000 in the 1930s (see chap. 1, n. 87), but this program was not resumed until 1946. See John Cruickshank, “Military Mapping by Russia and the Soviet

Notes to Pages 75–80

327

Union,” in Cartography in the Twentieth Century, ed. Mark Monmonier (Chicago: University of Chicago Press, 2015), 932– 942; Alexey V. Postnikov, “Maps for Ordinary Consumers versus Maps for the Military: Double Standards of Map Accuracy in Soviet Cartography, 1917– 1991,” Cartography and Geographic Information Science 29 (2002). 28. For German coverage, see index map 9 in E. Meynen, International Bibliography of the “Carte internationale du monde au millionième” (International Map of the World on the Millionth Scale) (Bad Godesberg: Bundesanstalt für Landeskunde und Raumforschung, 1962). For Japan, see William E. Davies, “Axis War Maps,” Surveying and Mapping 8 (July–Sept 1948): 130. 29. See Jacob Skop (chief of map design at the AMS), “The New 1:250,000 Scale Map of the Army Map Service,” Surveying and Mapping 7 (July–Dec 1947): 161. 30. Alfred H. Burton, Conquerors of the Airways: A Brief History of the USAF-ACIC and Aeronautical Charts (St. Louis: ACIC, 1953), 36– 46. 31. Progress of the IMW as reported in The International 1:1,000,000 Map: Report for 1938 (Southampton: Ordnance Survey, 1939) was 361 sheets; estimates for total number of sheets ranged from roughly 900 to 1,050, depending on how oceanic islands were treated. 32. And like the IMW, most of the WAC sheets were based on recompilations from existing maps; the WAC also used a similar convention of dashed contour lines to show areas that were insufficiently surveyed. Unlike the IMW, however, WAC maps of remote areas often included large expanses colored a conspicuous yellow and simply labeled “unsurveyed.” 33. For AGS participation, see “The Map of Hispanic America on the Scale of 1:1,000,000,” Geographical Review 36 (Jan 1946): 23– 24. 34. John K. Wright, “Highlights in American Cartography, 1939– 1949,” in Comptes rendus du Congrès International de Géographie, Lisbonne 1949, vol. 1, 300. 35. Sidney A. Tischler, “Procedural Developments in Trimetrogon Compilation,” Photogrammetric Engineering 14 (Mar 1948): 60. 36. Aeronautical Chart Service, Reconnaissance Mapping with Trimetrogon Photography (Washington DC, 1943), 8. 37. Both the Army Map Service and the Aeronautical Chart Service were subject to constant administrative reshuffling and renaming. For user-friendliness, I use one name consistently for each, reflecting the name most common in the published literature about wartime and postwar mapping; the AMS name was used officially from 1942 to 1968, the ACS name from 1944 to 1952. Full administrative chronology is detailed by NARA; see http://www .archives.gov/research/guide-fed-records/groups/456.html. 38. For the AMS, see “Arms and the Map,” Print 4 (Spring 1946); John H. Donoghue, “Maps Must Be Made by the Millions,” Military Engineer 34 (Sept 1942): 427– 430. For ACS, see Burton, Conquerors of the Airways, 43. 39. Donoghue, “Maps Must Be Made,” 429. 40. Wright, “Highlights”; Walter W. Ristow, Aviation Cartography, 2nd ed. (Washington DC: Library of Congress, 1960). 41. Davies, “Axis War Maps,” 128. 42. See chapter 1; cf. Peter Galison and Lorraine Daston, “Scientific Coordination as Ethos and Epistemology,” in Instruments in Art and Science, ed. Helmar Schramm, Ludger Schwarte, and Jan Lazardzig (Berlin: Walter de Gruyter, 2008). 43. Davies, “Axis War Maps,” 127. 44. Ibid., 130. 45. A postwar edition of these specifications is available at NOAA: Aeronautical Chart Service, Specifications for World Aeronautical Charts, 5th ed. (Washington DC, 1946); see esp. p. 8. 46. Tischler, “Procedural Developments,” 53. 47. Skop, “The New 1:250,000,” 163; J. G. Ladd, “Maps for Korea,” Military Engineer 42 (Nov–Dec 1950): 448– 450. 48. For details of the Loper-Hotine agreement (named for the heads of US and UK military mapping, Herbert Loper and Martin Hotine), see A. B. Clough, Maps and Survey (London: War Office, 1952), 43– 48.

328

Notes to Pages 81–85

49. Clough, Maps and Survey, 44– 45. 50. Military Mapping & Aeronautical Charting Conference Held at the Directorate of Military Survey 1st–11th September, 1947: Proceedings & Resolutions, marked “secret” (PRO, WO 163/487). 51. Agreements in the 1950s are detailed in Army Map Service, “Special Study” (of past agreements dividing mapping responsibility), 2 May 1956 (NARA Cartographic Records, RG 77, box 12 of 54, folder “Mapping & Geodetic Publications and Files”). Updated maps of responsibility are in Proceedings of the Commonwealth Military Survey Conference 1959 (NARA Cartographic Records, RG 77, box 3 of 3). 52. Schulten, Geographical Imagination; Stone, “Red Menace” (see n. 10 above); Denis Cosgrove, “Maps, Mapping, Modernity: Art and Cartography in the Twentieth Century,” Imago Mundi 57 (2005). 53. For wartime activity and dues, see “Central Bureau for 1/M Map,” 24 Sept 1941, and “Subscriptions Received during the War Years,” n.d. (both in PRO, OS 1/343, “Central Bureau for the International Map of the World and International General Aeronautical Map”). The UK tried to rid itself of responsibility for the Central Bureau at the first opportunity, while the US felt the project should become completely decentralized. See UK Steering Committee on International Organizations, “Survey of Inter-governmental Organisations,” 15 June 1949, and letter from OS director general to the UK secretary of state for foreign affairs, 20 Oct 1948 (both in PRO, OS 1/343, “Central Bureau for the International Map of the World and International General Aeronautical Map”). 54. See meeting minutes and reports of Subcommittee 7 of Committee II: Aeronautical Maps and Charts, 9– 21 Nov 1944 (PRO, AVIA 2/2722, “International Civil Aviation Conference Chicago 1944: U.S. Technical Annex Aeronautical Charts”). For an appraisal of the committee’s work, see “Note of Informal Discussion Held on Friday, 9th February, 1945: Maps for Civil Aviation” (PRO, AVIA 2/2722, “International Civil Aviation Conference Chicago 1944: U.S. Technical Annex Aeronautical Charts”). The final specifications can be found in Proceedings of the International Civil Aviation Conference (Washington DC: USGPO, 1948), 359– 364. 55. The American representative was careful to frame his offer as one motivated by sentiments of generosity and internationalism. At the conference he argued that because the WAC relied on previously published material from around the world, it should properly be seen as “an international achievement”; it was only “by chance” that the US had been the one to produce it. Paul A. Smith, citing his own remarks, in “Surveying and Mapping at Civil Aviation Conference,” Surveying and Mapping 5 (Apr 1945): 23– 24. 56. The mapping obligation was implicit until 1959, when it was formally included as part of annex 4. See ICAO Aeronautical Information Services and Aeronautical Charts Division, Report of the Meeting: Montreal, 28 April–25 May 1959, ICAO Doc 7993, AIS/MAP (Montreal: ICAO). For prioritization of WAC over IMW, see Canadian comment of 19 July 1951 or South African comment of 30 Oct 1951 in PRO, OS 1/769, “Central Bureau for the International Map of the World (and International General Aeronautical Map).” 57. A. G. Ogilvie, “Project for International Population Map on a Scale of 1:1.000.000,” in Comptes rendus du Congrès International de Géographie, Lisbonne 1949, vol. 1 (see n. 17 above), 267– 268; [L. Dudley Stamp], “A World Land Use Survey,” Geographical Journal 115 (Apr–June 1950): 223– 226; Henri Gaussen, “Projets pour diverses cartes du monde à 1/1.000.000: La Carte Ecologique du Tapis Végétal,” in Comptes rendus du Congrès International de Géographie, Lisbonne 1949, vol. 1, 224– 225. All three men were in their late fifties or early sixties; these were not upstart proposals, as the original IMW had been. 58. Comptes rendus du Congrès International de Géographie, Lisbonne 1949, vol. 1, 122. 59. International Geographical Union, Report of the Commission on the International Map of the World 1:1,000,000 (New York: IGU and AGS, 1952), 17. Hurault’s report was written in July 1950. 60. For IGU logic, see UK Steering Committee on International Organizations, “Survey of Inter-governmental Organisations,” 28 June 1949 (PRO, OS 1/343, “Central Bureau for the International Map of the World and International General Aeronautical Map”). For the UN, see United Nations Department of Social Affairs, Modern Cartography: Base Maps for World Needs (Lake Success, NY: UN, 1949).

Notes to Pages 86–92

329

61. John K. Wright to R. Ll. Brown, 19 Nov 1951, with attachment “Inquiry Regarding Relationship of IMW to WAC” (PRO, OS 1/769, “Central Bureau for the International Map of the World (and International General Aeronautical Map)”). 62. Most “international” thematic maps were simply overprints on whatever national topographic or aeronautical maps were available; at best the IMW only regulated the sheet lines and titles. 63. O. G. S. Crawford, “Mapping the Roman Empire,” Geographical Journal 20 (Sept 1954): 363. The project was taken over by the Union Académique Internationale and eventually reconceived as a series of books accompanied by folded maps, each organized by national archaeological agencies and published in the local language, rather than as maps accompanied by gazetteers. See R. A. Gardiner, “The International Map of the Roman Empire,” Geographical Journal 139 (Feb 1973); J. F. Drinkwater review of Tablua Imperii Romani sheet M31, Britannia 8 (1977): 466– 468. 64. Christopher Board, “Land Use Surveys: Principles and Practice,” in Land Use and Resources: Studies in Applied Geography (London: Institute of British Geographers, 1968). W. William-Olsson, “The Commission on a World Population Map: History, Activities and Recommendations,” Geografiska Annaler 45 (1963): 244. 65. The other geographers (P. Legris and M. Viart) were also French. Catalog of maps available online from the Institut Français de Pondichéry, http://www.ifpindia.org/ecrire /upload/ifp_ecology_publication.pdf. Maps were published covering all of India, Madagascar, Cambodia, Ceylon, and Sumatra. Isolated maps were published for areas of Tunisia, Algeria, Niger, Chad, and Mexico. 66. Crawford, “Mapping the Roman Empire,” 364. 67. For Gaussen epithet, see A. W. Küchler, “Vegetation Mapping in Africa,” Annals of the Association of American Geographers 50 (Mar 1960): 83. Even though thematic mapping was enthusiastically supported by specialized UN agencies like FAO and UNESCO, these organizations tended to provide only subventions and forums for debate or publicity rather than the strong managerial hand sought by geographers. For the role of UNESCO in the World Land Use Survey, see Dudley Stamp, “Ten Years On,” Transactions of the Institute of British Geographers 40 (Dec 1966): 11– 20. For the role of FAO, see IGU Commission on World Land Use Survey, Report of the Commission on World Land Use Survey for the Period 1949– 1952 (Worcester: IGU, 1952). 68. A. W. Küchler, “An International Vegetation Map on the Millionth Scale,” Geographical Review 54 (July 1964): 435. 69. Bruce Mitchell, Geography and Resource Analysis, 2nd ed. (New York : Wiley, 1989), 48– 52; International Geographical Union Commission on Inventory of World Land Use, Report of the Commission on Inventory of World Land Use (New York: IGU, 1956). 70. A series of articles appeared in Geografiska Annaler 45 (1963); see especially those by Norman Thrower and W. William-Olsson. See also James H. Johnson’s review of population maps of Scotland and California, Geographical Journal 133 (Dec 1967): 580– 581. 71. Crawford, “Mapping the Roman Empire,” 364. 72. For the World Land Use Survey, a Swiss geographer argued that globally uniform maps at 1:1,000,000 were only useful for schools and “politicians”; planners needed larger-scale maps tailored to local idiosyncrasies. See Hans H. Boesch, “The World Land Use Survey,” Internationales Jahrbuch für Kartographie 8 (1968): 139. Board, “Land Use Surveys,” 36, notes that these local variations meant that there was “no world survey in any strict sense.” For population mapping, see W. William-Olsson, “Data on Population Mapping from the Correspondence of the Commission,” Geografiska Annaler 45 (1963): 288– 289. 73. Namely India, Pakistan, Ceylon, and Nepal. See A. W. Küchler, “An International Vegetation Map.” 74. See again Gardiner, “International Map of the Roman Empire”; and Drinkwater review. The maps I could locate are much later, but seem consistent with these earlier descriptions. See, for example, Tabula Imperii Romani: Hoja K-29 (Madrid: Instituto Geográfico Nacional, 1991); or Tabula Imperii Romani: Iudaea-Palestina (Jerusalem: Israel Academy of Sciences and

330

Notes to Pages 93–96

Humanities, 1994). Notably, both only show detail in areas explored by Spanish or Israeli scholars. 75. Compare, for example, the presentation of the first AGS sheet in 1923 with the first World Land Use Survey project in 1958. The AGS map (of the central Andes) was designed to cover as many extremes as possible; the land use pilot (of Hong Kong) showcased local specificity. See A. R. H[inks], “The 1/Million Map of Spanish America,” Geographical Journal 61 (May 1923): 369– 371; Robert W. Steel, “The World Land Use Survey,” Nature 4645 (8 Nov 1958): 1285– 1286. 76. For a general overview of ICAO mapping activity, see Ristow, Aviation Cartography; or E. R. L. Peake, “The Activities of ICAO, with Particular Reference to Those of Its Map Division,” in Conference of British Commonwealth Survey Officers 1947: Report of Proceedings (London: HMSO, 1951). None of the delegates at the first meeting in 1944 were present at the fifth, in 1951. Of the sixty-three delegates at the sixth meeting in 1959, only three had participated eight years before. 77. Strict requirements for the location of marginalia were also eliminated. In Europe, the lowest elevations would be drawn in white; in the Americas they would be green. (Greenland could be drawn in either style.) The US objected to this change, but European countries were in strong support. See discussion in ICAO box “MAP— 3, 4, & 5: 1947– 1951”; for final results, see ICAO Aeronautical Charts Division, Report of the Fifth Session: Montreal, 9– 29 October 1951, ICAO Doc 7222-MAP/567 (Montreal: ICAO, 1951), 24. 78. ICAO, Report of the Fifth Session, 23, 51, 102. 79. James R. Park, “Aeronautical Charts: The Role of the International Civil Aviation Organization in Cartography,” Canadian Surveyor 13 (Jan 1956): 32. 80. There were differences. For example, in Canada, Australia, and North Africa, sheet lines were nudged to align with existing national map series. On the difficulty of voluntary agreements and the need for a strong central hand, see PICAO Monthly Bulletin, Mar 1947, 8– 9; Peake, “The Activities of ICAO.” The final sheet allocations were approved in March 1948. 81. For the uselessness of 1:1,000,000 for small islands, see letter from J. M. Buckeridge (UK) to I. T. Perry (Australia), 19 May 1953 (PRO, BT 249/2, “Maps and Charts: ICAO 1:1,000,000 Topographical Charts— Policy”). 82. For the logic of the original scheme, see Peake, “The Activities of ICAO,” 186. For Egypt, see note dated 27 Sept 1951, MAP V-WP/39 (ICAO, box “MAP— 3, 4, & 5: 1947– 1951”); ICAO, Report of the Fifth Session, 29– 30 and supp. 1, p. 4. 83. ICAO, Report of the Meeting: 1959 (see n. 56 above), p. 6-53. This was amplified further at an ICAO charting meeting in April 1966; see International Map of the World on the Millionth Scale: Report for 1966 (New York: UN, 1968), 3– 6. 84. By the early 1970s, the UK had stopped publishing charts for its former colonies; West Germany, France, and Brazil also discontinued their maps. See editions of the ICAO Aeronautical Chart Catalogue, ICAO Doc-7101 (Montreal: ICAO), published between 1955 and 1994. For the tendency toward regional coverage and concerns about “isolated blocks,” see “Map Policy— I.C.A.O. 1:1,000,000 Topographical Series,” n.d. [1950s] (PRO, BT 249/2, “Maps and Charts: ICAO 1:1,000,000 Topographical Charts— Policy”). 85. For worries of obsolescence and overstandardization, see, for example, R. Ll. Brown (future director-general of the Ordnance Survey), comment on Peake, “The Activities of ICAO,” 199; John D. Kay (US C&GS), “Capsule Charts for Universal Navigation,” Navigation (US) 2 (Sept 1950): 203; compare to ICAO, Report of the Fifth Session, 23. 86. For this research, undertaken by the contract research firm of Dunlap and Associates, see “Better Charts for High-Speed Avigation,” Research Reviews [ONR] (Dec 1951): 13– 17; Rolland H. Waters and Edward W. Bishop, “The Design of Aeronautical Charts,” Navigation (US) 3 (Mar 1952); John E. Murray and Rolland H. Waters, “The Design of Aeronautical Charts: II,” Navigation (US) 3 (Dec 1952). For later work, see, for example, “Jet Navigation Charts,” Revista Cartográfica 5 (1956): 183– 187; Lavoi B. Davis (of ACIC), “Design Criteria for Today’s Aeronautical Charts,” Surveying and Mapping 18 (Jan–Mar 1958); Richard W. Philbrick (of ACIC), “New Design Features for the World Aeronautical Chart,” Surveying and Mapping 17 (July–Sept 1957).

Notes to Pages 99–102

331

87. Philbrick, “New Design Features,” 312; emphasis in original. For explicit reference to psychology and psychophysics, see Robert J. Schreiber, “Psychologists Test Jet Charts,” Aviation Age 22 (Sept 1954): 70– 73; Burton, Conquerors of the Airways (see n. 30 above), 169– 170. The 1952 study was also published in a psychology journal: John E. Murray, “An Evaluation of Two Experimental Charts as Navigational Aids to Jet Pilots,” Journal of Applied Psychology 37 (June 1953): 218– 222. 88. John E. Dornbach (of ACIC), “Design Problems in Aeronautical Charting,” Professional Geographer 11 (Mar 1959): 8. 89. See again chapter 1; or Daniel R. Montello, “Cognitive Map-Design Research in the Twentieth Century: Theoretical and Empirical Approaches,” Cartography and Geographic Information Science 29 (2002). Montello sees the work of Arthur Robinson (head of wartime cartography for the US Office of Strategic Services), especially The Look of Maps, from 1952, as the major source of early map-design research. While Robinson was no doubt vitally important, none of his work is mentioned in aeronautical-chart research until in the late 1950s. For more on Eckert, especially his limited influence in the 1920s and 1930s, see Wolfgang Scharfe, “Max Eckert’s ‘Kartenwissenschaft’: The Turning Point in German Cartography,” Imago Mundi 38 (1986). 90. For “tools,” see Kay, “Capsule Charts,” 202. 91. Dornbach, “Design Problems,” 8. 92. Frederick O. Diercks (head of AMS), “The Role of the Topographic Map in the Missile Age,” Surveying and Mapping 19 (Sept 1959): 350. See also “Jet Navigation Charts.” 93. Dornbach, “Design Problems,” discusses relative relief. For maximum elevation data on the WAC, see Philbrick, “New Design Features.” 94. Waters and Bishop, “The Design of Aeronautical Charts,” 76; emphasis in original. See also Kay, “Capsule Charts.” 95. Paul Schauer, paraphrased in Waters and Bishop, “The Design of Aeronautical Charts,” 75; emphasis in original. 96. Dornbach, “Design Problems,” 8. 97. Philbrick, “New Design Features,” 305. 98. Richard W. Philbrick, “Charts for the Air Force,” Journal of the Surveying and Mapping Division 84 (Nov 1958), pp. 1843-5 and 1843-6. 99. First quote from ibid., p. 1843-6. Second from Philbrick, “New Design Features,” 305. See also “Jet Navigation Charts.” 100. Most important, it did not include information on communications, airways, or airspace, and it showed terrain using a system of “relative relief” that prioritized land cover and ruggedness over elevation above sea level. The Operational Navigation Chart was begun in 1960; over the next twenty years it replaced the American WAC outside North America. The WAC, however, was (and is) still maintained as a civilian series within US airspace. 101. From France: “Preliminary Comments and Suggestions of Col. M. Dévé Relating to the Agenda,” 10 Feb 1958; statistics: “Current Demand for Aeronautical Charts,” 20 Nov 1958; from the UK: “Distinction between Charts for Airline and for General Use,” 26 Nov 1958 (all in ICAO, box “MAP Panel 1 & 2, 1958”). 102. For proposals to reduce the number map types, see note from Canada, 3 Oct 1951, MAP V-WP/43 (ICAO, box “MAP— 3, 4, & 5: 1947– 1951”). 103. ICAO, International Standards and Recommended Practices: Aeronautical Charts: Annex 4 to the Convention on International Civil Aviation (Montreal: ICAO), 1957 and 1961 editions. 104. Cyrus G. Finley Jr., “Military Topographic Maps of the Future,” Surveying and Mapping 13 (Oct–Dec 1953): 491. 105. John I. Norris, “Mapping Activities and New Warfare Concepts,” 12 Dec 1956, in Notes of the Week (Tokyo: US Army Map Service, Far East, 1956– 7), 3– 4. Typescript papers available at NOAA. 106. Unlike pilots, ground troops tended to prefer layer tints over naturalistic shading, though preferences varied by the scale of the map. The program is described in Jacob Skop,

332

Notes to Pages 103–107

“Modern Warfare and the Map,” in Second International Cartographic Conference, Chicago, June 15– 21 1958 (spiral-bound papers available at NOAA). The techniques he describes were not wholly unprecedented; see Albert M. Walker, “The Military Map of the Future,” Military Engineer 15 (Sept–Oct 1923): 391– 397. Quote from Finley, “Military Topographic Maps of the Future,” 490. 107. When map use was discussed in the 1920s and 1930s, for example, the emphasis was more on training soldiers to read maps rather than designing maps for easy reading. See McK. W. Kriegh, “Preparation and Distribution of Maps,” Military Engineer 15 (May–June 1923): 241– 244; Walker, “The Military Map of the Future”; John W. Davis, “The Status of Military Mapping,” Military Engineer 17 (July–Aug 1925): 292– 294. 108. J. C. T. Willis to AMS, 2 Apr 1951 (NARA Cartographic Records, RG 77, box 36 of 89, folder “World 1:1,000,000: Specs. / Notes”). 109. “Plastic Map of Europe, Scale 1:1,000,000,” 14 Feb 1952 (NARA Cartographic Records, RG 77, box 30 of 109, folder “1:1,000,000 File (Program)”). 110. “Notes for Col. Diercks briefing 10/20/59” (NARA Cartographic Records, RG 77, box 3 of 54, folder “1301 World 1:1,000,000— General”). See also short notices in Military Engineer 341 (1959): 223; and Military Engineer 351 (1961): 56. Series 1301 is not mentioned in Diercks, “The Role of the Topographic Map.” The Defense Mapping Agency (successor to both AMS and ACIC) discontinued the series in 1979. 111. For example, the transliteration of Soviet place-names was standardized by the US, not the USSR. Symbols could also vary from international standards, and maps were often overlaid with the UTM grid (see chap. 4). 112. Numbers from Arthur Robinson, “The Future of the International Map,” Cartographic Journal 2 (June 1965): 24. 113. For a list of official resolutions on the IMW from the IGU, the UN (both ECOSOC and various regional cartographic conferences), the Pan American Consultation on Cartography, the Scientific Council for Africa South of the Sahara, ICAO, and the International Association of Cartography, see Meynen, International Bibliography (see n. 28 above), 41– 49. 114. For delegates, see UN Department of Economic and Social Affairs, United Nations Technical Conference on the International Map of the World on the Millionth Scale, vol. 1 (New York: UN, 1963), 5– 9. For exhibition catalog, see [E. Meynen and H. Schamp], I M W / C. I. M. 1:1,000,000 Map Exhibition (Bad Godesberg: Bundesanstalt für Landeskunde und Raumforschung, 1962). 115. Karl-Heinz Meine, “Problems of Coordinating the IMW with the ICAO Series 1:1,000,000 and Prospective Considerations on Smaller Scales,” Informations Relative to Cartography and Geodesy, Series II: German Contributions in Foreign Languages 10 (1960): 68– 69. 116. UN Department of Economic and Social Affairs, United Nations Technical Conference, vol. 1, 122. 117. D. M. P. minute note, 4 Sept 1962 (PRO, OS 1/801, “International Map of the World”). The UK position is given in R. A. Gardiner, “A Re-appraisal of the International Map of the World (IMW) on the Millionth Scale,” Internationales Jahrbuch für Kartographie 1 (1961), originally delivered at the 1960 IGU meeting in Stockholm. 118. Ministère des Travaux Publics et des Transports, Carte Internationale du Monde 1/1 000 000: Étude des signes conventionnels (Paris: IGN, 1962), available in PRO, OS 1/800, “International Map of the World.” 119. UN Department of Economic and Social Affairs, United Nations Technical Conference, vol. 1, 21. See also G. R. Crone, “The Future of the International Million Map of the World,” Geographical Journal 128 (Mar 1962): 36– 38, who recognized this logic in the original 1913 specifications. In the 1910s, however, it had never been explicitly discussed. 120. UN Department of Economic and Social Affairs, United Nations Technical Conference, vol. 1, 63– 64. 121. First quote from H. Knorr, “Ideas Concerning the Coordination of the International Map of the World and the World Aeronautical Chart 1:1,000,000,” Informations Relative to Car-

Notes to Pages 107–111

333

tography and Geodesy, Series II: German Contributions in Foreign Languages 12 (1962): 10. Second from Crone, “The Future of the International Million Map of the World,” 38. 122. UN Department of Economic and Social Affairs, United Nations Technical Conference, vol. 1, 42. 123. The final specifications are in UN Department of Economic and Social Affairs, United Nations Technical Conference on the International Map of the World on the Millionth Scale, vol. 2 (New York: UN, 1963). See also O. M. Miller (of AGS, US representative at the conference), “Conference on Specification Revisions for the International Map of the World 1:1,000,000,” Geographical Review 53 (Jan 1963): 142– 145; “I. M. W. Conference, Bonn 1962,” Geographical Journal 128 (Dec 1962): 563– 564. 124. UN Department of Economic and Social Affairs, International Map of the World on the Millionth Scale: Report for . . . (New York: UN); these annual reports were published continuously until 1969, and then again in 1973 and 1977. Brief pamphlet updates were also issued in 1971, 1972, 1976, 1982, 1983, and 1987. For UK designs, see Report for 1963, 11; these were published in 1965. For Japan, see Report for 1964, 14; these were published between 1964 and 1966. For Australia, see Report for 1967; these sheets were published through 1978. For reviews, see Gardiner, “The International Map of the World: Australia’s Contribution” (n. 3 above), 162; D. M. Castle review, Geographical Journal 144 (Feb 1978): 520– 521. For Canada, see Report for 1969, 3– 7. Other countries’ post-1962 maps are shown on index diagrams and text tables in the various annual reports. 125. Report for 1973, 5. 126. Robinson, “The Future of the International Map,” 24. 127. Pearson et al., “Cartographic Ideals and Geopolitical Realities” (see n. 4 above), 173. See also the “Global Map Version 1.1 Specifications” (available from the International Steering Committee for Global Mapping, http://www.iscgm.org), which headlines the “Purpose and intended use of the Global Map.” 128. This is certainly the case today. Googling “base map” gives many websites— everywhere from ESRI to Schlumberger— that discuss demographic, geological, or infrastructure maps as “base maps.” Reference atlases, however, are a notably conservative exception to this trend. 129. See, for example, Gardiner on French maps of southeast England: R. A. Gardiner, “International Map of the World at 1:1,000,000,” Geographical Journal 137 (Sept 1971): 424. 130. See especially J. B. Harley, “Deconstructing the Map,” Cartographica 26 (Spring 1989): 1– 20. This theme has had a long life; see, for example, Daniel Clayton’s discussion of Graham Huggan’s idea of the “mimetic fallacy” in his “On the Colonial Genealogy of George Vancouver’s Chart of the Northwest Coast of North America,” Ecumene 7 (2000): 385. 131. David Turnbull, Maps Are Territories (Chicago: University of Chicago Press, 1993 [orig. 1989]), 10. Harley follows in 1991 with a conference paper, “The Map Is the Territory: Cartography and the Frontier in Eighteenth-Century North America,” referenced in Harley, The New Nature of Maps, 224. See also Denis Wood with John Fels, The Power of Maps (New York: Guilford Press, 1992), 10, with subhead “Maps Construct— Not Reproduce— the World.” 132. This is especially explicit in Wood and Fels, The Power of Maps, but it is also the central theme in J. B. Harley, “Maps, Knowledge, and Power,” in The New Nature of Maps. It suffuses Turnbull, Maps Are Territories, as well. 133. For Harley, see “Deconstructing the Map,” 162 (esp. n. 58). For Edney, see “Cartography without ‘Progress’: Reinterpreting the Nature and Historical Development of Mapmaking,” Cartographica 30 (Summer/Autumn 1993): 54– 68; Edney also critiques the idea that “the map replicates the territory’s structure precisely and accurately.” 134. Edney has done the most to frame the 1980s as a radical break; he calls it a “paradigm shift” from the “empiricist paradigm” to the “critical paradigm.” See Matthew Edney, “Recent Trends in the History of Cartography: A Selective, Annotated Bibliography to the EnglishLanguage Literature,” version 2.1, Coordinates B 6 (Apr 2007), http://purl.oclc.org/coordinates /b6.htm. At the time, Harley did acknowledge the influence of working cartographers on the

334

Notes to Pages 111–114

history of cartography; see his “The Map and the Development of the History of Cartography,” in The History of Cartography, vol. 1, ed. J. B. Harley and David Woodward (Chicago: University of Chicago Press, 1987), 33– 34. 135. See Matthew Edney, “Cartography’s ‘Scientific Reformation’ and the Study of Topographical Mapping in the Modern Era,” in History of Cartography: International Symposium of the ICA Commission, 2010, ed. Elri Liebenberg and Imre Josef Demhardt (Berlin: Springer, 2012), 297– 298. 136. This is made explicit in two recent articles by Denis Wood and John Krygier: “Critical Cartography” and “Protest Maps,” both in The Encyclopedia of Human Geography (New York: Elsevier, 2009).

Chapter 3 1. In addition to simple geometry, the soldiers will also need a few simple correction factors for weather, elevation, and their overall position on the grid. For a description of the inter-Allied coordinate system, see GSGS (attributed to E. M. Jack), Report on Survey on the Western Front, 1914– 1918 (London: HMSO, 1920), 161– 163. For the timeline and strategy of the Battle of Amiens, see James Edmonds, ed., Military Operations: France and Belgium, 1918, vol. 4 (London: HMSO, 1947), 23, 34, 41, 59. 2. W. H. Mills, “‘Location by Agreement,’” Military Engineer 40 (1948): 205. 3. This quote specifically references the US domestic grid system. David Greenhood, Mapping (Chicago: University of Chicago Press, 1964 [originally Down to Earth: Mapping for Everybody, 1944]), 21. 4. Sometimes grids are shown only as marginal ticks. The USGS in particular has changed its graphic treatment of grids several times. Ticks were replaced by squares in 1979; this was reversed in 1992, and then again in the late 1990s. See http://www.usgs.gov/faq/list_faq_by _category/get_answer.asp?id=845. Actual practice has not always matched policy; see Alden P. Colvocoresses, “The Gridded Map,” Photogrammetric Engineering & Remote Sensing 63 (Apr 1997): 378. 5. Grids likewise prompted both the UK and the US to change from imperial to metric units in surveying; for a good illustration of these connections, see “Ordnance Survey Maps: Discussion, with Special Reference to the Grid System, at the Afternoon Meeting of the Society, 9 December 1935,” Geographical Journal 87 (Apr 1936): 308– 327. 6. Le service géographique de l’armée: Son histoire, son organisation, ses travaux (Paris: SGA, 1938), 81. 7. Matthew Edney discusses Carroll and Borges, citing Umberto Eco as well, in Mapping an Empire: The Geographical Construction of British India, 1765– 1843 (Chicago: University of Chicago Press, 1997), 25– 28. 8. “Study and Discussion of Military Grids,” n.d. [late 1940s] (NARA Cartographic Records, RG 77, box 59 of 215, folder “Background Reference UTM Papers”), 8. 9. Emilien Cordonnier, from his 1911 book on the Russo-Japanese war, quoted in translation in Christopher Bellamy, “The Russian Artillery and the Origins of Indirect Fire,” Journal of Soviet Military Studies 3 (1990): 507. See also Boyd Dastrup, The Field Artillery: History and Sourcebook (Westport, CT: Greenwood Press, 1994); J. B. A. Bailey, Field Artillery and Firepower (Annapolis: Naval Institute Press, 2004); Mark Monmonier, Rhumb Lines and Map Wars: A Social History of the Mercator Projection (Chicago: University of Chicago Press, 2004), 98. 10. “Le Service Géographique de l’Armée et la cartographie de guerre,” La Géographie 32 (1919): 468– 472; Dastrup, The Field Artillery, 46– 47. 11. The British, for example, had tried to use pure map firing in the Second Boer War, without much success. Dastrup, The Field Artillery, 46– 47. 12. The lack of precise elevation contours was also a major problem. For discussion of the defects of the État-major, see Paul Vidal de la Blache, “La carte de France au 50000e,” Annales de Géographie 13 (1904); Emmanuel de Martonne, “Pour la carte de France au 50000e,” Annales de Géographie 33 (1924); Jean Martin, “Comment réaliser la nouvelle carte de France à 1:50000,”

Notes to Pages 115–126

335

Annales de Géographie 44 (1935). For the original rationale for the Bonne, see Col. Henry, “Mémoire sur la projection des cartes géographiques, adoptée au Dépôt Général de la Guerre,” Mémorial du Dépôt Général de la Guerre 2 (1803– 1805 and 1810): 433. 13. These new maps were largely derived from preexisting maps of French fortresses and the French tax cadastre (either from local archives or captured by the enemy), but they were also supplemented by new surveys and some of the earliest aerial photography. Aerial photography was used to map daily changes in the enemy’s trenches, but the old-fashioned methods were crucial for creating the basic cartographic framework. For survey work, see “Le Service Géographique de l’Armée et la cartographie de guerre,” 468– 472; Georges Perrier, “Geographic Service in the French Armies during the War, 1914– 1918,” Military Engineer 32 (July–Aug 1940): 266– 270. For introduction of grids by various armies, see Peter Chasseaud, “British Artillery and Trench Maps on the Western Front, 1914– 1918,” Map Collector 51 (1990); Peter Chasseaud, “German Maps and Survey on the Western Front, 1914– 1918,” Cartographic Journal 38 (Dec 2001). 14. Chasseaud, “German Maps and Survey,” 132. 15. For a history of the carte de France, see Josef Konvitz, Cartography in France, 1660– 1848 (Chicago: University of Chicago Press, 1987). 16. For example, Gabriel James Morrison, Maps, Their Uses and Construction (London: Edward Stanford, 1901), 87; Charles F. Close, Text Book of Topographical and Geographical Surveying (London: HMSO, 1905), 25; A. Courtier, “Choix d’un système de répresentation pour la rédaction des levés hydrographiques,” Annales hydrographiques 32 (1912): 46. 17. Within a radius of about ten kilometers, any errors that come from assuming the earth is flat amount to less than one centimeter; see Courtier, “Choix d’un système de répresentation,” 24– 25. Nineteenth-century textbooks explain rectangular coordinates as the preferred method to plot local surveys. See, for example, Robert Gibson et al., A Treatise on Practical Surveying (Baltimore, 1818), 224f; George André, The Draughtsman’s Handbook of Plan and Map Drawing (London, 1874); Charles Breed and George Hosmer, The Principles and Practice of Surveying, 5th ed. (New York: J. Wiley and Sons, 1927), 299, 410, 449. 18. Soldner quoted in Wilhelm Jordan (with revisions by Otto Eggert), Handbuch der Vermessungskunde, 8th ed., vol. 3, book 2, Sphäroidische Berechnungen, konforme Abbildung des Erdellipsoids, und Aufgaben der Erdmessung (Stuttgart: Metzler, 1941), 261. Puissant from his Supplément au deuxième livre du traité de topographie, contenant la théorie des projections des cartes (Paris: Courcier, 1810), 36. In the 1950s, Puissant was seen as the originator of rectangular coordinates: a South African survey official wrote, “As you no doubt know, it is to the French mathematician Puissant, that the idea of the use of plane rectangular co-ordinates in surveying is attributed.” See G. H. Halliday, “Surveying for Title Purposes in South Africa,” Surveying and Mapping 19 (Sept 1959): 341. 19. For England: J. B. Harley, Ordnance Survey Maps: A Descriptive Manual (Southampton: Ordnance Survey, 1975). For Germany: Jordan and Eggert, Handbuch der Vermessungskunde, 250– 271. For Switzerland: S. Malchow, “Cadastral Survey and Its Geodetic Foundation in Swiss Practice,” Surveying and Mapping 13 (June 1963). For Massachusetts: H. F. Walling, “Topographic Surveys of States,” Journal of the Association of Engineering Societies 5 (Mar 1886). For New York City: New York Board of Estimate and Apportionment, Report on the Triangulation of Greater New York (New York: M. B. Brown Press, 1909). For South Africa: Report of the Sixth International Geographical Congress, Held in London, 1895 (London: John Murray, 1896), 339. For Egypt and German Africa: Max Eckert, Die Kartenwissenschaft, vol. 1 (Berlin: Walter de Gruyter, 1921), 185. 20. The Germans did not invoke the name Cassini; they instead called them “Soldner coordinates.” 21. Lt. Earl F. Church, “Report on Geodetic Work in France,” 24 Dec 1918 (NARA, RG 120, entry 1780, box 115, “AEF Historical Report 1917– 1919”), app. I, 5. 22. There are several counts for the number of grid systems used. See GSGS/Jack (n. 1 above), Report on Survey, 151; Arthur R. Hinks, “German War Maps and Survey,” Geographical Journal 53 (Jan 1919): 33; Chasseaud, “German Maps and Survey,” 124– 126.

336

Notes to Pages 126–130

23. The British began the war by overprinting yard-based squares onto French and Belgian maps drawn in meters— the result, besides much general confusion, was that no continuous grid was possible. See GSGS/Jack, Report on Survey, 161– 164 (quote on 151); Chasseaud, “British Artillery and Trench Maps,” 24– 25. 24. Having no angular distortion also means that at every point on a conformal projection (known as orthomorphic in the UK), east-west scale is always the same as north-south scale. Note, however, that simply connecting two distant points with a straight line will not give the correct bearing, since straight lines on the projection are not great-circle paths. It is only local angles that are preserved. For conformalism in general, see Franciszek Biernacki, Theory of Representation of Surfaces for Surveyors and Cartographers, trans. from Polish (Warsaw, 1965 [orig. 1949]), 206– 213; John Snyder, Flattening the Earth: Two Thousand Years of Map Projections (Chicago: University of Chicago Press, 1997). 25. Courtier’s innovation was to analyze not just the deformations inherent in various projections, but also the deformation equations necessary to translate high-precision survey data directly into a rectangular coordinate system without ever using latitude and longitude at all. See Courtier, “Choix d’un système de répresentation.” Compare to P. Hatt, Des coordonnées rectangulaires et de leur emploi dans les calculs de triangulations (Paris: Imprimerie Nationale, 1893); Nicolas Auguste Tissot, Mémoire sur la représentation des surfaces et les projections des cartes géographiques (Paris: Gauthier-Villars, 1881). For a discussion of Courtier’s novelty, see Service Géographique de l’Armée, Rapport sur les travaux exécutés du 1er août 1914 au 31 décembre 1919 (Paris: SGA, 1936), 296. Those outside France tended to see Tissot’s work as rather insulated and incomprehensible: see Arthur R. Hinks, “On the Projection Adopted for the Allied Maps on the Western Front,” Geographical Journal 57 (June 1921); A. E. Young, “Conformal Map Projections,” Geographical Journal 76 (Oct 1930). For a summary of Courtier’s career, see “La vie et l’œuvre de l’ingénieur général André Courtier (1877– 1958),” Annales hydrographiques, ser. 4, vol. 12 (1963– 1964 [orig. 1959]): 1– 4. See also A. Courtier, “Exposé de la projection de Lambert,” Annales hydrographiques, ser. 3, vol. 17, for 1940– 1945 (1946 [orig. 1916]): 114. 26. Gauss’s work did not appear in print until it was published by Oskar Schreiber in 1866. See Eckert, Die Kartenwissenschaft, vol. 1, 183– 189. 27. Today the projection is known as the Lambert Conformal Conic, to distinguish it from Lambert’s other projections. For a detailed narrative of the decision and conversion of existing projections, see SGA, Rapport sur les travaux 1914– 1919, 32– 38. The final projection involved a slight modification of Courtier’s original math: instead of using a tangent cone, a scale reduction of 1/2,037 was applied to approximate a secant condition. Courtier’s equations were also limited to the third order, so they were not strictly conformal and did not agree with Lambert tables subsequently produced by the British or Americans. Later, other Lambert zones were created, again with slightly different math; système Lambert usually refers to the original grid alone. 28. This decision is perhaps related to the fact that the Gauss projection can only be extended north-south, not east-west like the Lambert. In comparison to the French, the German command was also much more administratively fragmented. See Chasseaud, “German Maps and Survey,” 126. 29. The British surprise attack during the Battle of Cambrai in November 1917 is seen as the first successful use of the technique, but credit is also given to Georg Bruchmüller’s attack on Riga in September 1917. See Wilfrid Miles, ed., Military Operations: France and Belgium, 1917: The Battle of Cambrai (London: HMSO, 1948), 10– 13, 16. 30. For the eastern front, see Oskar Albrecht, Das Kriegsvermessungswesen während des Weltkrieges 1914– 18 (Munich: Bayerische Akademie der Wissenschaften, 1969), 24. For Austria, see Jordan and Eggert, Handbuch der Vermessungskunde, 269. Russia did not use grids during World War I. 31. Church, “Report on Geodetic Work in France,” 6. For standardization among the Allies, see Perrier, “Geographic Service in the French Armies,” 266– 270; Peter Chasseaud, Ar-

Notes to Pages 132–134

337

tillery’s Astrologers: A History of British Survey & Mapping on the Western Front, 1914– 1918 (Lewes: Mapbooks, 1999), 438. 32. As Harold Winterbotham, a British survey official, said in 1919, “Before this war no European nation thought of the necessity of a large-scale map. All were content to think of topographical maps as fulfilling tactical requirements; and we were just as bad as any of the rest of them in that respect”—comment on Hinks, “German War Maps and Survey,” 41. Postwar comments on wartime map use also include many anecdotes of soldiers and officers being quite unable to read maps at all. Peter Chasseaud reports that the UK printed 34 million maps, the French 30 million, and the Germans 775 million (a number which includes maps of the eastern front as well): Chasseaud, “German Maps and Survey,” 120. 33. For a good overview of intersection, resection, and traverse techniques, see War Department, Orientation for Artillery, Technical Manual 44-225 (Washington DC: USGPO, 30 June 1944). For a more technical discussion, see G. T. McCaw, “Resection in Survey,” Geographical Journal 52 (Aug 1918). For the problems of locating a gun in particular, see John O’Keefe, “The Universal Transverse Mercator Grid and Projection,” Professional Geographer 4 (Sept 1952): 19. 34. Plotting boards are also known as battery boards. This description based on GSGS/Jack, Report on Survey, 89– 92. For German boards, see Chasseaud, “German Maps and Survey,” 124. 35. Rate of fire as recommended by Giraloma Pallotta, “The Accompaniment of Infantry in the Attack,” Field Artillery Journal 10 (July–Aug 1920): 445. This is less than the maximum rate of fire for most guns. 36. Brigadier General Dwight E. Aultman, “Maps and Map Firing,” Field Artillery Journal 10 (July–Aug 1920): 376– 377. 37. Ibid. 38. Le service géographique de l’armée (see n. 6 above), 81. The SGA was also concerned with the “Canevas d’ensemble” (the set of master points) and especially the “Canevas de tir” (the more detailed set of points directly relevant to artillery). See “Le Service Géographique de l’Armée et la cartographie de guerre” (n. 10 above), 467. 39. The Battle of Arras. See Chasseaud, “German Maps and Survey,” 129. 40. For the UK, see J. A. Steers, An Introduction to the Study of Map Projections, 13th ed. (London: University of London Press, 1962 [orig. 1927]), 211. For the US, see William Bowie and Oscar Adams, Grid System for Progressive Maps in the United States, US Coast and Geodetic Survey Special Publication 59 (Washington DC: USGPO, 1919). For Germany, see Jordan and Eggert, Handbuch der Vermessungskunde (n. 18 above), 269. For France, see Clifford Mugnier, “The French Republic,” Photogrammetric Engineering & Remote Sensing 67 (Jan 2001), Grids and Datums, 35. 41. Most of these are listed in Federal Board of Surveys and Maps, Committee on Control, Report of the Subcommittee on Plane Coordinates, 11 Feb 1936 (available at NOAA). For Egypt, see discussion of Arthur R. Hinks, “The Grid in Civil Use,” Geographical Journal 63 (June 1924): 505. 42. See n. 40. For a specific discussion of consolidation in England, see H. St. J. L. Winterbotham, The National Plans (London: HMSO, 1934), 36– 39. 43. Enthusiasm for multipurpose grids was widespread, with many geographers predicting a day when every street sign would have a grid reference. For England, including links to local administration, see Hinks, “The Grid in Civil Use”; H. St. J. L. Winterbotham, “The Use of the New Grid on Ordnance Survey Maps,” Geographical Journal 82 (July 1933). For dreams of general multipurposeness in France, see Service Géographique de l’Armée, La nouvelle carte de France (Paris: SGA, 1923), 53. In the United States, see “The Use of Geodetic Control for Boundary Surveys (a Symposium),” Geodetic Letter (US Coast and Geodetic Survey) 2, no. 5 (May 1935). 44. James Scott, Seeing like a State (New Haven: Yale University Press, 1998). 45. New Jersey was the first, in 1935. Twenty-two more states had followed by 1947; another dozen had done so by 1977. See W. O. Byrd, “The State Coordinate Systems,” Surveying and Mapping 7 (July–Dec 1947): 146; Joseph Dracup, “The New Adjustment of the North American Datum,” ACSM Bulletin 59 (Nov 1977): 27– 28. For Adams’s biography, see http://www .knoxhistory.org/authors/adams.html.

338

Notes to Pages 134–141

46. Each zone is limited to about 150 miles on its short side; this keeps scale errors below about one part in ten thousand, which was seen as the precision limit of typical surveys. Also note that all grid coordinates were tied to the preexisting national network of high-precision triangulation, meaning that there would not be misalignments across state lines. For technical descriptions of the system, see three US Coast and Geodetic Survey special publications (SP) from 1935, all by Oscar Adams and his assistant Charles Claire: Manual of Plane-Coordinate Computation (SP 193), Manual of Traverse Computation on the Lambert Grid (SP 194), and Manual of Traverse Computation on the Transverse Mercator Grid (SP 195). See also Hugh Mitchell and Lansing Simmons, The State Coordinate Systems (a Manual for Surveyors), SP 235 (1945); or the many SPs after 1950 that are devoted to each state system individually. 47. For particular use of the phrase “common language,” see Colonel Beverly Ober, “Use of Coordinates in Police Work,” Surveying and Mapping 4 (Apr 1944): 36. 48. These clearinghouses were not all of uniform design. In New Jersey it was lodged in the Department of Conservation and Development, in Pennsylvania it was part of Internal Affairs, and in New York it was in Public Works. See New York State Planning Council, A State System of Plane Coordinates (Albany, 1938), quote on 19. For the lack of agencies in 1935, see John S. Dodds, “State Surveying and Mapping Bureaus,” Civil Engineering 5 (Apr 1935): 250. 49. For spacing, see George F. Syme, “North Carolina’s Geodetic Control System,” Military Engineer 26 (Jan–Feb 1934): 20– 24. For the goal of easily available publications, see Kissam’s comments in “The Use of Geodetic Control.” For relief work, see H. W. Hemple, “Cooperation of the United States Coast and Geodetic Survey and the Works Progress Administration in the Conduct of Geodetic Control-Surveys,” Transactions, American Geophysical Union 18 (1937): 63– 66; Oscar Adams, “Plane Coordinate-Systems for Individual States,” Transactions, American Geophysical Union 15 (1934): 36– 38; Frank Kane Jr., “Geodetic Surveys Will Be Extended,” New York Times, 5 Aug 1934, E7; Philip Kissam, “New Jersey Adopts Plane-Coordinate System,” Civil Engineering 5 (Nov 1935); Federal Board of Surveys and Maps, Report of the Subcommittee. 50. Oscar Adams, “Development of State Grid Systems,” Civil Engineering 7 (Jan 1937): 33– 37; Philip Kissam, “The Utilization of Plane Coordinates in New Jersey,” Geodetic Letter 1 (Jan 1937): 18. 51. Reprinted in Surveying and Mapping 6 (Jan–Mar 1946): 53. 52. The engineer was George Syme. His first idea was to use five north-south systems, based on a C&GS special publication from 1921; Adams created a single Lambert zone instead. See George Syme, “Geodetic Control for North Carolina Highways,” Civil Engineering 2 (Mar 1932): 182; Syme, “North Carolina’s Geodetic Control System”; Adams, “Development of State Grid Systems.” 53. For Kissam’s activism, see letter from Kissam to Harold W. Dodds, “Summary of Experience,” 18 Apr 1935; and letter from Kissam to Kenneth H. Condit, 3 Dec 1940 (both in Mudd, Kissam Papers, Departmental Correspondence File). See also “Professor Kissam Directs Unemployed Engineers,” Princeton Herald, 29 Dec 1933; “To Press Mapping Plan,” New York Times, 10 Mar 1936, 5; “Louisiana Adopts Coordinate System,” Civil Engineering 14 (Nov 1944): 485. For New Jersey as second grid, see Adams, “Development of State Grid Systems.” 54. Kissam, “New Jersey Adopts Plane-Coordinate System,” 684. John O’Keefe, the designer of the grid system discussed in chapter 4, put it even more baldly: without a grid, “a given area is likely to be surveyed ad infinitum and ad nauseum by a horde of surveyors . . . without leaving any residue of established points behind. Each surveyor sets up his own little coordinate system (often incredibly bad) which he refuses to let anyone else use. If, on the other hand, the Government sets up a coordinate system, and maintains a file of control points, the whole thing changes.” John O’Keefe to Inis Vignes, 16 June 1947 (NARA Cartographic Records, RG 77, box 59 of 215, folder “Background Reference UTM Papers”), 2. 55. One of Kissam’s contemporaries complained that because of reliance on “timehonored method[s]” of local surveys, “surveying proper, unlike other related branches of engineering, has scarcely advanced a step in this country for at least one hundred years.” E. W. Albrecht, “First-Order Geodetic Surveys,” Civil Engineering 2 (June 1932): 385.

Notes to Pages 141–144

339

56. Quote from Washington Bureau of Surveys and Maps, Washington’s Extended Use of State Plane Coordinates (Olympia, 1957), 3. See also Robert G. Watts, “Simplicity of State Plane Coordinate System in Surveying,” Surveying and Mapping 15 (Dec 1965); John W. Martin, “The Use of Coordinates in Land Surveying,” Surveying and Mapping 28 (Sept 1968). 57. Federal Board of Surveys and Maps, Report of the Subcommittee, 20. 58. Philip Kissam, “The United States Coast and Geodetic Survey and the Property Owner,” Scientific Monthly 44 (Apr 1937): 366; emphasis in original. For other discussions of permanence, see Syme, “North Carolina’s Geodetic Control System”; J. A. C. Callan, “Local Control Surveys in Alabama,” Civil Engineering 5 (Dec 1935): 790– 793; Oscar J. Marshall and Myron T. Jones, “State Bureaus of Maps and Surveys Aid Planning of Engineering Projects,” Civil Engineering 17 (Apr 1947): 40– 43; Charles M. Dixon, “A Practical Application of a Geodetic Control Survey and State Coordinate System,” Surveying and Mapping 9 (Apr–June 1949); Ralph Moore Berry, “Compulsory Ties to State Plane Coordinates?,” Surveying and Mapping 14 (July–Sept 1954). 59. Kissam wanted all markers to be inspected once every four years. Philip Kissam, “Function of a State Surveying Board,” Civil Engineering 5 (July 1935): 434– 435. 60. D. Graham Burnett, Masters of All They Surveyed: Exploration, Geography, and a British El Dorado (Chicago: University of Chicago Press, 2000), 133. 61. William Emory, Report on the United States and Mexican Boundary Survey (Washington DC, 1857), esp. 38. 62. For these examples, see Philip Kissam, “Local Control Surveys in New Jersey,” Civil Engineering 6 (Mar 1936), 205; Philip Kissam, “An Illustration of the Utility of State PlaneCoordinate Systems,” Transactions of the American Geophysical Union 18 (1937): 93– 96. 63. German and Austrian scientists began to be invited as observers from 1930. See W. D. L[ambert], “The International Geodetic and Geophysical Union,” Science 73, no. 1879 (2 Jan 1931): 14– 15; Walter D. Lambert, “Report of the Association of Geodesy,” Science 85, no. 2196 (29 Jan 1937): 122– 123. For the politics of international scientific exchange more broadly, see Daniel Kevles, “‘Into Hostile Political Camps’: The Reorganization of International Science in World War I,” Isis 62 (1971): 47– 60. 64. William Bowie, “Some International Problems in Geodesy,” Bulletin géodésique 26 (Apr–June 1930): 33, 38. 65. More precisely, Roussilhe, like Courtier, followed Tissot quite closely, and the stereographic was Tissot’s third projection. As with his other two projections, Tissot was not the first inventor of the stereographic, but before Courtier it had been unclear how Tissot’s work related to previous projections. 66. Compared to Lambert or Gauss zones, which could not be more than roughly 400 kilometers wide, his stereographic could grid a circle 1,120 kilometers in diameter. 67. H. Roussilhe, “Emploi des coordonnées rectangulaires stéréographiques pour le calcul de la triangulation dans un rayon de 560 kilomètres autour de l’origine,” Travaux de l’Association internationale de géodésie 1 (Paris, 1923 [presented 1922]), 4. The four systems were the old carte de l’État major (Delambre/Plessis ellipsoid, Bonne projection), the nouvelle carte de France (Clarke ellipsoid with three Lambert grid zones), coastal mapping (with eleven local origins and two different kinds of rectangular coordinates), and the cadastre (since the 1895 revision by Charles Lallemand, based on five Gauss belts). 68. Although his map shows fourteen origin points, only eight tables would be necessary, since many origins share the same latitude. His system thus required fewer tables than the Lambert, and gave greater accuracy than the Gauss. See H. Roussilhe, “Rapport sur les procédés de calcul en coordonnées rectangulaires et sur les systèmes de représentation plane de l’ellipsoïde applicables à la triangulation,” Travaux de l’Association internationale de géodésie 4 (Paris, 1927 [presented 1924]), quote on 43 (emphasis in original), discussion of tables on 46– 47. For IMW tie-in, see “Détail des questions proposées pour l’inscription à l’ordre du jour de la section de géodésie,” in Union Géodésique et Géophysique Internationale, Première assemblée générale: Rome, mai 1922: section de géodésie (Toulouse: Édouard Privat, 1922), 99. 69. H. Roussilhe, “Rapport sur les procédés de calcul,” 5.

340

Notes to Pages 144–150

70. Comptes rendus of the committee on projections from the 1927 meeting in Prague, Bulletin géodésique 18 (Apr–June 1928): 258. 71. Canada would be better served by a series of large Lambert zones. H. Roussilhe, “Rapport général sur les systèmes de représentation plane de l’ellipsoïde terrestre,” Travaux de l’Association internationale de géodésie 6 (Paris, 1928 [presented 1927]), 10. 72. Report of the commission on projections, Bulletin géodésique 17 (Jan–Mar 1928): 53. 73. Comptes rendus of the commission on projections from the 1933 meeting in Lisbon, Bulletin géodésique 45 (Jan–Mar 1935): 52. 74. Note that in geodesy, “spheroid” and “ellipsoid” are used roughly interchangeably; I use the latter throughout for consistency. For the eighteenth-century controversy, see Mary Terrall, The Man Who Flattened the Earth: Maupertuis and the Sciences in the Enlightenment (Chicago: University of Chicago Press, 2002). For a detailed chronology of ellipsoid measurements, see Georg Straßer, Ellipsoidische Parameter der Erdfigur (1800– 1950) (Munich: Bayerische Akademie der Wissenschaften, 1957). In figure 3.16, pre-1957 values are from Straßer, with surveying use as reported in Clifford Mugnier’s monthly Grids and Datums columns in Photogrammetric Engineering & Remote Sensing (1998– 2013). Values since 1957 are from D. H. Maling, Coordinate Systems and Map Projections, 2nd ed. (Oxford: Pergamon, 1992), 11. (No new ellipsoids have been adopted since.) Note that some of Maling’s pre-1850 values are incorrect. Newton’s flattening from the Principia is actually 1/230. For confusion about early French ellipsoids, see Colonel Lamotte, “Réponses au questionnaire du 1er avril 1924,” Bulletin géodésique 8 (1925): 652– 654. 75. For example, see G. H. Darwin, “On the Precession of a Viscous Spheroid, and on the Remote History of the Earth,” Philosophical Transactions of the Royal Society of London 170 (1879). 76. “Comptes rendus de la première assemblée générale” (Rome, 1922), Bulletin géodésique 2 (Apr 1923): 51. 77. William Bowie, “Propositions des États-Unis,” Bulletin géodésique [no. 1] (May 1922): 95. 78. In classical geodesy, the combination of ellipsoid and origin point is called a “datum” (plural “datums”), and different datums may orient their ellipsoids differently. Thus the other main (nonaccidental) cause of misalignments is called the “deflection of the vertical,” which is a problem of local gravity variation at a datum origin, where the vertical axis of the main telescope does not point directly to the center of the earth. This effect was first discovered in surveys near the Himalayas in northern India. Many textbooks give overviews of this story; for example, William Lowrie, Fundamentals of Geophysics (Cambridge: Cambridge University Press, 1997), 308; Peter Dehlinger, Marine Gravity (Elsevier, 1978), 10. For the seven-hundredmeter figure, see Floyd Hough, “European First Order Triangulation and Its Adjustment,” Photogrammetric Engineering 15 (Mar 1949): 26– 32. 79. First quote from “Comptes rendus de la deuxième assemblée générale” (Madrid, 1924), Bulletin géodésique 7 (1925): 554. Second from Harold Winterbotham response to G. T. McCaw, “The Proposed Adoption of a Standard Figure of the Earth,” Geographical Journal 64 (Aug 1924): 136. (In context, “rivals” seems to suggest both scientific and nonscientifc competition.) 80. See Swiss comments in “Comptes rendus de la deuxième assemblée générale,” 545. There were also disagreements— both within and between countries— about whether an international standard should be purely arbitrary and convenient (using only round numbers, for example) or should attempt to capture final truth. A round-number standard was defeated by a vote of nineteen to seventeen. 81. For this reasoning, see McCaw, “The Proposed Adoption of a Standard Figure of the Earth.” See also the various national responses to an IUGG inquiry, collected in Bulletin géodésique 8 (1925). For Perrier’s endorsement, see “Comptes rendus de la deuxième assemblée générale,” 546. See also John Hayford, Geodesy: The Figure of the Earth and Isostasy from Measurements in the United States (Washington DC: USGPO, 1909); John Hayford, Geodesy: Supplementary Investigation in 1909 of the Figure of the Earth and Isostasy (Washington DC: USGPO, 1910). Naomi Oreskes describes the importance of Hayford’s work from the point of view of isostasy; see her The Rejection of Continental Drift (New York: Oxford University Press, 1999). It seems

Notes to Pages 150–155

341

that Hayford may also have been the first to use the word datum to describe what the French called les éléments de départ. 82. “Comptes rendus de la deuxième assemblée générale,” 554. 83. For quote, see “Comptes rendus de la deuxième assemblée générale,” 543. The tables were published in 1928. See Georges Perrier, “Note au sujet du calcul des tables de l’ellipsoïde de référence international,” Bulletin géodésique 18 (1928): 361– 369. 84. For Portugal, see “Enquête sur l’emploi de l’ellipsoïde de référence international,” Bulletin géodésique 61 (1939): 98. For the USSR, see Maling, Coordinate Systems and Map Projections. 85. The resulting datum— the North American Datum of 1927, or NAD 27— continued to use the Clarke 1866 ellipsoid. It would remain in place until the creation of NAD 83, a massive project which took most of the 1970s to compute. 86. Remarks of Kasper Weigel, Jean Maury, and Ilmari Bonsdorff, from the comptes rendus of the commission on the European recalculation from the 1933 meeting in Lisbon, 219– 220. Last quote from Otto Eggert in the comptes rendus of the commission on the European recalculation from the 1936 meeting in Edinburgh, Bulletin géodésique 61 (1939): 259. 87. One by Pierre Tardi, in the comptes rendus of the commission on the European recalculation from the 1933 meeting in Lisbon, 220– 221; and one by K. Weigel, “Jonction des réseaux de triangulation des grand continents,” Bulletin géodésique 38 (1933): 198– 209. 88. Comptes rendus of the commission on the European recalculation from the 1933 meeting in Lisbon, 220. It was likewise deemed impracticable to undertake any new observations to strengthen the net before recalculation. 89. Oscar Adams, The Bowie Method of Triangulation Adjustment, US Coast and Geodetic Survey Special Publication 159 (Washington DC: USGPO, 1930), 4: “from 8 to 12 mathematicians” completed 3,800 working days of calculations. See also O. S. Adams, “The Readjustment of the First-Order Triangulation in the Western Half of the United States,” Bulletin of the National Research Council 11 (Nov 1926): 51– 53; O. S. Adams, “Loop-Closures Resulting from the Readjustment of the First-Order Triangulation in the Western Part of the United States,” Bulletin of the National Research Council (July 1927): 62– 65. 90. N. E. Nörlund and J. de Graaff Hunter, “Note on the Adjustment of the European Network of Triangulation,” Bulletin géodésique 61 (1939): 261– 272. Although there were rival schemes by Karsten Wold and Otto Eggert, Nørlund and de Graaff Hunter had been appointed as the subcommittee to consider calculation methods. See Georges Perrier, “Étude d’une compensation d’ensemble du réseau européen,” Bulletin géodésique 53 (Jan–Mar 1937): 7. 91. Quote from the resolution. See comptes rendus of the commission on projections from the 1936 meeting in Edinburgh, Bulletin géodésique 61 (Jan–Mar 1939): 129. For Tardi on Perrier, see his remembrances in Bulletin géodésique 1 (1946): 18– 21. 92. Zones five degrees wide were installed during World War II, but disagreement still lingered. See comments from Martin Hotine in “Comptes rendus de l’assemblée générale d’Oslo, section des triangulations” (1948), Bulletin géodésique 11 (1949): 15. 93. The colonies included in the 1929 proposal of Harold St. John Loyd Winterbotham were Kenya, Uganda, North Rhodesia, Nyasaland, and Tanganyika. See Gerald McGrath, “‘From Hills to Hotine,’” Cartographic Journal 13 (June 1976): 9. The colonies would be connected by an arc of triangulation along the thirtieth meridian; see A. Macdonald, “Two Continents, One Meridian, Two Visionaries, One Goal,” Survey Review 35 (Jan 2000). 94. Colonial administrators saw six-degree belts as much too wide for cadastral purposes; South Africa, for example, used a system of two-degree belts that dated from the 1840s. McGrath, “‘From Hills to Hotine,’” 12. 95. Pierre Tardi, “Étude d’un système de projection de Gauss en fuseaux de 6° d’amplitude, pouvant s’appliquer à l’ensemble du continent africain,” annex to his “Rapport sur les projections,” Travaux de l’Association internationale de géodésie 14 (1938 [presented 1936]), 14. 96. Or for Tardi, forty grades. Tardi, “Étude d’un système de projection.” Practical equations applicable to the entire globe were not derived until 1945 by the British surveyor E. H. Thompson. More prominent in the published record from the 1940s is Laurence Patrick Lee

342

Notes to Pages 155–160

(of the New Zealand Land Survey Office), whose later work is now the general reference on the Transverse Mercator. Both Thompson and Lee published in the Empire Survey Review from the mid-1940s. See John Snyder, Map Projections: A Working Manual, US Geological Survey Professional Paper 1395 (Washington DC: USGPO, 1987), 48. The most thorough midcentury work on the Transverse Mercator is attributed to the Bulgarian Vladimir K. Khristov; he presented at IUGG meetings, but his work does not seem to have been widely known in English until later. See Maling, Coordinate Systems and Map Projections (n. 74 above); and John Snyder and Harry Steward, Bibliography of Map Projections, 2nd ed., US Geological Survey Bulletin 1856 (Washington DC: USGS, 1997). Full solutions are of mostly of academic interest; for an introduction to the more practical Transverse Mercator solutions by Gauss (1825), Oskar Schreiber (1880), Louis Krüger (1912), and Giovanni Boaga (1948), see Snyder, Flattening the Earth (n. 24 above), 160– 161. 97. Tardi, “Étude d’un système de projection,” 14– 15. 98. Although these cadastral coordinate systems were indeed too limited in extent to control extended topographic mapping and had largely preceded any major triangulation, it is notable that they were never even mentioned at the IUGG. 99. H. Roussilhe, “Rapport sur les projections,” Travaux de l’Association internationale de géodésie 8 (1931). Roussilhe himself chose the stereographic for Syria and Lebanon; see Adib Fares, “The Cadastral System in Lebanon Comparing to the Other International Systems” (paper presented at the 22nd International Congress of the Fédération Internationale des Géomètres, Washington DC, Apr 2002). 100. “Enquête sur l’emploi de l’ellipsoïde de référence international.” Except for Portugal, no country recalculated its existing network; they simply used the International Ellipsoid for new work. 101. For comment from General Bellot, see McGrath, “‘From Hills to Hotine,’” 9. For British complaints, see A. R. H[inks], review of Physics of the Earth— II: The Figure of the Earth, by the US National Research Council, in Geographical Journal 79 (Mar 1932): 239– 241. 102. Perrier, “Étude d’une compensation d’ensemble du réseau européen.”

Chapter 4 1. The main source for this map is a 1980 UN survey published in World Cartography 17 (1983): annex 6. Additional information comes from earlier reports published in the same journal in 1970 and 1976 (another survey published in 1990 found “practically no changes” in the 1980s); Alden P. Colvocoresses, “A Unified Plane Co-ordinate Reference System,” World Cartography 9 (1969); Clifford Mugnier’s monthly Grids and Datums columns in Photogrammetric Engineering & Remote Sensing (since 1998). There is conflicting information in these sources; I have chosen to include Bolivia, Iceland, Egypt, and the Philippines, but not Panama or South Africa. France, West Germany, and Luxembourg are included because of their NATO membership. Information on the Communist system comes mostly from Colvocoresses, with Mongolia and China from Mugnier. 2. Estimate from Clifford Mugnier, “The Basics of Classical Datums,” Photogrammetric Engineering & Remote Sensing 66 (Apr 2000): 368. 3. Spherical coordinates are known to have been used by Eratosthenes in the third century BCE. 4. Boyd Dastrup, The Field Artillery: History and Sourcebook (Westport, CT: Greenwood Press, 1994), 60– 61. 5. For coast artillery, see “Adoption of Military Characteristics for Local Plane Projections, Grid Systems, and Precise Surveys of Harbor Defense Areas,” 22 Jan 1943 (NARA Cartographic Records, RG 77 box 54 of 215, folder “Background Papers for Coordinates of the Military Grid System”). Grids on plan-position indicators are discussed in “Notes on the Graticule-Grid,” 15 Aug 1943 (NARA, RG 331, entry 268, box 75, folder “H.Q. A.E.A.F. Maps: Referencing and Reporting Methods”). The automatic reporting of range and bearing in grid values is discussed

Notes to Pages 160–167

343

in “Conversion Diagrams,” 18 Apr 1944 (NARA, RG 331, entry 268, box 75, folder “H.Q. A.E.A.F. Maps: Referencing and Reporting Methods”). 6. John O’Keefe, “The Universal Transverse Mercator Grid and Projection,” Professional Geographer 4 (Sept 1952): 19. In Vietnam, about 85 percent of fire would be unobserved. 7. See chap. 2, n. 5. 8. The usable width of a grid depends on the accuracy required; artillerists require less accuracy than engineers. Mathematically, Lambert zones can be somewhat wider than Transverse Mercator belts. They are not strictly oriented east-west, but rather along a great circle that intersects a chosen meridian at ninety degrees. Stereographic and Oblique Mercator projections were also used, but much less commonly. 9. For Africa, see Gerald McGrath, “‘From Hills to Hotine,’” Cartographic Journal 13 (June 1976): 16. These zones were only five degrees wide, not the six degrees of Tardi’s scheme or UTM. 10. For this logic as applied to Libya, see A. B. Clough, Maps and Survey (London: War Office, 1952), 77; for the Balkans, 112. 11. For a description of the system, see “World Polyconic Grid,” AMS Bulletin, no. 1 (Aug 1943): 1– 5. It was first designed in 1919; see William Bowie and Oscar Adams, Grid System for Progressive Maps in the United States, US Coast and Geodetic Survey Special Publication 59 (Washington DC: USGPO, 1919). See also Mark Monmonier, “Practical and Emblematic Roles of the American Polyconic Projection,” Wiener Schriften zur Geographie und Kartographie, Band 16 (Vienna, 2004), 93– 99. 12. At the edges of each zone, the map error was about three or four times the gun error. The polyconic was seen as unsuitable even before it was adopted, and the reasons for its adoption remain unclear. See Earl F. Church, “Report on Geodetic Work in France,” 21 Dec 1918 (NARA, RG 120, NM-92 entry 1780, box 115), 7. For postwar disparagement, see Jacob Skop, “The Evolution of Military Grids,” Military Engineer 43 (Jan–Feb 1951): 15– 18. 13. This is the same Loper-Hotine agreement mentioned in chapter 2; see Clough, Maps and Survey, 43– 48. 14. See maps published during the war by the Soviet General Staff (Генеральный Штаб Рабоче-Крестьянской Красной Армии), available online at Mapster (http://igrek.amzp.pl/). These show the grid extended into Poland, Hungary, Czechoslovakia, and eastern Germany, with first-edition (Первое издание) 1:200,000 maps published mostly during 1944 and 1945. See also US Department of the Army, Foreign Maps, Technical Manual 5-248 (June 1956), which describes wartime extension of the Russian Gauss-Krüger grid into the Baltic states. Parts of Iran, Iraq, and Turkey were also covered in a series “published primarily as coverage for the Soviet Union” (p. 44). 15. When applied to England, the boundaries were changed to avoid awkward disjunctions, but this led to accuracy errors of up to twice what was accepted by the Americans. See “Notes on German Map Grids, and German Maps of Great Britain,” 16 Jan 1943 (NARA, RG 331, SHAEF Map Survey Section Numeric File, box 240, folder “German Grid Systems— General”). For its extension to Slovakia, Hungary, Poland, Lithuania, and parts of Austria and Denmark, see map indexes from the Berkeley map library for the Mitteleuropa series, 1 Apr 1943, http://oskicat.berkeley.edu/record=b10890696~S1, and the Deutsche Heereskarte, http://oskicat.berkeley.edu/record=b10891302~S1. 16. For example, it was not used in place of French grids in occupied territories. “The German Army Grid,” 6 Mar 1945 (NARA, RG 331, SHAEF Map Survey Section Numeric File, box 240, folder “German Grid Systems— General”). For its use on the eastern front and conversion of preexisting grids, see indexes to the Osteuropa series, 28 Nov 1942, http://oskicat.berkeley .edu/record=b10890696~S1; and again the Deutsche Heereskarte. For incomplete conversion, see J. L. Cruickshank, “German Military Maps of UK & Ireland of World War II,” Sheetlines 69 (Apr 2004). For use in Turkey, see Department of the Army, Foreign Maps, 146. For similarities to the Russian grid, see Central Intelligence Agency, German Cartographic and Map Collecting Agencies: The Geodetic Bases of German Cartography, Mar 1948 (NARA, RG 319, “Reports and Messages, 1918– 1951, Central Intelligence Agency,” box 21).

344

Notes to Pages 167–171

17. “German Grid and Reference Systems, and German Operational Maps of Great Britain,” n.d. (NARA, RG 331, SHAEF Map Survey Section Numeric File, box 240, folder “German Grid Systems— General”), 4. 18. For the thirty-second goal, see Joint Chiefs of Staff, Joint Intelligence Group, minutes from Working Group on Position Referencing Procedures, 20 July 1953 (NARA Cartographic Records, RG 77, box 1 of 3, folder “Map Projections & Grids— 1948– 1955”), 12. 19. Clough, Maps and Survey, 379. 20. For the magnitude of the gaps, see Floyd Hough, “European First Order Triangulation and Its Adjustment,” Photogrammetric Engineering 15 (Mar 1949): 26– 32. Distortion procedures are discussed in GSGS Technical Instruction 50, “Definition of the 6th Meridian,” 31 Aug 1943 (NARA, RG 331, 6th Army Engineer Section Subject File, box 217, folder “Geodesy— General”). 21. Clough, Maps and Survey, 121– 125. 22. Letter from Bob Cloud (?) to Louis Wirak, 30 Nov 1944 (NARA, RG 331, 6th Army Engineer Section Subject File, box 217, folder “Geodesy— General”); letter from Commandant Batteux to Wirak, 18 Jan 1945 (NARA, RG 331, 6th Army Engineer Section Subject File, box 217, folder “Geodesy— General”). For conversions in southern Germany, see Clough, Maps and Survey, 425. 23. For distrust of discrepant lists, see memo from Louis Wirak, dated 28 Nov 1944 (NARA, RG 331, 6th Army Engineer Section Subject File, box 217, folder “Geodesy— General”). 24. In 1944, for example, AMS issued a new military grid table which superseded five older ones. The new one applied to all latitudes instead of just specific latitude ranges and gave grid values at a resolution of one arc minute instead of five. This meant that equally accurate results could be obtained without the use of double interpolation. See “New Table Replaces Five Five-Minute Intersection Tables,” AMS Bulletin, no. 9 (July 1944): 17. 25. See War Department, Orientation for Artillery, Technical Manual 44-225 (Washington DC: USGPO, 30 June 1944). The “thrust-line” method was adopted from German precedent; see “Notes on the German ‘Stosslinie’ System of Map Reference” (NARA, RG 331, SHAEF Map Survey Section, Numeric File, Nov 1943– July 1945, box 240, folder “German Grid Systems— General”). 26. The first is War Department, Orientation, Technical Manual 4-225 (Washington DC, 15 July 1941); the second is War Department, Orientation for Artillery. 27. Both systems are described in War Department, Orientation for Artillery. 28. See the correspondence in NARA, RG 165, “ABC Files,” box 130. 29. The first letter referred to an area five hundred thousand units on a side (either yards or meters, depending on the grid), and the second referred to a one-hundred-thousand-unit square. O’Keefe to Skop, “Use of the 30 Quadrangle in Conjunction with the Point Grids,” 28 Dec 1948 (NARA, RG 337, Army Field Forces HQs Engineer Section Decimal File, box 15). 30. Joint Intelligence Group minutes, 78. 31. For these dates, see Floyd W. Hough, “A Conformal and World-Wide Military Grid System,” Transactions of the American Society of Civil Engineers 121 (1956 [orig. 1954]): 634; William C. Hall, “Reference Systems for Close Air Support,” Military Engineer 45 (1953): 339; John B. Robison, “Military Grids: Theory, History, and Utilization,” in Papers on Cartography, 1958– 59, by US Army Map Service, Far East (available at NOAA), 2. For use in Korea, see Joint Intelligence Group minutes. 32. John O’Keefe, “The New Military Grid of the Department of the Army,” Surveying and Mapping 8 (Oct–Dec 1948); Hough, “A Conformal and World-Wide Military Grid System.” I have seen no mention of the Soviet system in English or in IUGG sources before the war. For biographical details on O’Keefe, see obituaries in EOS (Jan 2001), 55; and by Bernard Chovitz in the IAG Newsletter (Jan 2001). 33. With the American system, east-west location was pegged to the east coast of the US. 34. A basic introduction to the MGRS is given in AMS, Grids and Grid References, Technical Manual 36 (Washington DC, 1950). Note that the regional two-letter codes have changed over time; my examples use present-day designations. For all rectangular systems, x and y coordi-

Notes to Pages 171–178

345

nates are known as eastings and northings, respectively— with E and N the preferred variables. This terminology comes from rectangular coordinates in survey, and ultimately from the use of latitude and “departures” in seafaring; for an early example, see Robert Gibson et al., A Treatise on Practical Surveying (Baltimore, 1818), 224. Also note that the polar system is properly known as the Universal Polar Stereographic, since it uses a stereographic projection. The referencing system is simply an extension of UTM. 35. GEOREF— i.e., GEOgraphical REFerence— coordinates are a string of letters and numbers, such as DJMQ4651. The letters label a globally unique one-degree area, and the four digits break down this area into smaller units of latitude and longitude. See William E. Johnson, “The World Geographic Reference System,” Air University Quarterly Review 4 (Summer 1951). 36. Between seven hundred and two thousand kilometers, depending on latitude. 37. The air force complained that printing the grid on smaller-scale maps made them too cluttered for cockpit use in combat situations. The army was happy to concede this point, since navigational charts at 1:500,000 or 1:1,000,000 would not be used for joint operations requiring close air support for ground troops. See “Note by the Secretaries to the Joint Intelligence Committee on the Geographic Reference System (GEOREF),” 29 Oct 1951 (NARA Cartographic Records, RG 77, box 12 of 54, folder “Mapping & Geodetic Publication Files: Grids”). 38. John O’Keefe to Inis Vignes, 16 June 1947 (NARA Cartographic Records, RG 77, box 59 of 215, folder “Background Reference UTM Papers”), 4. 39. Original proposals for UTM in 1945 and 1946 instead designated one degree of overlap, identical to the overlap in the World Polyconic system. The switch to miles was an explicit reference to ground-level reality. See Joint Intelligence Group minutes (n. 18 above), 26. 40. Early 1950s speed estimate from the army’s artillery school. See Cyrus Finley and John O’Keefe, “Requirements for Reference Systems,” 6 Nov 1953 (NARA Cartographic Records, RG 77, box 77 of 109, folder “Map Projections and Grids”), 5. 41. In 1953, grid zone 32V was extended west by four degrees after Norway complained of its inconvenience; a few years later Norway managed to get several zones near Svalbard widened and the entire UTM system extended north from 80°N to 84°N. See H. E. Sewell, “Changes in Grid Limits,” 22 Oct 1957 (NARA Cartographic Records, RG 77, box 12 of 54, folder “Mapping & Geodetic Publication Files: Grids”). The extension north to 84° was made beginning in 1956 over much AMS opposition; see H. E. Sewell, “UTM Grid on 1:250,000 Greenland Maps,” 18 July 1956 (NARA Cartographic Records, RG 77, box 3 of 15, folder “Map Projections & Grids, 1956– 1964”). 42. O’Keefe, “The Universal Transverse Mercator,” 23. 43. The UTM projection minimizes distortion by reducing the scale of the map uniformly by a factor of 1/2,500. This ensures that no inadmissible errors will occur at latitudes higher than about 42°. (A factor of 3/500 was used for the polar areas, meaning that errors were acceptable below about 82.5°.) In contrast, the Soviet (and German) system used no scale factor at all, which meant that its errors exceeded 1/2,500 below 57°—the latitude of Scotland or Latvia. The original système Lambert used a scale reduction of 1/2,037, and State Plane zones use factors of 1/10,000 or less. See O’Keefe, “The Universal Transverse Mercator,” 23; or the Joint Intelligence Group minutes, 27. Note that the UTM scale factor, while historically justified because of its suitability for Europe, is not much different from the value that minimizes overall global distortion, roughly 1/2,370; see E. W. Grafarend, “The Optimal Universal Transverse Mercator Projection,” Manuscripta Geodætica 20 (1995). 44. The precise boundaries were largely determined by the history of triangulation; this is why Manchuria was grouped with Japan, and Tibet with India. See Joint Intelligence Group minutes, 24. Mathematically, using these different ellipsoids simply meant that the translation of latitude and longitude to grid coordinates used a different set of tables in each area. This is quite different from the datum consolidation discussed in the next section. 45. See John Davies, “Uncle Joe Knew Where You Lived,” pts. 1 and 2, Sheetlines 72 (Apr 2005); 73 (Aug 2005). 46. See Hough’s obituary in the ACSM Bulletin 52 (Feb 1976), 43.

346

Notes to Pages 179–185

47. O’Keefe’s full comment was that “UTM is made the vehicle for datum adjustments.” See his “Note on the Universal Transverse Mercator Grid,” 2 Feb 1953 (NARA Cartographic Records, RG 77, box 77 of 109, folder “Map Projections and Grids, 1950– 1953”). 48. AMS memos regarding negotiations with the British made this clear: “Our objective in the field of map grids is . . . to achieve the UTM grid application world wide.” From J. D. Abell, “Application of UTM Grid,” 3 May 1956 (NARA Cartographic Records, RG 77, box 12 of 54, folder “Mapping & Geodetic Publication Files”), 1. 49. See chapter 3. For relationship between Bowie and Hough, see William Bowie, Comparison of Old and New Triangulation in California, CGS Special Publication 151 (Washington DC: USGPO, 1928), 1, 8. Hough worked for the Coast and Geodetic Survey from 1919 to 1930. 50. For the capture of Gigas on 11 April 1945, see Hough, “European First Order Triangulation” (n. 20 above), 26– 32. For the capture of the archive on 17 April, see John Cloud, “American Cartographic Transformations during the Cold War,” Cartography and Geographic Information Science 29 (2002): 264, 266, 267. Hough was still recounting this story just before his death; see Steve Fehr, “Honoring the Makers of Maps for Various Achievements,” Washington Post, 4 Sept 1975, DC7, which also gives the size of the Hough Team. 51. Hough, “European First Order Triangulation,” 26– 32. 52. For visits, see Floyd Hough, “International Cooperation on a Geodetic Project,” Transactions, American Geophysical Union 32 (Feb 1951): 107. For quote, see Floyd Hough, “The Readjustment of European Triangulation,” Bulletin géodésique 2 (Oct 1946): 29– 34. 53. Hough’s final summary of the project is his “International Cooperation on a Geodetic Project.” For the 1947 conference, see the comptes rendus and various annexes in the Bulletin géodésique 7 (1948): 1– 93. Because of a lack of adequate astronomical data in western Europe, the Bowie method ultimately had to be replaced there with a brute-force calculation of two thousand simultaneous equations using electronic computers; see C. A. Whitten, “Progress of the European Triangulation Adjustment,” Transactions, American Geophysical Union 30 (Dec 1949): 882– 883; J. G. Ladd, “The European Grid Conversion Program,” Aug 1953 (NARA Cartographic Records, RG 77, box 59 of 215, folder “Background Reference UTM Papers”). For Hough’s efforts at NATO, see Irene Fischer, Geodesy? What’s That? (New York: iUniverse, 2005), 13. 54. Quote from Ladd, “The European Grid Conversion Program.” Statistics from Jacob Skop, “Summation of the UTM Grid Conversion Program in Europe,” May 1952 (NARA Cartographic Records, RG 77, box 26 of 109, folder “UTM Conversion Progress”). 55. For a partial timeline of UTM adoption through 1963, see Mike Nolan, “The Introduction of Universal Transverse Mercator (UTM) Grid on Military Maps: A Sixty Year Retrospect,” Sheetlines 96 (2013): 28– 29. 56. The Soviet ellipsoid block in figure 4.13 was shown reduced only to Japan and Manchuria by 1952; see AMS, Universal Transverse Mercator Grid, Technical Manual 19 (Washington DC, 1952). The work on the Soviet triangulation was part of Harry Liberman’s AMS Technical Report 14, “An Investigation of the Geoid in Europe and Asia” (1953), which was published, without Asia, as “An Investigation of the Geoid in Europe,” Bulletin géodésique 37 (1955). See Fischer, Geodesy?, 23, 28. 57. For purposeful UTM delays, see letters exchanged between the director of military survey for Britain and the AMS, late 1953 and early 1954 (NARA Cartographic Records, RG 77, box 77 of 109, folders “Map Projections and Grids, 1950– 1953” and “Map Projections and Grids, 1954”). For rollout of finished maps starting in 1956, see Nolan, “The Introduction of Universal Transverse Mercator,” 29. Parts of Syria, Iraq, and Iran were not complete until the early 1960s. 58. Special arrangements were made with nearly every country in Europe. See “Notes on Grid Conversion of European Maps,” 27 Sept 1950 (NARA Cartographic Records, RG 77, box 25 of 109, folder “Cartographic Publication Files (1951– 1952) Memos”). 59. However, AMS was also prepared to map wherever it saw fit. For worries of the US-UK agreement breaking down during the Korean War because of US mapping in British areas, see attachment to J. D. Abell, “Application of UTM Grid,” 3 May 1956 (NARA Cartographic Records, RG 77, box 12 of 54, folder “Mapping & Geodetic Publication Files”), 3– 4.

Notes to Pages 185–189

347

60. Floyd Hough, “The Universal Transverse Mercator Grid (with Particular Reference to Africa),” in Conference of Commonwealth Survey Officers, 1951: Report of Proceedings (London: HMSO, 1955), 66. 61. Floyd Hough, “Geodesy and the Universal Transverse Mercator System (with particular reference to the Caribbean, Central and South America)” (paper presented at PAIGH, Oct 1952; available at NOAA). Mugnier reports that Hough also gave a presentation on UTM at PAIGH in 1947, but his 1952 presentation began essentially from scratch. See Clifford Mugnier, “Ecuador,” Photogrammetric Engineering & Remote Sensing 65 (May 1999): 542. 62. H. H. Brazier comment on Hough, “The Universal Transverse Mercator Grid (with Particular Reference to Africa),” 69. 63. Hough, “The Universal Transverse Mercator Grid (with Particular Reference to Africa),” 63. Achieving cadastral accuracy without the use of correction equations would require belts two degrees wide, rather than six. With correction equations, all conformal grids could provide perfect accuracy, but it was found that the average surveyor usually ignored such corrections. 64. Floyd Hough, “Use of UTM on Civilian Maps,” 12 Apr 1955 (NARA Cartographic Records, RG 77, box 1 of 3, folder “Map Projections & Grids, 1948– 1955”). For an AMS map showing fifteen systems in Latin America; see Hough, “Geodesy and the Universal Transverse Mercator System (with Particular Reference to the Caribbean, Central and South America).” This had been a consistent policy; see John O’Keefe, “Projections for Military Maps and Grids,” 16 June 1947 (NARA Cartographic Records, RG 77, box 59 of 215, folder “Background Reference UTM Papers”). 65. Quotes from J. D. Abell, “Grids in Thailand and Malaya,” 21 Dec 1956 (NARA Cartographic Records, RG 77, box 3 of 15, folder “Map Projections & Grids, 1956– 1964”). For the British point of view, see G. C. Stubbs, “Malayan Map Projection and Military Grids,” in Conference of British Commonwealth Survey Officers 1947: Report of Proceedings (London: HMSO, 1951). The proposed projection was a Rectified Skew Orthomorphic; the United States also used a similar projection for the State Plane system in the Alaskan Panhandle. AMS itself had used similar projections; see D. H. Woodyard, “Laborde Grid Systems for Air Force Missile Test Center,” 18 July 1955 (NARA Cartographic Records, RG 77, box 1 of 3, folder “Map Projections & Grids, 1948– 1955”). 66. For example, Matthew Edney has argued that the British geodetic efforts in India only led to “cartographic anarchy,” and Josef Konvitz notes how Gaspard Riche de Prony’s dream of a scientifically coordinated cadastre in France dissolved in the face of less expensive, and less centralized, alternatives. See Matthew Edney, Mapping an Empire: The Geographical Construction of British India, 1765– 1843 (Chicago: University of Chicago Press, 1997); and Josef Konvitz, Cartography in France, 1660– 1848 (Chicago: University of Chicago Press, 1987). See also Matthew Edney, “Politics, Science, and Government Mapping Policy in the United States, 1800– 1925,” American Cartographer 13, no. 4 (1986): 295– 306; Jeffrey Stone, “Imperialism, Colonialism, and Cartography,” Transactions of the Institute of British Geographers 13 (1988). 67. Coordinate mismatch was unavoidable for Soviet areas, but it was also common in most of western Europe. See E. A. Early, “Spheroids Adopted or in Use by Various Countries,” 15 Apr 1955 (NARA Cartographic Records, RG 77, box 1 of 3, folder “Map Projections & Grids, 1948– 1955”). By the late 1960s, the only western European countries that used the International Ellipsoid for domestic purposes seem to have been Denmark, Italy, and Belgium; see “The Status of World Topographic Mapping,” World Cartography 10 (1970). 68. O’Keefe to Vignes, 16 June 1947 (see n. 38 above). 69. See H. F. Rainsford, “The African Arc of the 30th Meridian,” in Conference of Survey Officers, 1951: Report of Proceedings (London: HMSO, 1955); D. L. Mills, “The African Arc of the 30th Meridian: Completion of the Triangulation,” in Conference of Survey Officers, 1955: Report of Proceedings (London: HMSO, 1959). 70. Most countries joined in the late 1940s, but Mexico only in 1953 and Paraguay in 1962. Argentina, Uruguay, and Surinam were the only countries that did not take part. (The agree-

348

Notes to Pages 189–191

ment with Cuba lapsed after the revolution.) For details, see John W. Granicher, “The Inter American Geodetic Survey: Twenty-Five Years of Cooperation” (paper prepared at the Army War College, 27 Dec 1972; available at dtic.mil). 71. For Brazil, see letters between O’Keefe, Abell, and Dupuy, 15 Oct 1955 and 12 Jan 1956 (NARA Cartographic Records, RG 77, box 54 of 215, folder “Raydist Decca”). For Shell Oil, see Lawton to Baker, 21 Oct 1953 (NARA Cartographic Records, RG 77, box 77 of 109, folder “Map Projections & Grids, 1950– 1953”). 72. New computational techniques were central to John Hayford’s work before World War I, and geodetic computation had been an important form of white-collar work relief during the Great Depression. One of the largest projects undertaken by the WPA-sponsored Mathematical Tables Project— which employed more than two hundred (human) computers in a large warehouse on the west side of Manhattan— had been the calculation of tables for the World Polyconic Grid in 1941 and 1942. See David Grier, “Table Making for the Relief of Labour,” in The History of Mathematical Tables, ed. M. Campbell-Kelly et al. (Oxford: Oxford University Press, 2003). 73. See discussion of Hough’s article, and the final resolution, in Conference of Survey Officers, 1951: Report of Proceedings (London: HMSO, 1955), 251– 254. 74. See exchange of letters between Ladd and Brazier, 23 and 30 June 1953; letter of 18 Sept 1953 shows the calculation had been finished (NARA Cartographic Records, RG 77, box 77 of 109, folder “Map Projections & Grids, 1950– 1953”). 75. See Clifford Mugnier, “Kingdom of Thailand,” Photogrammetric Engineering & Remote Sensing 77 (Feb 2011): 111. As with other arrangements, the Coast and Geodetic Survey was subcontracted to perform the calculations. 76. When in 1970 President Nixon proposed scaling back the Inter-American Geodetic Survey, the president of Costa Rica wrote in protest, “I believe I speak for all Central America in saying we definitely need this type of cooperation,” calling it “one of the most successful programs of the United States Government in Latin America.” Similar letters were received from other countries. See Granicher, “The Inter American Geodetic Survey,” 16– 17. 77. First quote from Fischer, Geodesy?, 189. Second from Irene Fischer, “The Development of the South American Datum 1969,” Survey Review 20 (Oct 1970): 354. 78. Fischer and Bernard Chovitz collaborated on a new ellipsoid in early 1956; Fischer did her own calculations over the next few months, and the Hough Ellipsoid was first presented at the IUGG in 1957. For details, and for its use as the first world datum (the Hough Datum, also known as the Vanguard Datum), see Fischer, Geodesy?, 50. 79. For WGS and the air force approach, see Deborah Jean Warner, “Political Geodesy: The Army, the Air Force, and the World Geodetic System of 1960,” Annals of Science 59 (2002): 363– 389. Weikko Heiskanen was the major force here; see his “New Era of Geodesy,” Science 121 (14 Jan 1955); or his “The Practical Significance of the Geoid Determinations,” Geofisika pura e applicata 18 (1950): 99– 102. For a later, more theoretical recapitulation, see R. A. Hirvonen, “The Reformation of Geodesy,” Journal of Geophysical Research 66 (May 1961). Theoretical approaches to global gravity dated from the mid-nineteenth century, but empirical results were not feasible until the 1930s; see J. de Graaff Hunter, “The Figure of the Earth from Gravity Observations and the Precision Obtainable,” Philosophical Transactions of the Royal Society of London, ser. A, 234, no. 743 (15 July 1935): 377– 431. 80. These are all shown on the map of major world datums in Geodesy for the Layman, 5th ed. (Washington DC: Defense Mapping Agency, 1983), 36, and discussed in Fischer, Geodesy?, chapter 10. For more detail, see Irene Fischer, “A Continental Datum for Mapping and Engineering in South America,” Surveying and Mapping 34 (Dec 1974); and Clifford Mugnier’s discussions of the Blue Nile (Adindan), Cape/Arc, and Australian datums in his columns for all the relevant countries. The latter two systems were not primarily American projects, but the US did provide input. For the Australian spheroid for UTM, see the map included with the 1964 revision of US Department of the Army, Universal Transverse Mercator Grid, Technical Manual TM 5-241-8 (Washington DC: 1958).

Notes to Pages 191–195

349

81. Irene Fischer, F. Ray Shirley, and Sandra Todd, “Specific Practical Applications of Satellite Geodesy,” Surveying and Mapping 30 (Dec 1970): 551. 82. Australia was converted in the mid-1960s; Indonesia and Sudan would change in the 1970s. See the sources in n. 1. 83. Hotine’s comments on Hough’s “The Universal Transverse Mercator Grid (with Particular Reference to Africa)” (see n. 60 above), 67; Colvocoresses, “A Unified Plane Co-ordinate Reference System” (see n. 1 above), 12. 84. In favor of UTM was Frederick Doyle (of USGS); see his “Federal Mapping and a National Grid,” Surveying and Mapping 33 (Sept 1973). The proposal from the National Geodetic Survey (successor agency to the Coast and Geodetic Survey) was to replace the State Plane with twodegree Transverse Mercator belts; see Buford Meade (of National Geodetic Survey), “Coordinate Systems for Surveying and Mapping,” Surveying and Mapping 33 (Sept 1973); William Pryor (of USDOT), “Plane Coordinates for Engineering and Cadastral Surveys,” Surveying and Mapping 33 (Sept 1973). The eventual solution is discussed in Joseph Dracup (of National Geodetic Survey), “The New Adjustment of the North American Datum,” ACSM Bulletin 59 (Nov 1977): 27– 28. Since 2001 this system has been known as the United States National Grid (USNG), but it remains identical with UTM; see http://www.fgdc.gov/standards/projects/FGDC-standards-projects/usng. 85. Namely, the Baltic countries, Moldova, Mongolia, and Albania; see Mugnier’s monthly columns. 86. First quote— with emphasis in original— from Charles E. Dills, “Coordinate Location of Archaeological Sites,” American Antiquity 35 (July 1970): 389. Second from Robert L. Edwards, “Archaeological Use of the Universal Transverse Mercator Grid,” American Antiquity 34 (Apr 1969): 182. Full-text searching reveals dozens of examples of actual archaeological use of UTM; one of the earliest is Carl H. Strandberg and Ray Tomlinson, “Photoarchaeological Analysis of Potomac River Fish Traps,” American Antiquity 34 (July 1969). 87. Rodney L. Crawford, “Grid Systems for Recording Specimen Collection Localities in North America,” Systematic Zoology 32 (Dec 1983), prefers the graticule but discusses contemporary use of UTM for specimen location, especially in Europe. And again, full-text searching reveals many more examples from the 1970s, in journals such as Science, Population Index, the American Journal of Botany, Wildlife Society Bulletin, and the Journal of Paleontology. The British National Grid had been used for similar purposes since at least 1949. 88. Annex I of Treaty of Peace between the Hashemite Kingdom of Jordan and the State of Israel, 26 Oct 1994, http://www.kinghussein.gov.jo/peacetreaty.html; Permanent Court of Arbitration, Eritrea-Ethiopia Boundary Commission, Annex to the Commission’s Statement of 27 November 2006, List of Boundary Points and Coordinates, http://www.pca-cpa.org/upload/files/Statement %20Annex%20271106.pdf. 89. For example, rectangular coordinate systems were used to stabilize the Missouri-Iowa border in the late nineteenth century, the Belgium-Germany border after World War I, and the US-Mexico border in 1970. The last used Texas State Plane coordinates. See, respectively, testimony of W. C. Hodgkins (of C&GS) in State of Missouri v. State of Iowa, United States Supreme Court Reports 165 (1896 term), 660; A. R. H[inks], “The Belgian-German Boundary Demarcation,” Geographical Journal 57 (Jan 1921): 49; Treaty to Resolve Pending Boundary Differences and Maintain the Rio Grande and Colorado River as the International Boundary, 23 Nov 1970, available at http://www.ibwc.state.gov/Files/1970Treaty.pdf. 90. See Dennis Rushworth, “Mapping in Support of Frontier Arbitration: Coordinates,” IBRU Boundary and Security Bulletin 5 (Autumn 1997): 59. See also Haim Srebro (Director of the Israel Survey) and Maxim Shoshani, “The Order of Precedence of Boundary Delimitations” (paper presented at the 2007 General Assembly of the International Federation of Surveyors, http://www.fig.net/pub/fig2007/techprog.htm). 91. First is Don Bartlett, “A Practical Guide to GPS— UTM,” first posted to rec.backcountry 10 Feb 1995, available at: http://www.dbartlett.com. Second is Casanova’s Outdoor Adventure Store, at http://casanovasadventures.com/catalog/gps/p3034.htm; archive.org indicates that the quote is from 2003.

350

Notes to Pages 195–197

92. Google Earth and Wikipedia, as well as most GPS receivers, do not use the regional two-letter codes of the Military Grid Reference System; coordinates are simply measured in meters from the equator and (roughly) the left edge of the grid zone, such as “5306468 325687 49G.” This reversal of northings and eastings is also the opposite of military convention. For Google Earth, see the release notes to version 4.0, summarized at http://ogleearth.com/2006 /09/new-google-earth-beta-out-supports-wms-time-tags-utm-grid. For Wikipedia, see the revision history for the “geohack” widget. 93. As early as the mid-1950s, the cost of changing the system even slightly amounted to a significant fraction of the initial cost of the entire project. AMS estimated the cost of updating various training manuals to reflect the changes near Norway at $25,000, while the estimated cost of “conversion of points and correction of drafting copy” for all of Europe had been $103,000. Neither estimate included the cost of printing new maps. 94. This system goes by several names. In Russian it is mostly just known as the GaussKrüger coordinate system or projection (проекция Гаусса-Крюгера). In the US it is usually known as the Soviet Unified Reference System, but Mugnier calls it the “Russia Belts” (cf. figure 4.3). For details, see Conversion of Warsaw Pact Grids to UTM Grids (Washington DC: Department of the Army, 1981); Colvocoresses, “A Unified Plane Co-ordinate Reference System”; D. H. Maling, Coordinate Systems and Map Projections, 2nd ed. (Oxford: Pergamon Press, 1992); Mugnier’s monthly columns. The Soviets also engaged in large-scale datum consolidation, expanding their SK-42 datum (Система координат 1942 года; known as Pulkovo 1942 in the West) at least throughout Eurasia. See Ю. А. Комаровский, Использование различных

референц-эллипсоидов в судовождении, 2nd ed. (Vladivostok, 2005), 248– 249. 95. The notable exceptions are the Oblique Mercator grid for Malaya designed in the 1950s and the stereographic system for New Brunswick. The most expansive rethinking of the problem is Colvocoresses, “A Unified Plane Co-ordinate Reference System,” which is based on his 1965 dissertation at Ohio State. Grids were also much discussed in Canada in the early 1960s; see J. E. Lilly, “New Brunswick Plane Co-ordinates,” Canadian Surveyor 14 (Oct 1959); W. V. Blackie, “Co-ordinate Control for Legal Surveys,” Canadian Surveyor 14 (Oct 1959); W. V. Blackie, “Legal Surveys for Oil and Gas Rights in the Yukon and Northwest Territories,” Canadian Surveyor 15 (Nov 1961); J. Saastamoinen, “A Canadian Grid System in 3° Transverse Mercator Zones,” Canadian Surveyor 18 (June 1964). See also the discussion of replacements for the State Plane system, n. 84 above. For a more recent version, see Grafarend, “The Optimal Universal Transverse Mercator Projection” (n. 43 above). 96. “Air Force Analysis of the Universal Transverse Mercator Grid,” and “Geographic Reference System (GEOREF),” both n.d. [late 1940s] (NARA Cartographic Records, RG 77, box 59 of 215, folder “Background Reference UTM Papers”). 97. See, for example, the Boy Scout Handbook; or Dava Sobel, Longitude (New York: Walker, 1995). But note that even nineteenth-century navigators were using a governmental system: their telescope and chronometer had to be used together with an up-to-date ephemeris. 98. John O’Keefe, “Graticule Mesh,” n.d. [late 1940s] (NARA Cartographic Records, RG 77, box 59 of 215, folder “Background Reference UTM Papers”); strikethrough in original. The scale error of the graticule reaches the UTM design requirement of 1/2,500 roughly 180 kilometers from the equator. 99. Quote from B. Chovitz, “A General Theory of Map Projections,” Bollettino di Geodesia e Scienze Affini 38 (1979). Both O’Keefe and Chovitz first articulated this stance in a series of technical articles in the early 1950s: John O’Keefe, “Some Applications of Tensor Analysis to Geodesy,” Transactions of the American Geophysical Union 32 (Apr 1951); B. Chovitz, “Classification of Map Projections in Terms of the Metric Tensor to the Second Order,” Bollettino di Geodesia e Scienze Affini 11 (1952); B. Chovitz, “Some Applications of the Classification of Map Projections in Terms of the Metric Tensor to the Second Order,” Bollettino di Geodesia e Scienze Affini 13 (1954). 100. Precise values of latitude and longitude are just as historically specific and unreproducible as grid coordinates, since local gravity fluctuations (i.e., deflections of the vertical) and

Notes to Pages 197–200

351

the least-squares adjustment of triangulation together assure that any coordinates found by local astronomical methods will differ from the official coordinates shown on the map. Even the coordinates of the Greenwich observatory are unstable over time: after the advent of satellite methods, the line of 0° longitude shifted about one hundred meters east of the main telescope. These issues often arise in critiques of boundary treaties; see Rushworth, “Mapping in Support of Frontier Arbitration,” or Hasanuddin Abidin et al., “Geodetic Datum of Indonesian Maritime Boundaries: Status and Problems” (paper presented at the 2005 General Assembly of the International Federation of Surveyors).

Chapter 5 1. For RAF sorties, see campaign diaries (published 2002) at http://webarchive.national archives.gov.uk/20070706011932/http://www.raf.mod.uk/bombercommand/oct44.html. 2. For an overview of Gee, see R. J. Dippy, “Gee: A Radio Navigational Aid,” Journal of the Institution of Electrical Engineers— Part IIIA: Radiolocation 93, no. 2 (1946). On interpretation systems, see “Minutes of a Meeting Held in Air Ministry, Abbey House on Thursday, 25th May, 1944 to Agree Action Required to Provide Interpretation Systems for Use with Stations Type 7000 on the Continent” (NARA, RG 331, entry 268, box 75, folder “Lattice Charts Production— Policy”). For Robert Watson-Watt, see his The Pulse of Radar (New York: Dial Press, 1959), 338. For G meaning grid, see Robert I. Colin, “Robert J. Dippy: The Hyperbolic Radio Navigation System,” IEEE Transactions on Aerospace and Electronic Systems 2 (July 1966): 476. 3. Elmer M. Lipsey (of Sperry Gyroscope Co.), “Tactical Air Navigation,” Navigation (US) 16 (Fall 1969 [orig. 1965]): 296, describing Loran-D. For “common grid,” see, for example, Keith D. McDonald, “A Survey of Satellite-Based Systems for Navigation, Position Surveillance, Traffic Control, and Collision Avoidance,” Navigation (US) 20 (Winter 1973– 1974): 306; or Bradford W. Parkinson interview by Steven R. Strom, Aerospace Headquarters, 1 Apr and 4 June 2003 (transcript at http://www.aero.orgcorporation/parkinson.html), 12. 4. These are postwar terms, evidently of British coinage. The earliest uses I have found for each are in Charles B. Bovill, “A Review of Radio Aids in Aviation,” Journal of the British Institution of Radio Engineers 6, no. 6 (1946): 257, and Dippy, “Gee,” 470. “Area coverage” was still being scare-quoted in the mid-1950s; see “Airline Navigation: A Summary of Facilities and Procedures,” Flight, 11 Mar 1955, 333. Track guide systems are also known as point-source systems, and most are also theta or rho-theta systems, where theta refers to bearing (to or from a point source) and rho is distance. Other variants are also possible, such as rho-rho or even rho-rhorho. (The latter evidently does not require navigating gently down a stream.) 5. The acronym RDF is likewise used, but since this was also used as an intentionally misleading codename for early British radar efforts, I avoid it here; see Louis Brown, A Radar History of World War II (Philadelphia: Institute of Physics, 1999), 83. 6. Quote from Paul Henderson (general manager of National Air Transport, Inc.), “Airways and Airdromes,” Engineers and Engineering 43 (May 1926): 139; he also compares airways to highways and navigable rivers, both of which require investment and maintenance. For lists and discussion of the facilities needed on an airway, see Frederick H. Sykes, “Imperial Air Routes,” Geographical Journal 55 (Apr 1920): 249; Dennis H. Handover, “A New Empire Link: West African Colonies Linked with Empire’s Air System,” Journal of the Royal African Society 35 (Oct 1936): 414. 7. In particular, the French government paid for railroad infrastructure, while private companies supplied the superstructure. This dichotomy was first used by railroad engineers in the 1860s and had acquired its political-economic connotation by the 1870s. It was applied to aviation in the 1920s. See, for example, Congrès des Transports Aériens, Rapports et discussions, 29 Nov–2 Dec, 1934 (Paris, 1934), especially the reports by Laignier, Bregi, and Alessandri, where the financial analogy with railroads is explicit. 8. The “transcontinental” in the US was initiated by the post office in 1919; by 1941 there were five transcontinentals in the US and one in Canada. See US Office of Assistant Secretary

352

Notes to Pages 205–211

for Aeronautics, “Civil Aeronautics in America,” Information Bulletin 1, 5th ed., 1 Oct 1927 (Harvard University, Loeb Library, Vertical Files NAC 2795 US), 5; Civil Aeronautics Administration press release, “New Airways Set-Up Groups Canadian and American Facilities,” 1 June 1941 (NARA, RG 237, box 444, “Central Files, 935.1 Australia— 940 Canada,” folder 940). For Africa, see Sykes, “Imperial Air Routes,” 246; Robert Brenard, “The Romance of the Air Mail to East and South Africa,” Journal of the Royal African Society 38 (Jan 1939): 47. For the Americas, see George E. Sanford, “The Intercontinental Airways System,” typescript for a presentation made 20 Sept 1926 (NARA, RG 237, box 445, “Central Files, 940 Central America— 943 Canada,” folder 940); William A. M. Burden, The Struggle for Airways in Latin America (New York: Council on Foreign Relations, 1943), 188. 9. The use of mileage for comparative purposes— either miles of airways or miles flown— was ubiquitous. For some particularly bald examples, see Stephen B. Sweeney, “Some Economic Aspects of Aircraft Transportation,” Annals of the American Academy of Political and Social Science 131 (May 1927): 161; P. R. C. Groves, “The Influence of Aviation on International Affairs,” Journal of the Royal Institute of International Affairs 8 (July 1929): 289; Russell E. Hall, “Expanding Airways in the Far East,” Far Eastern Survey 6 (28 Apr 1937): 95– 101; Melvin Hall and Walter Peck, “Wings for the Trojan Horse,” Foreign Affairs 19 (Jan 1941): 347– 369. 10. “On the beam” evidently became a common phrase outside aviation as well; see J. M. Ramsden, “Air Navigation,” Flight International, 18 Sept 1975, 407. 11. For the origin of the name, see Ronald Keen, Wireless Direction Finding, 3rd ed. (London: Iliffe and Sons, 1938), 476. For the railway analogy, see C. H. McIntosh (pilot instructor for American Airlines), Radio Range Flying (Chicago: Ringer, 1941), 17. 12. Substantially the same invention was described in German patents by Otto Scheller from 1907 and 1916 (and additionally by Franz Kiebitz in 1911), but none of the developers of the Radio Range knew of these until about 1926. See Robert I. Colin, “Otto Scheller and the Invention and Applications of the Radio-Range Principle,” Electrical Communication 40, no. 3 (1965): 365. For the system’s early history, see J. H. Dellinger, H. Diamond, and F. W. Dunmore, “Development of the Visual-Type Airway Radiobeacon System,” Proceedings of the Institute of Radio Engineers 18 (May 1930): 799– 802. 13. Quote from H. Dellinger and Haraden Pratt, “Development of Radio Aids to Air Navigation,” Proceedings of the Institute of Radio Engineers 16 (July 1928): 894. For explicit comparison with railroad practice, see “Air Lines to Span Nation,” Science News-Letter 12 (2 July 1927): 13. 14. See H. J. Walls, “The Civil Airways and Their Radio Facilities,” Proceedings of the Institute of Radio Engineers 17 (Dec 1929); F. G. Kear and W. E. Jackson, “Applying the Radio Range to the Airways,” Proceedings of the Institute of Radio Engineers 17 (Dec 1929); H. Diamond, “Applying the Visual Double-Modulation Type Radio Range to the Airways,” Proceedings of the Institute of Radio Engineers 17 (Dec 1929); Dellinger, Diamond, and Dunmore, “Development of the Visual-Type Airway Radiobeacon System”; W. E. Jackson and S. L. Bailey, “The Development of a Visual Type of Radio Range Transmitter Having a Universal Application to the Airways,” Proceedings of the Institute of Radio Engineers 18 (Dec 1930). 15. R. V. Jones, “Navigation and War,” Journal of Navigation (UK) 28 (Jan 1975): 5; “Radio Ranges in America,” Flight, 16 May 1940, 463; Museum of Air Traffic Control, “Four-Course Radio Range,” http://www.atcmuseum.org/navigation/nav_sis/nav_sis_lfr.asp (captured by archive.org 17 July 2010). 16. Roger F. Williams, “Federal Legislation Concerning Civil Aeronautics,” University of Pennsylvania Law Review and American Law Register 76 (May 1928): 803, which also discusses foreign air law. Cf. John C. Cooper, “Air Transport and World Organization,” Yale Law Journal 55 (Aug 1946): 1196– 1197. 17. Arthur K. Kuhn, “The Beginnings of an Aërial Law,” American Journal of International Law 4 (Jan 1910); Arthur K. Kuhn, “International Aerial Navigation and the Peace Conference,” American Journal of International Law 14 (July 1920); S. W. Buxton, “Freedom of Transit in the Air: The Present Position and How It Has Been Reached,” Economica 16 (Mar 1926); H. Burchall, “The Politics of International Air Routes,” International Affairs 14 (Jan–Feb 1935):

Notes to Pages 211–215

353

98– 99; John C. Cooper, “Some Historic Phases of British International Civil Aviation Policy,” International Affairs 23 (Apr 1947); Cooper, “Air Transport and World Organization,” 1197; D. Goedhuis, “Sovereignty and Freedom in the Air Space,” Transactions of the Grotius Society 41 (1955): 137. For a broader legal discussion, see Stuart Banner, Who Owns the Sky? (Cambridge, MA: Harvard University Press, 2008). 18. Women were explicitly prohibited from serving on the crew of any aircraft engaged in public transport; see ICAN, Convention Relating to the Regulation of Aerial Navigation Dated 13th October 1919 (Paris, 1925), 28. 19. There were different reporting norms in different countries. In the UK and France, for example, locations were usually reported using nearby town names, while elsewhere latitude and longitude were used. See ICAN Maps Sub-commission, “Minutes No. 18: Sittings of 22nd November 1938” (typescript copy available in vitrine of the ICAO Legal Bureau, Montreal), 7– 8. Ground D/F could also be used to “talk down” a pilot to a landing field using only bearings rather than point locations; see Keen, Wireless Direction Finding, 616. 20. For problems with interference from insufficiently screened engines, see Brian Kendal, “Air Navigation Systems Chapter 3: The Beginnings of Directional Radio Techniques for Air Navigation, 1910– 1940,” Journal of Navigation (UK) 43 (Sept 1990): 325. For the navigational problems involved (and the need for a good compass), see Robert I. Colin, “Survey of Radio Navigational Aids,” Electrical Communication 24 (June 1947): 221– 225. 21. The tension between pilot autonomy and air traffic efficiency in the US was largely a question of competing technological systems; see Erik Conway, “The Politics of Blind Landing,” Technology and Culture 42 (2001): 81– 106. In Europe, however, D/F could enable either autonomy or control, depending on what information was transmitted to the pilot (see n. 19). For use of D/F in traffic control, see H. A. Taylor, “Radio and Air Traffic,” Flight, 30 Jan 1936, 120– 122. For a lament about the “problem of the itinerant aircraft not flying over a regular route,” see Roderick Denman, “Radio Air Navigation: The Trend of British Development,” Flight, 21 Jan 1937, 54– 56. Yet Denman’s proposed solution, which he called “fanciful,” was not a system of rigid routes but a “radio grid” of stable coordinates! 22. The physics of directional radio had been investigated since the 1890s by Hertz, Marconi, and others. The first patents specifically for direction finding came just after those for directional transmission, in the early 1900s. See Keen, Wireless Direction Finding, 6– 10. For use in the 1910s, see Ken Beauchamp, History of Telegraphy (London: Institution of Electrical Engineers, 2001), 243, 269– 272, 315, 324– 326. 23. For details of using the “Position Line Method” at sea (used since the mid-nineteenth century), including both ship and ground D/F, see Edward J. Willis, The Methods of Modern Navigation (New York: D. Van Nostrand, 1925). For maps and D/F, see Keen, Wireless Direction Finding, chap. 8. 24. For 1920s stations, see Kendal, “Beginnings of Directional Radio,” 324; C. Powell, “Radio Navigation in the 1920s,” Journal of the Institution of Electronic and Radio Engineers 56 (Aug– Sept 1986): 297; Gerald C. Gross, “European Aviation Radio,” Proceedings of the Institute of Radio Engineers 19 (Mar 1931): 346. For expansion into 1930s, see “Short-Wave Direction Finding,” Flight, 8 July 1937, 61; Keen, Wireless Direction Finding, 549, 577. For coastal D/F provided by the US Lighthouse Service and the US Navy, see George R. Putnam, “Radio Fog Signals for the Protection of Navigation: Recent Progress,” Proceedings of the National Academy of Sciences of the United States of America 10 (15 June 1924); Dellinger and Pratt, “Development of Radio Aids,” 892. For Canada, see Harold S. Patton, “Canada’s Advance to Hudson Bay,” Economic Geography 5 (July 1929): 233. 25. For requirement for carrying wireless apparatus, see ICAN, Official Bulletin 20 (Oct 1932): 58; ICAN, Official Bulletin 24 (Dec 1936): 148– 149. Likewise, the Regulations for the International Radioelectric Service of Air Navigation (Paris: ICAN, July 1934) are almost entirely devoted to communications. For standardization of frequencies in 1934, see Kendal, “Beginnings of Directional Radio,” 324. For standardized beacon services (again modeled on marine precedent), see Denman, “Radio Air Navigation,” 54. ICAN mapping was discussed for several

354

Notes to Pages 215–216

years, especially in response to increasing congestion; see “I.C.A.N., 1937,” Flight, 8 July 1937, 61. As with the IMW derivatives discussed in chapter 1, there was much debate about whether the IMW graphics would be suitable. For semifinalized recommendations (which were never officially adopted before the war), along with discussions about potential D/F problems presented by the IMW’s elevation colors, see ICAN Maps Sub-commission, “Minutes No. 18,” 6– 16 and annexes 1, 5, 8, 13, and 14. 26. For changing routes in response to airspace restrictions by Europe countries, see Burchall, “The Politics of International Air Routes.” 27. Reports of experimental Radio Range flights specifically emphasized the lack of maps. For a flight from Philadelphia to College Park, see Dellinger, Diamond, and Dunmore, “Development of the Visual-Type Airway Radiobeacon,” 834. For a flight from Detroit to Washington, see “Shaken Reed Shows Path through Clouds,” Science News-Letter 17 (24 May 1930): 335. Both tests used a visual interface that was never embraced by pilots, but the radio system was the same. 28. For landing apparatus, see Robert I. Colin, “Ernst L. Kramar: Pioneer Award 1964,” IEEE Transactions on Aerospace and Navigational Electronics 11 (June 1964): 82; Keen, Wireless Direction Finding, 616f; Kendal, “Beginnings of Directional Radio,” 321. Germany had built fourteen full-size Radio Range stations (there were also two in Austria); see Walter Blanchard, “Another Look at the Great Area-Coverage Controversy of the 1950’s,” Journal of Navigation (UK) 58 (2005): 351. These blurred the line between railroading and international standardization; compare “All Landings Blind When Necessary, Experts Forecast,” Science News-Letter 35 (10 June 1939): 357; Denman, “Radio Air Navigation,” 56. For radio compasses, see C. B. Collins, “Inter-aerodrome Navigation,” Flight, 7 Dec 1933, 1221– 1222; H. M. Samuelson, “The Future of Aircraft Radio,” Flight, 20 Jan 1938, 69; “New Inventions Add Greatly to Safety of the Aviator,” Science News Letter 22 (Oct 1938): 259; “Navigation— Fourth Rate,” Flight, 19 Jan 1939, 54. For ideas about confining traffic to routes in Europe, see R. H. S. Mealing, “Air Traffic Control,” Flight, 19 Oct 1933, 1049; Denman, “Radio Air Navigation,” 55, which discusses “track beacons.” For presentation of the Omnirange in explicitly lighthouse-like terms, see “The Radio Range Beacon,” Science 92, no. 2376 (12 July 1940): 7. For British interest, see “An Aerial Radio Lighthouse,” Flight, 19 Sept 1940, 227; “Omni-directional Radio Range,” Flight, 8 May 1941, 328. The full treatment appeared in three parts (the third delayed by the war): David Luck, “An Omnidirectional Radio-Range System,” RCA Review 6 and 7 (July 1941, Jan 1942, and Mar 1946). 29. For an overview of German systems, including those that never advanced beyond proposals or experimental trials, see Fritz Trenkle, Bordfunkgeräte— Vom Funkensender zum Bordradar (Koblenz: Bernard & Graefe, 1986). For Japan, see US Naval Technical Mission to Japan, “Japanese Navigational Aids,” Feb 1946 (scanned microfilm copy available at http:// www.scribd.com/doc/32941274/USN-Technical-Mission-to-Japan-Japanese-Navigational -Aids-US-1946); Roger I. Wilkinson, “Short Survey of Japanese Radar,” pts. 1 and 2, Electrical Engineering 65 (Aug–Sept and Oct 1946): 372, 459. For Soviet proposals, see R. V. Whelpton and P. G. Redgment, “The Development of C. W. Radio Navigation Aids, with Particular Reference to Long-Range Operation,” Journal of the Institution of Electrical Engineers— Part IIIA: Radiocommunication 94 (Mar–Apr 1947): 246. For wartime Radio Range, see “Radio Ranges on the Continent,” “Radio Ranges— United Kingdom,” and letter from H. V. Satterly (Tiger Force) to Air Ministry, 17 June 1945 (all in PRO, AIR 14/2122, “Tiger Force— Radio Ranges— Policy”). See also postwar COM reports from ICAO Pacific, Middle East, Caribbean, and North Atlantic regions (ICAO, box “PICAO COM 1945– 47”). For an overview of British systems, see F. C. Richardson, “Radionavigation in the UK in World War II,” Journal of the Institute of Navigation (UK) 45, no. 1 (1992): 60– 69. 30. See Watson-Watt quoted in Colin, “Robert J. Dippy” (n. 2 above), 479; R. A. Smith, Radio Aids to Navigation (Cambridge: Cambridge University Press, 1947), 49. 31. Stuart William Seeley (the system’s inventor), “Shoran Precision Radar,” Electrical Engineering 65 (Apr 1946): 232. See also Stuart William Seeley, “Shoran— a Precision Five Hundred Mile Yardstick,” Proceedings of the American Philosophical Society 105 (15 Aug 1961).

Notes to Pages 216–218

355

32. Chronologically, the narrow-beam X-Gerät (also known as Wotan I) was developed before Knickebein; the latter was produced as a more user-friendly version of the former. See Brown, Radar History (n. 5 above), 113– 114; Alfred Price, Instruments of Darkness (London: William Kimber, 1967), 21; Karl Hecks, Bombing 1939– 1945: The Air Offensive against Land Targets in the World War Two (London: Robert Hale, 1990), 53. 33. Winston Churchill, Their Finest Hour (Boston: Houghton Mifflin, 1949), 381f. For a detailed narrative see R. V. Jones, Most Secret War (London: Hamilton, 1978), esp. chap. 16 and 127– 129, 179; Brown, Radar History, 115, 119. For an overview of countermeasures, see Robert Cockburn (wartime head of Radio Countermeasures at TRE), “The Radio War,” IEE Proceedings 132, pt. A (Oct 1985 [orig. 1945]): 423– 434. 34. For other systems— a transportable version of Knickebein named Zyklop, a rotatable landing aid known as Karussell, and a multibeam system called Elektra— see Trenkle, Bordfunkgeräte, 107, 109, 120. See also “Radio Navigation Systems and Equipment,” an August 1945 Allied translation of a captured German original written sometime after November 1944 (NARA, RG 165, entry 79, box 1954, folder “Radio Navigation Systems and Equipment”), 12f, which describes pilots flying toward the Zyklop tower and includes a map showing five installations pointing away from the coasts of Denmark, France, Italy, and Greece. 35. These systems developed from Identification Friend-or-Foe (IFF) equipment that had originally been designed by Watson-Watt; see Lord Bowden of Chesterfield, “The Story of IFF,” IEE Proceedings 132, pt. A (Oct 1985). For other applications, see K. A. Wood, “200-Mc/s Radar Interrogator-Beacon Systems,” Journal of the Institution of Electrical Engineers— Part IIIA: Radiolocation 93, no. 2 (1946): 481– 495. For German IFF, see David Pritchard, The Radar War: Germany’s Pioneering Achievement, 1904– 45 (London: Patrick Stephens, 1989), 178– 181. 36. For the continuous-phase Y-Gerät (also known as Wotan II), see R. V. Jones, “Navigation and War” (n. 15 above), 10– 13; for intelligence against it, see Reginald Jones, “Scientific Intelligence,” Journal of the Royal United Services Institution 42 (1947): 352– 369; or Reginald Jones, Reflections on Intelligence (London: Heinemann, 1989). On the use of Y and Egon for fighter control, see Donald Caldwell and Richard Muller, The Luftwaffe over Germany: Defense of the Reich (London: Greenhill, 2007), 127– 132, 243– 429; Gebhard Aders, History of the German Night Fighter Force, 1917– 1945 (London: Jane’s, 1979 [orig. German 1978]), 76, 127; Francisco Gallei, “American and German Fighter Control through 1945” (master’s thesis, Air University, 2008). On the use of Egon for bombing over England, see R. V. Jones, “Navigation and War,” 21. 37. For Oboe, see F. E. Jones, “Oboe: A Precision Ground-Controlled Blind-Bombing System,” Journal of the Institution of Electrical Engineers— Part IIIA: Radiolocation 93, no. 2 (1946): 496– 511; A. H. Reeves and J. E. N. Hooper, “Oboe: History and Development,” IEE Proceedings 132, pt. A (Oct 1985 [orig. 1945]). For Shoran, see Seeley, “Shoran Precision Radar,” 232– 240; Frederick J. Green Jr., “Shoran Stations,” 8 Mar 1945 (NARA, RG 331, entry 268, box 78, folder “Shoran and Rebecca-H Policy”). 38. For quotes, see Henry E. Guerlac, Radar in World War II (New York: American Institute of Physics, 1987), 836, 903. For operational accuracy of Oboe, see F. E. Jones, “Oboe,” 496; Reeves and Hooper, “Oboe,” 398; F. E. Jones et al., “D.S.R. Historical Monograph: Oboe,” Apr 1946 (PRO, AVIA 44/519, “Oboe”), 208– 213. For Shoran, see H. R. Crowley, “Shoran,” 8 Mar 1945 (NARA, RG 331, entry 276F, box 131, “SHAEF Air Staff, Air Signal Division, Radar Section, Numeric Subject File, July 1943– Dec 1944”). On the Norden bombsight, see Stephen L. McFarland, America’s Pursuit of Precision Bombing, 1910– 1945 (Washington DC: Smithsonian Institution Press, 1995). 39. “H” was the British codeletter for the range-range technique. Rebecca-H used IFF-based equipment; see Wood, “200-Mc/s Radar Interrogator-Beacon Systems,” 493. For Micro-H, see Charles W. McArthur, Operations Analysis in the U.S. Army Eighth Air Force in World War II (Providence: American Mathematical Society, 1990), 175– 176. For surveying use of Oboe, Gee, and G-H (also known as Gee-H, since it piggybacked on Gee equipment) from 1943 through the late 1940s, see “Discussion on ‘Radar Navigation,’” Journal of the Institution of Electrical Engineers— Part IIIA: Radiolocation 93, no. 2 (1946): 511– 512, and C. A. Hart, “Surveying from

356

Notes to Pages 218–221

Air Photographs Fixed by Remote Radar Control,” in The Royal Society Empire Scientific Conference, June–July 1946: Report, vol. 2 (London, 1948), 649. 40. In 1946 Seeley reported that before the war, “it was not known, at first, to what uses a system for accurate transmission [of] path length measurement could be put”; see “Shoran Precision Radar,” 232. For quote, see Crowley, “Shoran,” 1. For the bombing of Belgium, see R. V. Jones, Most Secret War, 276– 277. 41. For introduction of Gee, see UK Ministry of Civil Aviation, “An Outline of the Technical Performance of the ‘Gee’ Radio Navigation System during the War of 1939– 45,” Aug 1946 (NARA, RG 319, MLR NM3 82, “Publications (‘P’) Files, 1946– 51,” box 2727), 3. For Sonne, see “Radio Navigation Systems and Equipment.” For Loran, see J. A. Pierce, “An Introduction to Loran,” Proceedings of the IRE 34 (May 1946): 216– 234. For Decca, see Claud Powell, “Early History of the Decca Navigator System,” Journal of the Institution of Electronic and Radio Engineers 55 (June 1985): 203– 209. 42. For Gee, see Watson-Watt quoted in Colin, “Robert J. Dippy,” 478; for Loran, see Pierce, “An Introduction to Loran,” 219. 43. For German experiments with hyperbolic systems, see Trenkle, Bordfunkgeräte, 134, 137. For an overview, see W. F. Blanchard, “Air Navigation Systems, Chapter 4: Hyperbolic Airborne Radio Navigation Aids— a Navigator’s View of Their History and Development,” Journal of Navigation (UK) 44 (Sept 1991). 44. For details on Sonne, see A. H. Jessell, “The Range and Accuracy of Consol,” Journal of the Institute of Navigation (UK) 1 (July 1948): 241– 256; “Consol Navigation System,” International Hydrographic Review 26 (Nov 1949): 66. Many years after the war, Sonne (and the VHF Omni-Range) were sometimes described as “collapsed” hyperbolic systems; see E. Kramar, “Hyperbolic Navigation— History and Outlook,” Interavia (Feb 1969): 174– 177. 45. See Dippy’s remarks in Colin, “Robert J. Dippy,” 476. Reginald Jones notes that resistance to radio navigation was still widespread in the British high command a year later, when a scathing report was issued calling attention to the problem; see Most Secret War, 210, 217. 46. Although Loran is traditionally credited to the American millionaire polymath Alfred Loomis, recent evidence suggests that his ideas came from conversations with a loose-lipped British engineer during the famous Tizard Mission of September 1940; see E. G. Bowen, Radar Days (Bristol: A. Hilger, 1987), 171– 174; Jennet Conant, Tuxedo Park: A Wall Street Tycoon and the Secret Palace of Science That Changed the Course of World War II (New York: Simon and Schuster, 2002), 199– 200, 231– 234. Immediately after the war one of the leaders of the Loran effort suggested that Loran “may be said to have been invented in America in the sense in which Galileo is said to have invented the telescope”: Pierce, “An Introduction to Loran,” 217. 47. J. A. Pierce, A. A. McKenzie, and R. H. Woodward, eds., Loran: Long Range Navigation (New York: McGraw Hill, 1948), 20. 48. Wilfred Lewis, cited in J. W. S. Pringle, “The Work of TRE in the Invasion of Europe,” IEEE Proceedings 132, pt. A (Oct 1985): 356. For Sonne in the Bay of Biscay, see A. G. Watson, “Radio Aids to Navigation,” Navigation (US) 3 (June 1952): 130. For development, see Colin, “Ernst L. Kramar” (n. 28 above), 83– 84; Kramar, “Hyperbolic Navigation,” 174– 177; Trenkle, Bordfunkgeräte, 123. For Decca, see Powell, “Early History of the Decca Navigator System.” 49. For the inadequacy of fixed-path navigation in war, see Pierce, McKenzie, and Woodward, Loran, 28, 35. 50. Letter from office of the senior air staff officer, Headquarters, Tactical Air Force, “Gee Topographical Lattice Maps,” 8 Oct 1943 (NARA, RG 331, entry 268, box 78, folder “Lattice Charts Production— Policy”). 51. Pierce, “An Introduction to Loran,” 227. 52. For “map reading,” see letter from office of Air Marshal Trafford Leigh-Mallory (air officer commanding-in-chief, Fighter Command), “Gee Topographical Lattice Maps,” 14 Oct 1943 (NARA, RG 331, entry 268, box 75, folder “Lattice Charts Production— Policy”); or “Organi-

Notes to Pages 221–228

357

zation and Operational Employment of ‘Gee’ in the AEAF,” Mar 1944 (NARA, RG 331, “SHAEF Special Staff, Signal Division, Requirements & Administrative Section, Numeric-Subject File 1943– 45,” box 43, folder “Radar Navigational Aid (Gee)”). For “interpretation,” see letter from office of Leigh-Mallory (as commander-in-chief, Allied Expeditionary Air Force), “Mobile ‘Gee’ Operational Use by AEAF; Provision of Charts,” 2 Mar 1944 (NARA, RG 331, entry 268, box 75, folder “Lattice Charts Production— Policy”). 53. See letters from Air Marshal Arthur Coningham (air officer commanding-in-chief, Second Tactical Air Force), “Siting of Gee Stations,” 3 Oct 1944, from office of Air Vice-Marshal Victor Hubert Tait (director-general of signals), “Gee Chains on the Continent— Preparation of Charts,” 23 Oct 1944 (both NARA, RG 331, entry 276F, box 128, “SHAEF Air Staff, Air Signal Division, Radar Section, Numeric Subject File, July 1943– Dec 1944”), and again from office of Coningham, “R.N.A. Plans (Overlord and Eclipse),” 4 Apr 1945 (NARA, RG 331, entry 268, box 77, folder “Continental Gee and G-H Cover Plan”). For ongoing plans and problems, see, for example, the loose minute from Squadron Leader David Bruce, 10 Apr 1945 (NARA, RG 331, entry 268, box 77, folder “Continental Gee and G-H Cover Plan”); or folder “Continental Cover Plan, GEE and G-H” (NARA, RG 331, entry 268, box 77). 54. See Pierce, McKenzie, and Woodward, Loran, 93– 94, 182– 186. 55. Dickie Richardson, Man Is Not Lost: The Log of a Pioneer RAF Pilot/Navigator, 1933– 1946 (Schrewsbury: Airlife, 1997), 233– 234; R. V. Jones, “Navigation and War,” 14. 56. Quote from “Closing Speech by Sir Robert Watson-Watt,” in “Report on International Meeting on Radio Aids to Marine Navigation, London, 1946” (NARA, RG 43, “International Meeting on Marine Radio Aids to Navigation,” box 8, folder “Jansky Papers”), iv. The bombing and rebuilding is described in a letter from Sven Pran (a Swedish radio engineer) to Jerry Proc, http://jproc.ca/hyperbolic/consol.html (posted Dec 2008); this letter also mentions the repair of the station at Lugo, Spain. See also Blanchard, “Air Navigation Systems,” 312. The modifications required to give asymmetric coverage are described in A. H. Brown, “The Consol Navigation System,” Journal of the Institution of Electrical Engineers— Part IIIA: Radiocommunication 94, no. 16 (1947): 973– 974. 57. For German jamming and British countermeasures, see UK Ministry of Civil Aviation, “An Outline of the Technical Performance,” 3. The Germans did continue to try to jam Gee in certain areas— Normandy, for example— but British bombing of jammers and use of alternate frequencies thwarted these efforts; see R. A. Smith, “Radar Navigation,” Journal of the Institution of Electrical Engineers— Part IIIA: Radiolocation 93, no. 1 (1946): 335. For German plans for Truhe (a codename meaning “storage chest”), see “Radio Navigation Systems and Equipment” (n. 34 above), 15f, 26. The associated ground equipment was known as Bodentruhe; see Trenkle, Bordfunkgeräte (n. 29 above), 134– 137. 58. Col. Talbot (of IX Bombing Division), “Oboe and Gee Jamming,” 14 Mar 1945 (NARA, RG 331, entry 276F, box 131, “SHAEF Air Staff, Air Signal Division, Radar Section, Numeric Subject File, July 1943– Dec 1944”). 59. For the first and second Commonwealth and Empire Conferences on Radio for Civil Aviation, see “C.E.R.C.A.,” Flight, 16 Aug 1945, 182. A third such conference was held in August 1945, chaired by Watson-Watt; its final report was filed as ICAO doc. 409 (ICAO, box “PICAO— Documents (1945– 46): 401– 1800”). Glossy pamphlets explaining the postwar virtues of Loran were available in August 1945. See US Navy, “Loran: Long Range Radio Navigational Aid,” Aug 1945 (ICAO, box “Com— Sub. 1, 2, & 3: 1945– 1949”). 60. For Decca, see Powell, “Early History of the Decca Navigator System,” 208; C. Powell, “Radio Aids to Surveying,” in Conference of Commonwealth Survey Officers: 1955 Report of Proceedings (London, 1955), 91. For others, see n. 39. 61. Conrad Crane, “Raiding the Beggar’s Pantry: The Search for Airpower Strategy in the Korean War,” Journal of Military History 63 (Oct 1999): 909, 917. For a discussion of some of the problems associated with Shoran in Korea, see Daniel Kuehl, “Refighting the Last War: Electronic Warfare and U.S. Air Force B-29 Operations in the Korean War, 1950– 53,” Journal of Military History 56 (Jan 1992): 95– 100.

358

Notes to Pages 229–232

62. Seeley, “Shoran— a Precision Five Hundred Mile Yardstick” (n. 31 above), 450– 451. 63. Shoran work in Canada was very well publicized within the survey profession; see J. E. R. Ross, “Shoran Triangulation in Canada,” Bulletin géodésique 24 (1952); J. E. R. Ross, “Shoran Operations in Canada,” Surveying and Mapping 12 (Oct–Dec 1952): 363– 370; J. E. R. Ross, “Canadian Shoran Effort, 1949– 1953,” pts. 1 and 2, Empire Survey Review 12 (Apr and July 1954): 242– 254, 290– 303. On the use of Hiran and Shiran, see Carl I. Aslakson, “The Influence of Electronics on Surveying and Mapping” Surveying and Mapping 10 (July–Sept 1950); B. B. Hunkapiller, “Aerial Electronic Surveying,” Navigation (US) 5 (June 1956); Richard B. H. Shepherd, “Shoran and Hiran in Geodetic Surveying,” 4 Sept 1958, Notes of the Week (Tokyo: US Army Map Service, Far East, 1958– 9)— typescript papers available at NOAA; E. M. Salkeld Jr., “Development of a Precise Geodetic Survey System,” IEEE Transactions on Aerospace and Electronic Systems 2 (Jan 1966). 64. For connections between surveying and missile guidance, see Deborah Jean Warner, “From Tallahassee to Timbuktu: Cold War Efforts to Measure Intercontinental Distances,” Historical Studies of the Physical Sciences 30, no. 2 (2000). 65. Carl I. Aslakson, “Geodetic Control in Jungle Areas,” Transactions of the American Geophysical Union 39 (Oct 1958); Simo H. Laurila, “Across the Jungle by Hiran,” Surveying and Mapping 24 (Mar 1964). For US work in Ethiopia and Iran, see Clifford Mugnier, “Federal Democratic Republic of Ethiopia,” Photogrammetric Engineering & Remote Sensing 69 (Mar 2003), Grids and Datums, 213; Clifford Mugnier, “Islamic Republic of Iran,” Photogrammetric Engineering & Remote Sensing 79 (Aug 2013), Grids and Datums, 684. 66. In addition to hyperbolic coordinates, offshore surveys could also be performed using distance-measuring apparatus similar to Shoran. Sea-Fix and Hi-Fix were developed from Decca, and EPI was a cross between Shoran and Loran; see Clarence A. Burmister, “Electronics in Hydrographic Survey,” International Hydrographic Review 26 (May 1949). The new systems (the American Raydist and French Lorac, Rana, and Toran) were derived from a 1934 French patent that finally found life in the late 1940s. For details, see Charles E. Hastings, “The Application of Raydist to Hydrographic Surveying,” International Hydrographic Review 26 (Nov 1949); Seismograph Service Corporation, “Lorac: A Radiolocation System Having Long Range Accuracy,” International Hydrographic Review 26 (May 1949); Étienne Honoré and Émile Torcheux, “Les radionavigateurs Rana,” Navigation (France) 1 (Jan 1953): 38– 44; P. Laurent, “Toran,” in Radio Aids to Maritime Navigation and Hydrography, International Hydrographic Bureau Special Publication 39, 2nd ed. (Monaco, 1965). 67. “Nigerian Venture,” Decca Navigator News 26 (Nov 1959): 9. For Shell in Qatar, see “This Too Used the Navigator,” Decca Navigator News 12 (Dec 1954): 9. For BP, “150 Tankers,” Decca Navigator News 28 (Sept 1960): 2– 3. 68. Eventually there were even proposals for precise trilateration networks on the ocean floor. See A. G. Mourad and N. A. Frazier, “Improving Navigational Systems through Establishment of a Marine Geodetic Range,” Navigation (US) 14 (Summer 1967); Andrew C. Campbell, “Geodetic Positioning at Sea,” Navigation (US) 15 (Spring 1968). 69. For comparison of nautical charts and topographic maps, see G. D. Dunlap, “Major Developments in Marine Navigation during the Last 25 Years,” Navigation (US) 18 (Spring 1971): 76. For underwater relief, see Ronald E. Doel, Tanya J. Levin, and Mason K. Marker, “Extending Modern Cartography to the Ocean Depths: Military Patronage, Cold War Priorities, and the Heezen-Tharp Mapping Project, 1952– 1959,” Journal of Historical Geography 32 (2006); Naomi Oreskes, The Rejection of Continental Drift (New York: Oxford University Press, 1999), 267 (the cover of which shows a very similar shaded-relief map). For Harrison’s map, see “The Newly Discovered World beneath the Waves,” Fortune, Nov 1959. 70. See S. N. Nandan, “The Exclusive Economic Zone: A Historical Perspective,” in FAO Essays in Memory of Jean Carroz: The Law and the Sea (Rome: FAO, 1987); Lewis M. Alexander, “The Expanding Territorial Sea,” Professional Geographer 11, no. 4 (1959): 6– 8; Legislation Branch, FAO, “Limits and Status of the Territorial Sea, Exclusive Fishing Zones, Fishery Conservation Zones and the Continental Shelf,” International Legal Materials 8 (1969): 516– 546. For the con-

Notes to Pages 232–235

359

tinental shelf in particular, see A. D. Couper, “The Marine Boundaries of the United Kingdom and the Law of the Sea,” Geographical Journal 151 (July 1985): 228– 236. 71. At the 1958 UNCLOS meetings, radiolocation was seen as sufficient grounds for initiating “hot pursuit”; see comments by Denmark in UNCLOS Document A/CONF.13/5 (5 Aug 1957), 81, and by the US in A/CONF.13/C.2/SR26-30 (8 Apr 1958), 80, in vols. 1 and 4 of Official Records of the United Nations Conference on the Law of the Sea (Geneva: UN, 1958). On the links between marine navigation, surveying, and grid conflicts, see J. D. Nares (director of the International Hydrographic Bureau), “Coordination of Geographical Grids of the World,” in J. M. Tienstra, “Comptes rendus des séances de travail de la Section des Triangulations de l’Association Internationale de Géodésie,” Bulletin géodésique 23, no. 1 (1949): 15– 16; Pierre Tardi, “L’unification des réseaux géodésiques et les problèmes de navigation,” Bulletin géodésique 26, no. 4 (1952); Floyd Hough, “The Universal Transverse Mercator Grid (with Particular Reference to Africa),” Conference of Commonwealth Survey Officers, 1951: Report of Proceedings (London: HMSO, 1955), 60. For the North Sea, see T. C. Haile, “Political Aspects of the Charting of the Seas,” Journal of Navigation (UK) 34 (Jan 1981): 66. 72. For marine meetings, see R. B. Michell, “The Second International Meeting on Radio Aids to Marine Navigation,” Journal of the Institute of Navigation (UK) 1 (Jan 1948). For frequency standardization, see, for example, records of the Special Administrative Conference for the North-East Atlantic, Jan–Feb 1949 (PRO, MT 9/5133, “LORAN Conference, Geneva, January 1949”); or various ITU meetings. For specific use of “technical merits,” see testimony of Harold Watkinson (minister of transport and civil aviation) before the House of Commons, 4 Feb 1959, Parliamentary Debates, 5th series, vol. 599, col. 383. 73. The main forum was the standing COM committee at ICAO, but there was also a special conference to discuss short-range aids in particular; see ICAO, boxes SP/COM/OPS/RAC and COM-1 through COM-7. 74. Loran coverage in 1945 was about sixty-two million square miles; see Pierce, McKenzie, and Woodward, Loran (n. 47 above), 50. For Loran plan, see pamphlet from US Coast Guard, Electronic Navigational Aids: Loran— Radar— Racon, n.d. (NARA, RG 319, MLR NM3 82, “Publications (‘P’) Files, 1946– 51,” box 1899), 3. 75. Watson-Watt quotes from “Radio Aids to Air and Sea Navigation,” 12 July 1946 (PRO, MT 9/4457, “Radio Aids to Marine Navigation: Proposal for a Combined GEE/LORAN (GLORAN) System for Marine Use”); and “Gee,” Dec 1946 (PRO, BT 217/323). For his initiative in offering free equipment, see his minute of 29 Mar 1946 (PRO, BT 217/323). Much of this was part of secret plans by TRE and the British military for Europe-wide civilian Gee coverage; see “Gee Chains for Civil Aviation: Minutes of Meeting Held at Headquarters, No. 60 Group,” 18 Dec 1945 (PRO, BT 217/323). 76. For US offering Radio Range equipment at “give-away” terms, see “Minutes of a Meeting Held in Inveresk House at 1500 hours,” 15 Apr 1946 (PRO, BT 217/323). See also Blanchard, “Another Look” (n. 28 above), 349– 363. In the late 1940s, the US successfully pushed its allies for temporary extensions for its North Atlantic Loran stations, despite Loran having been excluded from the formal frequency allocations decided at the International Radio Conference of 1947; see records of the Special Administrative Conference for the North-East Atlantic. For US policy on its VOR equipment versus British ILS manufactures, see Jean Jacques Duprez, “The Strategic Embargo: Doctrine and Practice,” World Today 19 (Sept 1963): 378. 77. Watson-Watt, The Pulse of Radar (see n. 2 above), 340. It is worth noting that WatsonWatt suggested a Gee-Loran hybrid known as Gloran at the first International Meeting on Radio Aids to Marine Navigation, where it was enthusiastically received by the US. In the end, however, Gloran would have been “hopelessly uneconomic” for marine use, and developing the necessary airborne equipment (and dealing with patent issues) would have taken years; see Watson-Watt, “Radio Aids to Air and Sea Navigation,” 12 July 1946; or minute from Oates to O’Neill, 12 Mar 1948 (both in PRO, MT 9/4457, “Radio Aids to Marine Navigation: Proposal for a Combined GEE/LORAN (GLORAN) System for Marine Use”).

360

Notes to Pages 235–241

78. For ships, the lines separating harbor, coastal, and oceanic navigation were drawn roughly at five and one hundred kilometers from the nearest obstruction. For aircraft, local operations were airport specific, and the line between short- and long-range navigation was at about four hundred kilometers. For marine navigation, see Michell, “The Second International Meeting on Radio Aids to Marine Navigation,” 73; for aviation, see the final report of the ad hoc committee on systems, 8 Nov 1945, doc. 693 (ICAO “COM— Sub. 1, 2, & 3: 1945– 1949”), 6; or comments from Canada on draft specifications for nondirectional beacons, app. A to COM IV-WP/33, 1 Feb 1951 (ICAO, box “COM-4, 1951”), 6. The standardized blind-landing system— a hybrid of American and British technology known simply as the instrument landing system, or ILS— is still in use today. It blended a British radar-based system known as ground-controlled approach (GCA) and a US military beam landing system named SCS-51. The US put forward the basic ILS-with-GCA-supplement proposal immediately after the war; see “U.S. Proposal, Section VII: Technical Characteristics of Communications Systems,” 18 Oct 1945, doc. 467 (ICAO, box “COM— Sub. 1, 2, & 3: 1945– 1949”). This had been accepted in principle by the British within a few months; see opening address by Robert Watson-Watt, in “P.I.C.A.O. Demonstrations in the United Kingdom of Radio Aids to Navigation,” 24– 25 Sept 1946 (NARA, RG 319, MLR NM3 82, “Publications (‘P’) Files, 1946– 51,” box 2727), 8. For adoption, see “Division Meetings: Communication,” ICAO Monthly Bulletin, Apr 1949, 5– 8. For technical details, see Colin, “Ernst L. Kramar” (n. 28 above), 83; or the British brochure prepared for PICAO, Demonstrations of Radio Aids to Civil Aviation (London: HMSO, 1946), 35– 50. For conflict between ILS and GCA in the US, see Conway, “The Politics of Blind Landing” (n. 21 above). In the UK, see F. R. Willis, “Radar in Civil Aviation,” Flight, 20 Mar 1947, 242– 243. In the 1970s a replacement for ILS known as the microwave landing system (MLS) was endorsed by ICAO, but installation was slow and never complete due to competition from GPS-based systems; see Per Enge et al., “Terrestrial Radionavigation Technologies,” Navigation (US) 42, no. 1 (1995): 98– 106. 79. The proposal paired the directional VOR with a new American system known as DME (distance measuring equipment); the full system is known simply as VOR/DME. When combined with an ITT-designed military system known as TACAN, it became VORTAC. By 1960, there were about six hundred VOR beacons in the US, eighty-five of which were of the TACANcompatible variety; eleven hundred VORTACs were planned for 1965. See “DME Procedures and Implementation,” Flight, 1 Apr 1960, 441. For VOR/DME in Europe, the US, and Australia, see “Airline Navigation” (n. 4 above), 333; J. H. Grover, “Long-Range Navaids,” Flight, 27 Apr 1956, 495. For TACAN, see Peter C. Sandretto, “Development of Tacan at Federal Telecommunication Laboratories,” Electrical Communication 33 (Mar 1956): 4– 10; W. L. Garfield, “Tacan: A Navigation System for Aircraft,” Proceedings of the IEE, Part B: Radio and Electronic Engineering 105, no. 9, pt. S (1958): 298– 306. 80. Blanchard, “Another Look,” 349– 363. For “débâcle,” see “Short Range Aide-Memoire,” Flight, 29 May 1959, 733. For vote packing, see “Britain Opposes VOR/DMET,” Flight, 6 Mar 1959, 308– 309. For US agenda strategy— and British reaction— see statement by the US “concerning the attitude of the meeting” in “Draft Report: Special COM/OPS/RAC Meeting,” 28 Feb 1959, doc. COR-WP/70 (ICAO, box “SP/COM/OPS/RAC 1958”), p. IV-3. For the existential threat to ICAO, see “Britain Accepts DMET,” Flight, 20 May 1960, 682. For Decca in helicopters, see J. G. Adam, “Decca for Helicopter Operations,” Journal of the Institute of Navigation (UK) 9 (Oct 1956): 385– 389. For Decca on ships, see C. Powell, “The Decca Navigator System for Ship and Aircraft Use,” Proceedings of the IEE, Part B: Radio and Electronic Engineering 105, no. 9, pt. S (1958): 225. 81. For use of Consol by fishing fleets, see J. C. Farmer, “Survey of Long-Range Radio Navigation Aids,” Proceedings of the IEE, Part B: Radio and Electronic Engineering 105, no. 9, pt. S (1958): 219. For stations, see Ernst Kramar, “Consol and Consolan,” in Radio Navigation Systems for Aviation and Maritime Use: A Comparative Study, ed. W. Bauss (New York: Macmillan, 1963), 31; Geoffrey Edward Beck, Navigation Systems: A Survey of Modern Electronic Aids (New York: Van Nostrand Reinhold, 1971), 113. The US stations were slightly different in design and were known by the name Consolan.

Notes to Pages 241–243

361

82. These new systems included Navaglobe, Navarho, Dectra, Delrac, Radio Mailles, Radux/Omega, and Loran-C. They were presented widely in navigation journals and discussed in “Report of the Sixth Session,” Oct 1957, doc. 7831 (ICAO, box “COM-6, 1957”). 83. P. C. Gaudillère, “Un système de base pour la radionavigation: ‘Radio-Mailles,’” pts. 1 and 2, Navigation (France) 1 (1953): no. 1, 45– 48, and no. 2, 25– 43; “Radio Mailles: A French Navigation and Traffic Control System,” Flight, 19 July 1957, 87; “And Now Omega,” Flight, 25 Oct 1957, 640; Robert L. Frank, “History of Loran-C,” Navigation (US) 29 (Spring 1982): 1– 6; Edward L. McGann, “The Evolution of Loran-C Coverage,” Navigation (US) 29 (Spring 1982): 89– 101. See also the previous note. 84. Memorandum of conversation, “High Precision Long-Range Navigation System (Loran C) for Use in Connection with United States Ballistic Missile Program,” 15 Jan 1958 (NARA, RG 59, “Bureau of European Affairs, Office of European Regional Affairs, Political-Military Numeric Files 1953– 1962,” box 19). The expense of Loran-C receivers was noted widely in navigation journals. 85. First from memo to Norman Brook, “Loran-C,” n.d. (PRO, HO 255/858, “US Proposals for Loran ‘C’”), 2; second from memo from UK Post Office Engineering Department, “Loran C,” 2 July 1959 (PRO, HO 255/858, “US Proposals for Loran ‘C’”), 1. 86. US working paper COM VII-WP/111, 10 Jan 1962 (ICAO, box “COM-7, 1962”), 2; this paper explicitly references the US policy position laid out in 1958. UK working paper COM VII-WP/166, 30 Jan 1962 (ICAO, box “COM-7, 1962”). 87. Several other kinds of self-contained navigation were also available for military use, such as automatic star trackers and radio mapping, and inertial systems were especially important on nuclear submarines. For the history of self-contained navigation, see R. V. Jones, “Navigation and War” (n. 15 above), 21– 23; R. B. Horsfall, “Stellar Inertial Navigation,” IRE Transactions on Aeronautical and Navigational Electronics 5 (June 1958): 106; William J. Tull, “Doppler Navigation,” Navigation (US) 5 (Summer 1957): 290– 298; H. Hellman, “The Development of Inertial Navigation,” Navigation (US) 9 (Summer 1962), 83– 94; Donald MacKenzie, Inventing Accuracy: A Historical Sociology of Nuclear Missile Guidance (Cambridge, MA: MIT Press, 1990), chap. 2. For its commercialization, see reports on Doppler by the US and UK in “Report of the Sixth Session,” pp. VII-103 to VII-148; Richard Witkin, “Aviation: Congestion. Declassification of Military Radar Devices May Help Clear the Air,” New York Times, 11 Aug 1957, 114; “Navigation— Inertial Portents: Doppler Debut: Radio Aids Full House,” Flight, 16 May 1958, 659– 665; “Doppler in Practice,” three papers in Journal of the Institute of Navigation (UK) 14 (Jan 1961): 34– 63; Joseph F. Galigiuri, “SGN-10 First Commercial Inertial Navigator,” Navigation (US) 14 (Spring 1967): 85– 92; Alexander B. Winick, “Air Navigation Trends,” Navigation (US) 15 (Spring 1968): 79. 88. For the lack of any need for standardization, see in particular US working paper AN Conf/4-WP-17, 20 July 1965 (ICAO, box “AN-Conference 4”), 3. 89. Karl Ramsayer, “Integrated Navigation,” Journal of the Institute of Navigation (UK) 16 (Jan 1963). For integration in general, see A. Stratton, “The Combination of Inertial Navigation and Radio Aids,” Proceedings of the IEE, Part B: Radio and Electronic Engineering 105, no. 9, pt. S (1958): 266– 276. For Decca systems in particular, see C. Powell, “An Elementary Compound System,” Journal of the Institute of Navigation (UK) 16 (Oct 1963): 467– 472; M. G. Pearson, “The Use of an Airborne Digital Computer in a Compound Navigation System,” Journal of the Institute of Navigation (UK) 16 (Oct 1963): 472– 475; “Decca Developments,” Flight, 15 Apr 1960, 525; Blanchard, “Another Look,” 357. For similar solutions for Loran, see Loren E. De Groot, “Loran-Inertial Navigation Systems for Long-Range Use,” Journal of the Institute of Navigation (UK) 18 (July 1965): 319– 329. For both analog and digital computers, the chief computational difficulty was converting between polar and hyperbolic coordinates. For use of computers in the early 1970s, see P. M. Grindon-Ekins, “The Impact of Digital Computing,” Journal of Navigation (UK) 27 (July 1974): 323– 331; R. A. Severwright, “The Impact of Special-Purpose Computers on Aircraft Equipment,” Journal of Navigation (UK) 28 (July 1975): 398– 405. 90. “Report of Committee C to the Conference on Item 9,” 19 Nov 1965, doc. AN Conf/4WP/75 (ICAO, box “AN-Conference 4”), p. 9-2. The long-distance-aids section of ICAO annex

362

Notes to Pages 244–246

10 was updated in 1968 and remained unchanged until the mid-1990s, when all mentions of Loran-A and Consol were deleted. See ICAO, International Standards and Recommended Practices: Aeronautical Telecommunications; Annex 10 to the Convention on International Civil Aviation (Montreal: ICAO), 2nd through 6th editions of vol. 1, especially attachment A in editions of 1968, 1972, and 1985; this attachment did not appear in 1996 or 2006. 91. Analog computers for navigating away from strict VOR paths had been promised as early as the first ICAO meetings but had never materialized. For UTM in particular, see Lipsey, “Tactical Air Navigation” (n. 3 above), 307. 92. “D.I.A.N.,” Flight, 23 Aug 1957, 250. See also E. R. Wright, “The Use of the Flight Log,” Journal of the Institute of Navigation (UK) 9 (Oct 1956): 389– 393. 93. For briefing system, see G. Wikkenhauser, “A Roller Map Equipment,” Journal of the Institute of Navigation (UK) 13 (1960): 104. For projection, see Eric S. Guttmann, “Chart Logistics for Advanced System Requirements,” Navigation (US) 12 (Winter 1965– 1966): 339– 347. 94. Unfolding a large map in a cramped cockpit was, of course, impossible, and it was common for pilots to see only part of a map at one time. Common techniques included special folding patterns and manual roller-map holders. For early examples, see Omar B. Whitaker, “Aeronautical Charts,” Geographical Review 4 (July 1917). 95. For the US, see Edward Hudson, “Use of ‘Area Navigation’ System Planned to Help the Air Traffic,” New York Times, 13 Mar 1969, 94; Robert Lindsey, “New Air Navigator Blazes Its Own Trail,” New York Times 21 May 1970, 70. On the institution of RNAV in Britain, see memos from late 1972 in PRO, DR 37/162, “Short Range Navigational Aids— Overall Policy.” See also W. P. Robinson, “Area Navigation,” Journal of the Institute of Navigation (UK) 24 (July 1971): 379– 391. 96. “F.A.A. Is Criticized on Flight Delays,” New York Times, 29 June 1968, 58. 97. RNAV in the US was eventually rolled back in the early 1980s when it was discovered that pilots were switching off the ground inputs to their computers. Since the introduction of GPS, RNAV has been revived; for the elimination of RNAV routes in 1983 and their resuscitation after 2000, see FAA, “Establishment of Area Navigation Routes (RNAV),” Federal Register 68 (9 May 2003): 24864– 24866. 98. Quotes from “Report of Committee C,” p. 9-1; for “system,” see US working paper AN Conf/4-WP-17, 2. 99. For ICAO, see “Report of Committee C,” p. 9-1. For marine navigation, see Nicholas S. Christopher, “Marine Integrated Navigation System,” Navigation (US) 16 (Winter 1969– 1970): 419. 100. Bjørn A. Rørholt, “Electronic Aids to Navigation for Fishing Vessels and Other Open Sea Users,” Navigation (US) 16 (Fall 1969): 248. The new Consol stations gave coverage of the Norwegian Sea, with stations at Andøya, Bjørnøya, Jan Mayen, and Varhaug. 101. The importance of traffic control was widely stressed; see, for example, Winick, “Air Navigation Trends.” For marine traffic, see Rørholt, “Electronic Aids,” 247; R. B. Richardson, “The New Reach in Navigation at Sea,” Journal of the Institute of Navigation (UK) 23 (July 1970): 359– 365; Thomas D. Mara, “Automation of Merchant Ship Navigation Systems,” Navigation (US) 18 (Summer 1971): 215– 220. For ICAO, see, for example, memo from the ICAO Secretariat for the Special Committee on Future Air Navigation Systems, “The Systems Approach,” 14 May 1984, doc. FANS/1-WP/4 (ICAO, vol. “FANS-1 WP’S O-B DP’S ETC EFRS”), or revisions of ICAO doc. 9613, “Manual on Required Navigation Performance (RNP),” 1994, and “PerformanceBased Navigation (PBN) Manual,” 2008 (both in ICAO, box “Documents 9610– 9620”). See also the FAA’s Roadmap for Performance-Based Navigation, first issued in 2003 (linked at http:// www.faa.gov/news/fact_sheets/news_story.cfm?newsId=8768). 102. The US and USSR began exchanging information on Loran-C and Chayka in 1980, and agreements were signed in 1987 and 1988 for standardized signals and cooperative operation in the Bering Sea. See Gary R. Westling, “Joint Soviet/American Loran Operations: The Bering Sea Chain,” IEEE Aerospace and Electronic Systems Magazine 4 (Feb 1989). 103. Sven. H. Dodington, “Ground-Based Radio Aids to Navigation,” Navigation (US) 14 (Winter 1967– 1968): 386; John Larsen, “The Integration of Air Navigation Systems,” Navigation (US) 15 (Winter 1968– 1969): 410.

Notes to Pages 246–249

363

104. Christopher, “Marine Integrated Navigation System,” 419. Rørholt, “Electronic Aids,” 247. 105. For conversion away from hyperbolic coordinates to a Gauss-Krüger grid, see Simo Laurila, “Decca in Off-Shore Survey,” Bulletin géodésique 39 (1956): 66; more generally, see C. Powell, “Two-Range Decca as an Aid to Hydrography,” Journal of the Institute of Navigation (UK) 11 (Jan 1958): 94. 106. As of 2014, the US government planned to reduce VOR to a “minimum operating network” by 2020 that would act only as a backup to GPS. See Department of Defense, Department of Homeland Security, and Department of Transportation, 2014 Federal Radionavigation Plan (available at ntl.bts.gov).

Chapter 6 1. Bradford W. Parkinson and Stephen T. Powers, “Fighting to Survive,” GPS World 21 (June 2010): 12; Bradford W. Parkinson and Stephen T. Powers, “Origins of GPS,” GPS World 21 (May 2010): 30. The Atheros AR1520 chip released in 2010 measured five millimeters square; see Louis E. Frenzel, “Chips Improve GPS, Wi-Fi Performance,” Mobile Dev & Design, 22 Jan 2010 (available at http://mobiledevdesign.com/hardware_design/chips-improve-gps-wifi -performance-012210). Location-based services (LBS) do not always use GPS, but they share a similar chronology; the first LBSs are generally dated to the late 1990s. See, for example, Anind Dey, Jeffrey Hightower, Eyal de Lara, and Nigel Davies, “Location-Based Services,” IEEE Pervasive Computing 9 (Jan–Mar 2010): 11– 12. 2. The technical literature on the design and operational details of GPS is staggeringly vast; a full GPS bibliography would include thousands of reports, articles, books, and even several dedicated GPS periodicals. (Most archival records, unfortunately, remain classified or otherwise inaccessible.) I am grateful for the help of Nathaniel S. Patch, NARA archivist, who searched diligently without success for GPS records in the US National Archives; navy records have not been transferred to NARA, and the bulk of DOD records remain unorganized and undescribed. For bibliographies, see Wendlynn Wells, David E. Wells, and Alfred Kleusberg, Global Positioning System Bibliography, Dredging Research Report DRP-92-2 (Washington DC: US Army Corps of Engineers, 1992; available at dtic.mil); Robert R. Whitlock and Thomas B. McCaskill, NRL GPS Bibliography (Washington DC: Naval Research Laboratory, 2009); or Tomás Soler, “GPS/GNSS Current Bibliography,” published annually since 1998 in GPS Solutions and available at http://www.ngs.noaa.gov/CORS/GPS_Bibliography. GPS periodicals include GPS World (since 1990), GPS Report (since 1991; later Global Positioning and Navigation News), Earth Observation Magazine (since 1992), GPS Solutions (since 1995), and Inside GNSS (since 2006). 3. William F. Buckley Jr., “Precision Sailing,” New York Times Magazine, 19 May 1985, 135. GPS was described as “universal” as early as 1976; see F. James Heindl, “Study of the Impact of the Global Positioning System on Army Survey,” DBA Systems, Inc. report ETL-0070, Sept 1976 (available from dtic.mil), 29. 4. Vernon I. Weihe, “World-Wide Coverage Radio Navigation for Aviation and Marine Services Using Land Based Facilities in the I.T.U. [100 KC-S] Band,” Navigation (US) 2, no. 7 (1950): 200– 201. See also R.C.T.A. specification for “ultimate global navigational aid” in Decca Navigator News 11 (July 1954): 3, and discussion of Navaglobe, Navarho, and Delrac in chapter 5. 5. Very little has been published about Soviet systems; basic descriptions and chronology are available from the Encyclopedia Astronautica, http://www.astronautix.com. The counterpart to Omega was known as Alpha (Альфа, also РСДН-20), which entered trials in 1972. The counterpart to Transit was Tsiklon (Циклон; “Cyclone”), which was first approved for development in 1962 and accepted for service in 1972. A more advanced version of Tsiklon known as Parus (Парус, “Sail”) was launched in the mid-1970s; it was matched by a civilian system known as Tsikada (Цикада, “Cicada”) that came online a few years later. 6. Pierce’s first postwar system was called Radux. This became Radux-Omega, then Omega. Pierce also worked on an analogous system called Draco. The main difference between Pierce’s systems and low-frequency (LF) Loran was that the former were phase-comparison systems,

364

Notes to Pages 249–258

while LF Loran, like all Loran variants, used pulses. For relevant VLF work, see J. A. Pierce, “The Navigational Uses of Low Radio Frequencies,” Navigation (US) 3 (Dec 1952): 198– 204; C. J. Casselman, D. P. Heritage, and M. L. Tibbals, “VLF Propagation Measurements for the RaduxOmega Navigation System,” Proceedings of the IRE (May 1959): 829– 839. 7. For a general evaluation, see Eric R. Swanson, “Omega,” Proceedings of the IEEE 71 (Oct 1983): 1140– 1155. For narratives of its development, see John A. Pierce, “J. A. Pierce and the Origin of Omega” (HUARC, HUG(B)-P461.4); John Alvin Pierce, “Memoirs of John Alvin Pierce,” pts. 1 and 2, Navigation (US) 36 (1989): 1– 8, 147– 155; John Alvin Pierce, “Technical Extracts from the Memoirs of Dr. J. A. Pierce,” pts. 1, 2, and 3, ed. Walter F. Blanchard, Journal of Navigation (UK) 55 (2002): 1– 22, 157– 183, 337– 362; Peter B. Morris et al., Omega Navigation System Course Book, vol. 1 (Alexandria, VA: US Department of Transportation, 1994; available at dtic.mil). Omega was similar enough to Decca’s Delrac proposal that Decca eventually won a patent-infringement case against the US government: “Decca Wins $40 Million in Court,” Flight International, 21 Apr 1979, 1222. 8. Quote from Krafft A. Ehricke (of Convair), in US House Select Committee on Astronautics and Space Exploration, The Next Ten Years in Space, 1959– 1969 (Washington DC: USGPO, 1959), 51; see also comments by John T. Hayward (Navy) in the same volume, 12, 55, 78. See also Alton B. Moody, “Use of Satellites for Navigation,” Navigation (US) 6 (Summer 1958): 95– 101; Paul D. Thomas, “From Landmarks to Satellites,” Military Engineer, no. 355 (Sept–Oct 1961): 337– 340; William H. Guier and George C. Weiffenbach, “Genesis of Satellite Navigation,” Johns Hopkins APL Technical Digest 19, no. 1 (1998): 15. 9. For general overviews, with accuracy data, see R. B. Kershner and R. R. Newton, “The Transit System,” Journal of the Institute of Navigation (UK) 15 (Apr 1962): 129– 144; Thomas A. Stansell Jr., “The Navy Navigation Satellite System: Description and Status,” Navigation (US) 15 (Fall 1968): 229– 243; T. A. Stansell Jr., “The Many Faces of Transit,” Navigation (US) 25 (Spring 1978): 55– 70; Helen Gavaghan, Something New under the Sun: Satellites and the Beginning of the Space Age (New York: Copernicus, 1998), 47– 126. The relentless increase in accuracy was due both to better determinations of the earth’s gravity field and to software improvements in Transit receivers. 10. Quote from Thomas A. Stansell Jr., “Transit: The Navy Navigation Satellite System,” Navigation (US) 18 (Spring 1971): 94. The original paper is William H. Guier and George C. Weiffenbach, “Theoretical Analysis of Doppler Radio Signals from Earth Satellites,” Nature 181 (31 May 1958): 1525– 1526. 11. Guier and Weiffenbach, “Genesis of Satellite Navigation,” 16. 12. With a suitably high-frequency signal, only a single small antenna was required, instead of the three-foot dish required for traditional direction finding or the periscope needed for an automatic star tracker. See Guier and Weiffenbach, “Genesis of Satellite Navigation,” 16; Bradford W. Parkinson, Thomas Stansell, Ronald Beard, and Konstantine Gromov, “A History of Satellite Navigation,” Navigation (US) 42, no. 1 (1995): 112– 114. 13. Pierce complained that navy administrators initially resisted engineering changes that would aid search-and-rescue operations; see Pierce, “Technical Extracts,” 343, 356. The first submarine receiver was only developed in 1967; see Morris, Omega Navigation System Course Book, p. 2-23. For British reaction to Omega, see “Notes on a V.L.F. Symposium Held at R.A.E. on Wednesday 7th April 1965” (PRO, AVIA 13/1429), 1. For low-accuracy use, see a navy engineer’s suggestion that it should not be relied upon as a primary means of navigation: “Notes on a U.S.-U.K. VLF Conference,” 25– 27 Nov, 1968 (PRO, AVIA 65/2100, “VLF Radio Navigation System: Project OMEGA”), p. 1-C. 14. For fix times in the early 1960s, see “Transit Proves System,” Science News-Letter 77 (11 Jun 1960): 375. Each one-knot error in speed would translate to about four hundred meters of uncertainty in location; see Stansell, “The Many Faces of Transit,” 56. 15. Eugene Erlich, “Current Developments in Navigation Satellites,” Navigation (US) 12 (Winter 1965– 1966): 330. 16. For the novelty of noncore uses as late as 1969, see Edward F. Gallagher, “Expanded Application of Navy Navigation Satellite System in Marine and Air Navigation,” Navigation (US)

Notes to Pages 258–261

365

16 (Spring 1969): 56– 60. For airborne applications in particular, see Benny R. Spicer, “Study of Aircraft Position Fixing Using the Navy Navigational Satellite System” (MS thesis, MIT, 1967; available from ntrs.nasa.gov). For receiver costs, see Dan Fales, “Use of Satellite Navigation Gains,” New York Times, 13 Jan 1980, S9. See also Stansell, “The Many Faces of Transit.” 17. For descriptions of antennas, see Pierce, “Technical Extracts,” 352; Swanson, “Omega,” 1141. 18. For signal variation, see, for example, Robert R. Morgan and Eugene L. Maxwell, “Omega Navigational System Conductivity Map,” ONR Report 54-F-1, Dec 1965 (available from dtic.mil); G. P. Asche, “Implementation Status of the Omega Navigation System,” Navigation (US) 19 (Summer 1972): 114. For differential techniques, see Swanson, “Omega,” 1146; Pierce, “Technical Extracts,” 355– 356; E. R. Swanson, D. J. Adrian, and P. H. Levine, “Differential Omega Navigation for the U.S. Coastal Confluence Region,” Navigation (US) 21 (Fall 1974): 264– 271; H. G. Miller, “Differential Omega in the Domestic Air Traffic Control Environment,” Navigation (US) 22 (Summer 1975): 165– 172; W. M. Hollister and S. M. Dodge, “An Evaluation of Differential Omega for General Aviation Area Navigation,” Navigation (US) 22 (Fall 1975): 259– 273. For the close links between differential techniques and civilian interest, see “Report on a Visit to the ION Symposium on Omega Held at Washington DC on 8 November 1971” (PRO, AVIA 65/2100, “VLF Radio Navigation System: Project OMEGA”). 19. For the importance of the early 1960s gravity mapping (by the so-called “Tranet” network), see Stansell, “Transit,” 100– 101. 20. In New Zealand, one pamphlet was titled “OMEGA: Nuclear Warfare Subsystem or International Navigational Aid?” See D. Tonkin to I. K. C. Ellison, “The Omega Controversy,” 21 May 1969 (PRO, AVIA 65/2100, “VLF Radio Navigation System: Project OMEGA”). In Trinidad, it was simply noted that “local politics are uncertain.” See “Omega Mission Report,” May 1970 (PRO, FCO 55/611, “Radio Sites for ‘Omega’ Navigation System of USA”). 21. For State Department concerns with any association between Omega and French nuclear sites in the South Pacific, see Pierce, “Technical Extracts,” 347– 349. For contingency of funding on international cooperation, see memo to C. C. H. Dunlop (British Navy Staff), “Information Exchange Project IEP B49: Omega Navigation System,” 28 Aug 1970 (PRO, FCO 55/611, “Radio Sites for ‘Omega’ Navigation System of USA”). 22. For UK involvement, see letter from Gerald G. Oplinger (US Embassy in London) to M. Gowlland (UK Foreign and Commonwealth Office), 16 Dec 1970 (PRO, FCO 55/611, “Radio Sites for ‘Omega’ Navigation System of USA”). The Liberian station was jointly operated by the Liberian government and a US contractor; see L. W. Campbell, T. M. Servaes, and E. R. Grassler, “North Atlantic Omega Navigation System Validation” (report to US Department of Transportation, 21 Jul 1980; available from dtic.mil), p. 2-3. 23. Namely, France, Japan, Argentina, Norway, Australia, and Liberia. Negotiations with these countries were not always smooth, and the bilateral agreements took much longer to secure than planned. See synopsis of a lecture by F. S. Stringer, “Omega— a World-Wide Navigation System,” 30 Jan 1969; and “Second Omega Mission Report,” Apr 1971, 1 (both in PRO, AVIA 65/2100, “VLF Radio Navigation System: Project OMEGA”). 24. Quotes from Stansell, “The Many Faces of Transit,” 60, 59. For the North Sea gas field, see N. A. G. Leppard, “Satellite Doppler Fixation and International Boundaries,” Philosophical Transactions of the Royal Society of London, Series A 294 (14 Jan 1980): 294. For particular angst from oil and gas users, see J. G. Morgan, “The Role of Navigation Satellites in Oil Exploration,” Navigation (US) 26 (Spring 1979): 37– 43. 25. Other corporate contacts included large companies like TRW, Philco, IBM, Cubic, Hughes Aircraft, and Westinghouse, as well as smaller companies like the Technology Audit Corporation and Page Communications. See, for example, the distribution list in Y. Morita, F. Zwas, and D. Colling, “Satellite Navigation Studies: Sixth Quarterly Progress Report,” Aug 1967 (available from ntrs.nasa.gov), 9– 11. 26. These agencies, along with NASA, were part of the Joint Navigation Satellite Committee (formed in 1964). See Alton ‘B’ Moody, “The Role of Satellites in Aircraft Navigation,” Journal of the Institute of Navigation (UK) 18 (Oct 1965): 507– 510; Panel 11 of the Summer Study

366

Notes to Pages 261–267

on Space Applications, National Research Council, Useful Applications of Earth-Oriented Satellites, vol. 11, Navigation and Traffic Control (Washington DC: National Academy of Sciences, 1969), 41. For the FAA’s Aerosat project, which came close to being launched in 1977 but was then canceled, see Nicholas Valéry, “Space Business Bridges the Atlantic,” New Scientist 68 (23 Oct 1975): 213; “Aerosat Postponed Again,” Flight International, 28 May 1977, 1507; “FAA ‘Can’t Afford Aerosat,’” Flight International, 1 Oct 1977, 969. For other FAA projects, see Keith D. McDonald, “A Survey of Satellite-Based Systems for Navigation, Position Surveillance, Traffic Control, and Collision Avoidance,” Navigation (US) 20 (Winter 1973– 1974); Bryant D. Elrod, “Aircraft Interrogation Scheduling with ASTRO-DABS,” IEEE Transactions on Aerospace and Electronic Systems 10 (Sept 1974): 595– 604. 27. I cannot pretend to synthesize all satellite work around the world. French geodetic satellites were used to test navigational concepts, but they were primarily dedicated to other projects and could not themselves be used for navigation. The French Argos project, launched in 1978, perhaps came closest, but this was primarily a tracking satellite for weather balloons and wildlife. For a sampling of French work, see “‘Dioscures’ pour le contrôle (tactique) des avions sur l’Atlantique Nord,” Air et Cosmos, 25 Nov 1967; Jean-Claude Trichet, “‘Dioscures’ satellites jumeaux proposés par le CNES et le SGAC pour la navigation aérienne,” Air et Cosmos, 2 Dec 1967, 16– 17; J. C. Husson and A. Banchereau, “Les satellites géodésiques français Diadème I et II: contribution de l’expérience à la navigation par satellites,” Centre National d’Études Spatiales (France), Report 1151-PR/PS, Sept 1967 (available from ntrs.nasa.gov); G. B. Rachet and J. L. Pieplu, “EOLE: Performances ultimes du système de localisation de balises fixes: Synthèse des résultats,” Centre Spatial de Bretigny, Division Mathématiques & Traitement, Departement Calcul d’Orbites, Report GB/73.364/CB/MT/OB, 1973 (available from ntrs.nasa.gov); A. Shaw and R. Rolland, “Marine Application of the ARGOS System,” in Proceedings OCEANS ’83 (Piscataway, NJ: IEEE, 1983), 262– 265. For German interest and projects, see, for example, H. C. Freiesleben, “Navigational Aid from Other Satellites,” Journal of the Institute of Navigation (UK) 15 (Apr 1962): 149– 154; H. Euler and G. Hoefgen, “Granas, a New Satellite-Based Navigation System,” Journal of the Institute of Navigation (UK) 37 (Sept 1984): 354– 359. For Japan, see Erlich, “Current Developments,” 338. For the UK, see F. G. Smith, “A New Navigation System Using Artificial Earth Satellites,” Journal of the Institute of Navigation (UK) 13 (Jan 1960): 109– 111. For the USSR, see C. D. Wood and G. E. Perry, “The Russian Satellite Navigation System,” Philosophical Transactions of the Royal Society of London, Series A 294, no. 1410 (14 Jan 1980): 307– 315. For involvement of various international organizations, see Panel 11, Useful Applications, 39– 40. 28. W. M. Hoover, “ATS-1/ATS-3 Dual Satellite Navigation Study,” Jan 1971 (report for NASA by Texas Instruments; available at ntrs.nasa.gov), p. 2-1. This referred to variation in basic system principles rather than specific proposals; there were slightly fewer of the latter. For a separate list of ten approaches and six specific engineering concepts, see Panel 11, Useful Applications, 56, 65. 29. One American engineer suggested that the navigational signal should be used mostly for “backup and gross error detection”—that is, for a redundant check on existing aids. See Moody, “The Role of Satellites,” 509. For a catalog of more than forty nonnavigational applications, see Panel 11, Useful Applications. See also Alfred E. Fiore and Paul Rosenberg, “Earth Satellite Systems for Marine and Transoceanic Air Navigation and Traffic Control,” Navigation (US) 17 (Fall 1970): 234– 245. 30. As a GE engineer put it, “a passive system offers merely another navigation aid in addition to the many useful aids that are already in existence”; see John E. Cutting, “Simple General Purpose Marine Navigation Using Navigation Satellite Data,” Navigation (US) 18 (Summer 1969): 174. 31. Again, there were many different ways this could be done. Some used range-range measurements (like Shoran); others used more exotic techniques like fan beams or interferometers. For examples, see Philco Corporation, “Fan Beam Navigation Satellite Study,” WDL Technical Report 2962, 13 July 1966 (available from ntrs.nasa.gov); G. S. Gopalapillai, G. T. Ruck, and A. G. Mourad, “Satellite Interferometer as an Advanced Navigation/Communication System,”

Notes to Pages 267–268

367

Navigation (US) 25 (Winter 1978– 1979): 405– 418. (Interferometers had been considered in the 1960s as well.) For lack of user input, see L. M. Keane, “Recent Progress in Navigation Satellites,” Navigation (US) 15 (Winter 1968– 1969): 415– 423. 32. For hybrid passive-active systems, see G. W. Casserly and L. D. Filkins, “The Potential Use of Satellites in Hyperbolic Position Finding,” Navigation (US) 13 (Winter 1966– 1667): 353– 366; Keane, “Recent Progress”; David D. Otten (of TRW), “Study of a Navigation and Traffic Control Technique Employing Satellites, Volume 1: Summary,” Dec 1967 (available from ntrs .nasa.gov), 20. As launched, OPLE could handle four thousand receivers; the full proposal called for three geosynchronous satellites collecting data from twelve thousand balloons. See Charles Laughlin, Gay Hilton, Roger Hollenbaugh, and Richard Lavigne, “Description of Experimental Omega Position Location Equipment (OPLE),” Goddard Space Flight Center report, Jan 1966 (available from ntrs.nasa.gov), 7; Charles Laughlin, Gay Hilton, and Richard Lavigne, “OPLE Experiment,” Goddard Space Flight Center report, June 1967 (available from ntrs.nasa.gov). 33. The first of NASA’s experimental satellites in the mid-1960s— the Applications Technology Satellites, of which six were eventually launched— responded specifically to objections from the DOD about the vulnerability of civilian designs for communications-only satellites. See Linda Neuman Ezell, NASA Historical Data Book, vol. 2, Programs and Projects, 1958– 1969 (Washington DC: NASA, 1988), 392. 34. Fiore and Rosenberg, “Earth Satellite Systems,” 238. This article refers only to geostationary (zero-inclination) orbits, but inclined and elliptical geosynchronous orbits were also proposed. The ground trace of noncircular geosynchronous orbits is a lopsided figure eight; they are often used for focused regional coverage, such as over the contiguous United States. 35. R. R. Bohannon, “An International Airline Views Navigation Satellites,” Navigation (US) 13 (Spring 1966): 23– 28; James T. Enzensperger, “The Views of Ship Operating Companies on Navigational Satellites,” Navigation (US) 13 (Spring 1966): 29– 30; Roy E. Anderson, “Satellite Navigation and Communication for Merchant Ships,” Navigation (US) 14 (Summer 1967): 127– 141; Panel 11, Useful Applications. 36. For one- and two-satellite systems, and comparison with Transit, see Panel 11, Useful Applications, 36, 65. Compare to Keane, “Recent Progress,” 418, which discusses the eighteenand sixteen-satellite proposals of RCA and TRW. These proposals, however, were still based on two- and three-satellite geosynchronous designs; see RCA, “Navigation|Traffic Control Satellite Mission Study,” Dec 1968 (available from ntrs.nasa.gov); or Otten (of TRW), “Study of a Navigation and Traffic Control Technique.” 37. Quote from Panel 11, Useful Applications, 79– 81; see also Fiore and Rosenberg, “Earth Satellite Systems,” 245. For US-only schemes, see E. S. Keats, “A Navigation System Using Distance and Direction Measurements from a Satellite,” Navigation (US) 11 (Autumn 1964): 335– 341; Elrod, “Aircraft Interrogation.” For international cost sharing, see the canceled Aerosat project (a Europe-US-Canada effort) or the successful (but functionally limited) French-US Argos system. Argos is still in operation today; see Rebecca Morelle, “Argos: Keeping Track of the Planet,” BBC News, 7 June 2007 (available at http://news.bbc.co.uk/2/hi/science/nature/6701221.stm). 38. For Omega integration, see Francis. J. Enge, “An OMEGA Receiver Navigation Set for High Performance Aircraft,” Navigation (US) 13 (Winter 1966– 1967): 343; “Hyperbolic WorldWide,” Flight International, 11 July 1968, 65– 66. For NASA integration, see Keane, “Recent Progress”; Herbert Winter, “Optimal and Suboptimal Methods of Satellite Surveillance for Traffic Control of Transoceanic Flights,” Navigation (US) 18 (Winter 1971– 1972): 417. For Transit, see Stansell, “Transit,” 105– 106; W. E. Warwick, “The System aboard Queen Elizabeth II,” Journal of the Institute of Navigation (UK) 23 (Oct 1970): 456– 457; Joseph Chernof, “Application of Satellite Navigation Techniques to Marine and Air Navigation,” Navigation (US) 16 (Summer 1969): 133– 140. 39. Scott Pace et al., The Global Positioning System: Assessing National Policies (Santa Monica, CA: RAND, 1995), 204: “The Global Positioning System is a simple idea that has some complex results. GPS satellites may be thought of as ‘clocks in space’ that broadcast a uniform time.”

368

Notes to Pages 269–272

40. Parkinson et al., “History of Satellite Navigation,” 126. For DOD impatience with multiplying systems and the ability of niche solutions to continue to receive funding based on their special characteristics, see Richard Easton quotation of Chester Kleczek, “Who Invented the Global Positioning System?,” Space Review, 22 May 2006 (available at http://www .thespacereview.com/article/626/1). As late as 1967, the DOD was not anticipating having a single system for all users; see Donald F. Spencer, “Navigation Satellites,” Navigation (US) 14 (Winter 1967– 1968): 380. 41. Timation was begun in 1964, following earlier work on time transfer for the US Naval Observatory. Navigation studies began at the Aerospace Corporation in 1963; the name “621B” dates from 1968. For elimination of other systems, see H. A. Cheilek and W. R. Seymour, Impact of Navstar Global Positioning System on Military Plans for Navigation and Position Fixing Systems, WSEG Report 289 (Arlington, VA: Institute for Defense Analyses, Oct 1975; available from dtic .mil), 5. 42. Quote from Roger Easton, “GPS Inventor,” 29 Mar 2005 (available at http:// gpsinventor.com/?page=memoir). See also Rick W. Sturdevant, “Tracing Connections— Vanguard to NAVSPASUR to GPS: An Interview with Roger Lee Easton, Sr.,” High Frontier 4 (May 2008): 52– 55. The first two Timation satellites used crystal oscillators, but later generations— launched after the merger with 621B, and rebranded Navigation Technology Satellites instead of Timation— used rubidium- and cesium-vapor clocks. See “Navigation Satellite Carries Atomic Clock,” Science News 112 (2 Jul 1977): 6. Easton received the National Medal of Technology in 2005 and was inducted into the National Inventors Hall of Fame in 2010. See NRL Press Release 34-10r, “Father of GPS and Pioneer of Satellite Telemetry and Timing Inducted into National Inventors Hall of Fame,” 31 Mar 2010 (available at http://www.nrl.navy.mil/media /news-releases/34-10r). See also Richard D. Easton (Roger Easton’s son), GPS Declassified: From Smart Bombs to Smartphones (Lincoln, NE: Potomac Books, 2013). 43. GPS uses a code division multiple access (CDMA) signal with pseudorandom noise (PRN) encoding; see Parkinson and Powers, “Origins of GPS,” 31. For the Aerospace contribution, see Bradford Parkinson interview by Steven R. Strom, Aerospace Headquarters, 1 Apr and 4 June, 2003 (available at http://www.aero.org/corporation/documents/ParkinsonInterview .pdf), 5. Parkinson, along with Ivan Getting, was awarded the Draper Prize in 2003 and was inducted into the National Inventors Hall of Fame in 2004. 44. Parkinson, for example, has consistently described Timation as a two-dimensional system and 621B as fundamentally three dimensional, but Timation proposals make it clear that three-dimensional applications were planned from the start. See Bradford W. Parkinson and Stephen W. Gilbert, “NAVSTAR: Global Positioning System— Ten Years Later,” Proceedings of the IEEE 71 (Oct 1983): 1177; compare to Naval Research Laboratory, “Timation Development Plan,” NRL Report 7227, 2 Mar 1971 (available from dtic.mil). Note, however, that even navy authors sometimes made this claim; see George Lowenstein, John Phanos, and Edward Rish, “Sole Means Navigation in U.S. Navy Aircraft,” IEEE Aerospace and Electronic Systems Magazine 3 (August 1988): 16. Parkinson has also protested that the precise orbits used in GPS, with a period of 11 hours and 58 minutes, were different from the 12-hour orbits originally considered by the navy. Both Easton and his son have likewise engaged in a long quest to show that none of the obvious features of GPS come from Project 621B, which is simply incorrect. See, for example, letters to the editor from Parkinson and Richard Easton in Inside GNSS 5 (May 2010): 12– 17. For more, see http://gpsinventor.com. 45. Brad Parkinson interview by Michael Geselowitz, Center for the History of Electrical Engineering, 2 Nov 1999 (available at http://www.ieeeghn.org/wiki/index.php/Oral-History: Brad_Parkinson). For an overall narrative of this process, see Pace et al., Global Positioning System, app. B. Parkinson’s account of the frantic redesign of 621B (which he describes taking place over Labor Day in the empty Pentagon) differs from Easton’s. Compare Parkinson and Powers, “Origins of GPS,” 38– 41, with Easton, “GPS Inventor.” 46. R. N. Parker (of DOD), “Radio Navigation Activities within the Department of Defense,” Navigation (US) 21 (Summer 1974 [orig. Nov 1973]): 113, 114.

Notes to Pages 273–274

369

47. Brad Parkinson, “True Origins and Major Original Challenges for GPS Success (1962– 1978)” (presentation at Stanford Center for Position, Navigation, and Time, Oct 2009; available at http://scpnt.stanford.edu/pnt/PNT09/presentation_slides/13_Parkinson_Creating _GPS.pdf), slide 26. 48. As of 1971, the preferred Timation constellation consisted of three planes of nine satellites in eight-hour orbits, but a wide variety of options were considered. See NRL, “Timation Development Plan,” 57. For a general description of 621B, see McDonald, “A Survey of SatelliteBased Systems” (n. 26 above), 310. Note, however, that the exact 621B configuration was constantly evolving; its leaders have objected in particular to oft-repeated claims that it would have placed the atomic clock on the ground and used the satellites only as transponders. See Parkinson and Powers, “Origins of GPS,” 36. On several occasions Parkinson has pointed to an Aerospace briefing that shows several concepts for satellite-based navigation, including one that would work much like GPS, though with a different constellation: J. B. Woodford and H. Nakamura, “Briefing— Navigation Satellite Study,” Aerospace Corporation report TOR-1001(2525-17)-1, 24 August 1966 (held by Aerospace Corporation library, El Segundo, CA; published as an appendix to Air Force Center for Systems Engineering [AFIT/SY], Global Positioning System Systems Engineering Case Study, 4 Oct 2007, available at http://www.afit.edu /cse/casedocs/files/GPS%20SE%20Study%20Final%20Report%20_4%20Oct%2007.pdf). 49. For example, J. A. Buisson and T. B. McCaskill, “Timation Navigation Satellite System Constellation Study,” NRL Report 7389, 27 June 1972 (available from dtic.mil), 2. 50. For the cost logic of 621B, see Parkinson and Powers, “Origins of GPS,” 36; part of the reason for the rejection of 621B in August 1973 was its price. The final approved phase-one funding for GPS in December 1973 was $104 million. The predicted cost of the Timation space segment (in 1971) was $200 million— or $245 million in 1973 dollars (deflated as a relative share of US GDP). The predicted cost of the proto-621B project at Aerospace (in 1964) was $64 million for the regional project and $111 million for the global version— or $133 and $231 million in 1973 dollars. See NRL, “Timation Development Plan,” 3; Woodford and Nakamura, “Briefing— Navigation Satellite Study,” slide 75; Air Force Center for Systems Engineering, Global Positioning System Systems Engineering Case Study, 26. 51. Quote from Parkinson and Gilbert, “NAVSTAR Ten Years Later,” 1177. For the USSR cluster, see Richard Easton citing Harry Sonneman (chairman of the Navigation Steering Committee), in “Who Invented the Global Positioning System?” 52. The initial approval of the program endorsed the “highest cost and lowest risk” of three alternatives presented; see Jeffrey A. Drezner and Giles K. Smith, An Analysis of Weapon System Acquisition Schedules (Santa Monica, CA: RAND, 1990), 180. 53. This also caused $66 million to be transferred from the navy to GPS; see Parkinson and Powers, “Fighting to Survive,” 15. See also Drezner and Smith, Weapon System Acquisition Schedules, 180– 184. For details of the nuclear-testing equipment, see Pace et al., Global Positioning System, 241– 242. 54. Parkinson and Powers, “Fighting to Survive,” 17. 55. Even with eighteen satellites, there would be periodic outages. See P. S. Jorgensen, “Navstar/Global Positioning System 18-Satellite Constellations,” Navigation (US) 27 (Summer 1980): 89– 100. 56. Drezner and Smith, Weapon System Acquisition Schedules, 184– 188. Parkinson remembers GPS coming “very, very close to termination on a number of occasions”; see Parkinson interview by Strom, 11. 57. “RCA to Build New USN Navsat,” Flight International, 1 Apr 1978, 939. 58. This was largely a question of signal design. GPS satellites were designed to transmit signals on two frequencies at once, with one encrypted for military use. Not only could the civilian signal be turned off in wartime, but access to both signals would also enable direct correction of the effects of the ionosphere, thereby giving authorized users an accuracy advantage. See Parkinson and Powers, “Origins of GPS,” 40. 59. Parker, “Radionavigation Activities,” 116; NRL, “Timation Development Plan,” 42.

370

Notes to Pages 274–276

60. For negotiations with Congress, see Geselowitz interview of Parkinson. For the civilian market reducing overall system cost, see Cheilek and Seymour, Impact of Navstar (n. 41 above), 15; “Navigation Satellite Carries Atomic Clock” (n. 42 above). For user fees, see Thomas A. Stansell Jr., “Civil GPS from a Future Perspective,” Proceedings of the IEEE 71 (Oct 1983): 1191. 61. Parkinson and Gilbert, “NAVSTAR Ten Years Later,” 1185– 1186. For civilian response, see Eliot Marshall, “Steering Clear of Sakhalin,” Science 222, no. 4621 (21 Oct 1983): 303– 304; W. J. Klepczynski and L. G. Charron, “The Civil GPS Service,” in Proceedings of the Twentieth Annual Precise Time and Time Interval (PTTI) Applications and Planning Meeting, Vienna, VA, 29 Nov–1 Dec 1988 (available from dtic.mil). 62. For conflicting press reports, see Mike Hirst, “Airborne Navigation Systems,” Flight International, 14 Jan 1978, 113; “Third NavStar Prototype Launched,” Flight International, 4 Nov 1978, 1689; “Nato Studies Navstar GPS,” Flight International, 29 Sept 1979, 1034. 63. Panel 11, Useful Applications (n. 26 above), 38; Parkinson and Powers, “Fighting to Survive,” 16. 64. “Satellites to Guide Air Traffic Backed,” New York Times, 2 Apr 1979, D7; “Department of Transportation National Plan for Navigation Executive Summary,” Wild Goose Association Radionavigation Journal (1978): 52. For civilian enthusiasm, see T. A. Stansell Jr., “Civil Marine Applications of the Global Positioning System,” Navigation (US) 25 (Summer 1978): 224– 235; Fales, “Use of Satellite Navigation Gains” (n. 16 above). 65. Larry Hogle, Kelly Markin, and Jerry W. Bradley, “Navigation Systems Performance versus Civil Aviation Requirements,” Proceedings of the IEEE 71 (Oct 1983): 1212. 66. Marshall, “Steering Clear of Sakhalin”; Richard Witkin, “F.A.A. Chief Says Jet Was beyond U.S. Radar,” New York Times, 19 Sept 1983, A7. 67. For details about restrictions on DOD management in the early 1990s, see Jonathan M. Epstein, “Global Positioning System (GPS): Defining the Legal Issues of Its Expanding Civil Use,” Journal of Air Law and Commerce 61 (1995): 258. The RAND study is Pace et al., Global Positioning System, 189. For Selective Availability, see Bill Clinton, Presidential Decision Directive NSTC-6, 28 Mar 1996 (available at http://www.fas.org/spp/military/docops/national/gps .htm); for discussion, see Jonathan M. Epstein, “The Role of the Global Positioning System in the Environment,” N.Y.U. Environmental Law Journal 6 (1997): 77– 78. 68. Parkinson et al., “History of Satellite Navigation,” 139. See again fig. 6.14. 69. Specifically the Casio Pathfinder, first introduced at the January 1999 Consumer Electronics Show. For a detailed description, see Sam Evans, “Casio GPS Pathfinder Watch,” 21 Sept 2000 (available at http://www.geek.com/hwswrev/conel/gpswatch). For price, see Sameer Kumar and Kevin B. Moore, “The Evolution of Global Positioning System (GPS) Technology,” Journal of Science Education and Technology 11 (Mar 2002): 69. 70. Stansell, “Civil GPS,” 1192. 71. Pace et al., Global Positioning System, 248– 249. The specification was issued by the National Oceanographic and Atmospheric Administration, successor to the Coast and Geodetic Survey. 72. Integration of horizontal and vertical, linear and angular, was also performed by “total station” equipment, which first appeared in the 1970s. Total stations only began to be integrated with GPS in the late 1980s. See P. Lloret, “Inertial + Total Station + GPS for High Productivity Surveying,” in Proceedings of IEEE Position Location and Navigation Symposium (PLANS 1990), 338– 346. Today total-station and GPS methods are the two main varieties of surveying, for high-precision and routine work, respectively. 73. For this earlier history, see chapter 4. For the 1980s, see Defense Mapping Agency, Geodesy for the Layman, 5th ed. (Washington DC: DMA, 1983), 35– 36. 74. There is an important distinction between the WGS 84 datum, which locates a GPS reading precisely on the earth, and the way that these coordinates are written mathematically. The GPS satellites themselves use earth-centered, earth-fixed Cartesian coordinates: the origin is at the earth’s center of mass, the X axis points to 0°E, 0°N, the Y axis points to 90°E, 0°N, and the Z axis runs through the North Pole. These same values can easily be translated

Notes to Pages 276–281

371

into latitude and longitude, or UTM, but it is often quite difficult to convert to other datums, which are not just mathematically different but are also aligned to the earth based on historical astronomical or trigonometric measurements. This was the problem that led to datum consolidation in the first place. 75. Transit, which began to use WGS 72 in 1975, was used to create surveying datums in Guadeloupe, Mauritius, Vanuatu, Samoa, Fiji, the Seychelles, Tuvalu, Macau, and the South China Sea; see Clifford Mugnier’s Grids and Datums articles for each country in Photogrammetric Engineering & Remote Sensing (1998– 2013). WGS 72 was also used to connect surveys in equatorial Africa and South America. See also Stansell, “Transit” (n. 10 above), 105. Transit was also the first practical manifestation of WGS 84 (in 1987); GPS followed soon thereafter. What ultimately ties the satellites to the earth is the coordinates of the satellite monitoring stations. See National Imagery and Mapping Agency, Department of Defense World Geodetic System 1984: Its Definition and Relationships with Local Geodetic Systems, 3rd ed., NIMA TR8350.2 (Washington DC: NIMA, 2000), pp. 2-2 to 2-5. 76. Ronald F. Abler, “Everything in Its Place: GPS, GIS, and Geography in the 1990s,” Professional Geographer 45 (May 1993): 137 (emphasis in original), trees on 136. As a nondigital precedent, Abler cites George Jenks, “‘Pointillism’ as a Cartographic Technique,” Professional Geographer 5 (Sept 1953): 4– 6. Of course, electronic data and GIS are not in fact transparent or unproblematic; see John Pickles, ed., Ground Truth: The Social Implications of Geographic Information Systems (New York: Guilford, 1994); Michael Curry, Digital Places: Living with Geographic Information Technologies (London: Routledge, 1998). 77. Bradford W. Parkinson, “Introduction and Heritage of NAVSTAR, the Global Positioning System,” in Global Positioning System: Theory and Applications, vol. 1, ed. Bradford W. Parkinson and James J. Spilker Jr. (Washington DC: American Institute of Aeronautics and Astronautics, 1996 [orig. 1994]), 24. GPS was used for locating minefields in the Persian Gulf during the 1987– 1988 war and for correcting maps during the 1989 intervention in Panama. See Pace et al., Global Positioning System, 245. 78. For civilian deaths with smart weaponry, see Neta C. Crawford, “Just War Theory and the U.S. Counterterror War,” Perspectives on Politics 1 (Mar 2003): 18. For popular fascination with smart weapons, see Caren Kaplan, “Precision Targets: GPS and the Militarization of U.S. Consumer Identity,” American Quarterly 58 (2006). For contemporary reports, see, for example, Vincent Kiernan, “Guidance from Above in the Gulf War,” Science 251, no. 4997 (1 Mar 1991): 1012– 1014; Andrew Pollack, “War Spurs Navigation by Satellite,” New York Times, 6 Feb 1991, D1. 79. For the “Left Hook”—also called the “Hail Mary”—see Michael Russell Rip and James M. Hasik, The Precision Revolution: GPS and the Future of Aerial Warfare (Annapolis, MD: Naval Institute Press, 2002), 126– 131. 80. Michael Russel Rip and David P. Lusch, “The Precision Revolution: The Navstar Global Positioning System in the Second Gulf War,” Intelligence and National Security 9 (Apr 1994): 167– 241; Rip and Hasik, Precision Revolution. 81. Barry R. Posen, “Command of the Commons: The Military Foundation of U.S. Hegemony,” International Security 28 (Summer 2003): 5– 46. 82. Quoted in “What the Russians Learned from the Gulf War,” Aviation Week & Space Technology 137 (5 Oct 1992): 78. 83. Pace et al., Global Positioning System, 250– 251; Andrew Pollack, “In U.S. Technology, a Gap between Arms and VCRs,” New York Times, 4 Mar 1991, A1. 84. There is some controversy about who first developed a usable in-car navigation system. Early systems used a dead-reckoning computer rather than satellites. The first GPS navigation equipment available for purchase debuted around 1990. For a rambunctious list of conflicting claims from various manufacturers’ websites, see http://en.wikipedia.org/wiki/Automotive _navigation_system#History. 85. For FAA, see Pace et al., Global Positioning System, 249. For early discussions at ICAO, see the records of the Future Air Navigation Systems conferences, which first met in 1984 based on

372

Notes to Pages 281–283

earlier discussions in 1982 and 1983 (bound volumes, ICAO archives). For eventual standards, see ICAO annex 10. 86. For experimental use on the Burlington Railroad, see Malcolm W. Browne, “Worldwide Satellite Network Promises Great Accuracy in Navigation,” New York Times, 8 Nov 1988, C1; for operational use of the same system, see Patrick T. Harker, “Services and Technology: Reengineering the Railroads,” Interfaces 25 (May–June 1995): 72– 80. 87. For early discussion of tectonic-plate monitoring, see Richard A. Kerr, “New Technology Aids Geophysicists,” Science 227, no. 4687 (8 Feb 1985): 619– 620; William H. Prescott, James L. Davis, and Jerry L. Svarc, “Global Positioning System Measurements for Crustal Deformation: Precision and Accuracy,” Science 244, no. 4910 (16 Jun 1989): 1337– 1340. For later environmental uses, see Epstein, “The Role of the Global Positioning System in the Environment.” For time, see Klepczynski and Charron, “The Civil GPS Service,” 53. For later uses— especially in time division multiple access (TDMA) communications traffic (an alternative to the CDMA used in GPS)— see Pace et al., Global Positioning System, 99– 102. For stoplights, see Chun Hau Thao and Szu Chin Chang, “Traffic Signal Control System Employing Universal Co-ordinated Time (UTC) of GPS as Time Base,” US Patent 6,847,307, 25 Jan 2005 (filed 29 May 2002). 88. For an extensive discussion of various uses, see Kumar and Moore, “The Evolution of Global Positioning (GPS) Technology” (n. 69 above). For other lists of predicted uses, see W. E. Ramsey and J. R. Page, “Navigation Satellites: Their Future Potential,” Philosophical Transactions of the Royal Society of London, Series A 312 (26 Jul 1984): 67– 73; Buckley, “Precision Sailing” (n. 3 above). 89. Parkinson, “Introduction and Heritage,” 3; B. W. Parkinson, “Overview,” Navigation (US) 25 (Summer 1978): 94; Parkinson and Powers, “Origins of GPS,” 30. 90. David Braunschvig, Richard L. Garwin, and Jeremy C. Marwell, “Space Diplomacy,” Foreign Affairs 82 (July–Aug 2003): 157; Joan Johnson-Freese, “Getting Our Bearings,” YaleGlobal Online Magazine, 23 May 2003 (available at http://yaleglobal.yale.edu/print/741). 91. Epstein, “Expanding Civil Use” (see n. 67 above), 256. For emphasis on military origins, see prefatory note from the editors in Crosslink 3 (Summer 2002): 1. 92. Laura Kurgan, “You Are Here: Information Drift,” Assemblage 25 (Dec 1994): 17, 24; Clinton, Presidential Decision Directive NSTC-6 (n. 67 above); Kumar and Moore, “The Evolution of Global Positioning System (GPS) Technology,” 69; Alice A. Wong and Ray E. Clore, “Promoting International Civil GNSS Cooperation through Diplomacy,” High Frontier 4 (May 2008): 25. 93. Richard W. Stoffle, David B. Halmo, Thomas W. Wagner, and Joseph J. Luczkovich, “Reefs from Space: Satellite Imagery, Marine Ecology, and Ethnography in the Dominican Republic,” Human Ecology 22 (Sept 1994): 373. 94. Claudio Aporta and Eric Higgs, “Satellite Culture: Global Positioning Systems, Inuit Wayfinding, and the Need for a New Account of Technology,” Current Anthropology 46 (Dec 2005): 729– 753. 95. Pace et al., Global Positioning System, 97. 96. Tracking of humans began in the late 1990s for convicts and other law enforcement activities. Tracking was also used for commercial vehicle fleet management. In the early 2000s the FCC issued a requirement that 95 percent of cell phones be location aware by the end of 2005. By 2006, cell-phone providers were selling tracking services to monitor family members. By the end of the decade, schoolchildren and the elderly were also being tracked. The legal literature on tracking is, understandably, quite large. See in particular A. Michael Froomkin, “The Death of Privacy?,” Stanford Law Review 52 (May 2000): 1496– 1498; Aaron Renenger, “Satellite Tracking and the Right to Privacy,” Hastings Law Journal 53 (Jan 2002): 549– 565; Paul M. Schwartz, “Property, Privacy, and Personal Data,” Harvard Law Review 117 (May 2004): 2056– 2128; Tracy J. Hasper and Gordon F. Lull, “Where on Earth: GPS Tracking Devices Raise Fourth Amendment Issues for Civil and Criminal Law Practitioners Alike,” Los Angeles Lawyer (Sept 2009): 30– 37; United States v. Jones, 132 S. Ct. 945 (2012).

Notes to Pages 283–286

373

97. First quote from Kaplan, “Precision Targets” (n. 78 above), 708. Second from Mark Palmer and Robert Rundstrom response to Aporta and Higgs, “Satellite Culture,” 748. 98. Chirac cited in Braunschvig, Garwin, and Marwell, “Space Diplomacy,” 160. For more discussion of European fears of US militarism, see Sang Wook Daniel Han, “Global Administrative Law: Global Governance of the Global Positioning System and Galileo,” ILSA Journal of International and Comparative Law 14 (2008): 571– 593. 99. Brian Klinkenberg, “Geospatial Technologies and the Geographies of Hope and Fear,” Annals of the Association of American Geographers 97, no. 2 (2007): 350– 360; Ingrid Wiesel, “Predicting the Influence of Land Development on Brown Hyena (Parahyaena brunnea) Movement and Activity” (poster at Society of Conservation Biology Conference, July 2007, Port Elizabeth, South Africa); Erica Hagen, “Putting Nairobi’s Slums on the Map,” Development Outreach 12 (July 2010): 41– 43. 100. Jerome E. Dobson and Peter F. Fisher, “Geoslavery,” IEEE Technology and Society Magazine 22 (Spring 2003): 47– 52; Jerome E. Dobson and Peter F. Fisher, “The Panopticon’s Changing Geography,” Geographical Review 97 (July 2007): 307– 323; William A. Herbert, “No Direction Home: Will the Law Keep Pace with Human Tracking Technology to Protect Individual Privacy and Stop Geoslavery?,” I/S: A Journal of Law and Policy 2, no. 2 (2006): 409– 473. 101. Alternately, arguing that activist mapping and targeted marketing are part of a pervasive military culture stretches the definition of military culture past its breaking point. I am reminded of some of the criticism of Foucault’s carceral society: in its purest form, Foucault’s model is not just analytically unhelpful, but self-refuting. See Marcel Gauchet and Gladys Swain, Madness and Democracy: The Modern Psychiatric Universe, trans. Catherine Porter (Princeton: Princeton University Press, 1999 [orig. 1980]). 102. Commercial power has been especially emphasized by geographers: see Matthew James Kelley, “The Emergent Urban Imaginaries of Geosocial Media,” GeoJournal 78 (2011): 181– 203; Matthew Wilson, “Location-Based Services, Conspicuous Mobility, and the LocationAware Future,” Geoforum 43 (2012): 1266– 1275; Mark Graham, Matthew Zook, and Andrew Boulton, “Augmented Reality in Urban Places: Contested Content and the Duplicity of Code,” Transactions of the Institute of British Geographers 38 (2013): 464– 479; Jim Thatcher, “Avoiding the Ghetto through Hope and Fear: An Analysis of Immanent Technology Using Ideal Types,” GeoJournal 78 (2013): 967– 980. 103. Steven Lazar, “Modernization and the Move to GPS III,” Crosslink 3 (Summer 2002): 42– 46. 104. Parkinson began teaching at Stanford in 1984. See Parkinson interview by Strom (n. 43 above), 11. 105. Some GPS pundits predict that these various augmentation systems may one day merge to create a hyperaccurate, hyperreliable global whole, but this seems far from certain; see Todd Walter, Juan Blanch, and Per Enge, “Future Augmented: Coverage Improvement for Dual-Frequency SBAS,” GPS World 21 (Mar 2010): 36– 41. For indoor tracking, see Charles R. Trimble, “Phantom Menace,” Foreign Affairs 82 (Sept–Oct 2003): 194. Paying for higher accuracy is the basic idea behind commercial systems like John Deere’s Starfire, but paid services are also planned for Galileo. 106. For promises to ICAO since 1993, see Pace et al., Global Positioning System, 248. For expansion of the ground-station network, see C. H. Yinger, W. A. Feess, V. Nuth, and R. N. Haddad, “GPS Accuracy versus Number of NIMA Stations,” in ION GPS/GNSS 2003: Proceedings of the 16th International Technical Meeting of the Satellite Division of the Institute of Navigation, September 9– 12, 2003, Portland, Oregon (Fairfax, VA: Institute of Navigation, 2003); Harold W. Martin III and Walter Petrofski, “Controlling GPS: Architecture Evolution Plan and Beyond,” High Frontier 4 (May 2008): 38– 41. 107. The GNSS acronym was used in early ICAO discussions to avoid prejudice for any existing system; the use has spread quite widely since. For ICAO, see ICAO news release, “Global Implementation Plan for Satellite-Based Air Traffic Management Completed,” PIO 11/93, Oct 1993 (available at http://www.icao.int/icao/en/nr/1993/). For later use, see, for

374

Notes to Pages 287–292

example, Epstein, “Expanding Civil Use” (n. 67 above), 271– 283; Langhorne Bond, “The GNSS Safety and Sovereignty Convention of 2000 AD,” Journal of Air Law and Commerce 65 (2000): 445– 451; Rosalind Lewis, Michael Kennedy, Elham Ghashghai, and Gordon Bitko, Building a Multinational Global Navigation Satellite System (Santa Monica, CA: RAND, 2005); Han, “Global Administrative Law.” 108. The US has issued many policy statements attesting to its commitment to GPS. See also John M. Beukers, “Global Radionavigation: The Next 50 Years and Beyond” (paper presented at the NAV99/ILA28 Conference, Nov 1999; available at http://www.loran.org /Meetings/Meeting1999/pdffiles/papers/013.pdf).

Conclusion 1. This is not meant to minimize the significance of, say, the worldwide navigational charts of nineteenth-century Britain; rather, the point is that GPS organizes far more activities than earlier state-sponsored projects. The point is also not to present the link between mapping and militarism as novel; indeed, this should be perfectly obvious, perhaps even banal. See the classic polemic by Yves Lacoste, La géographie, ça sert, d’abord, à faire la guerre (Paris: F. Maspero, 1976). 2. The parallel between technology and power here is quite intentional. In both cases, there is a temptation to criticize their mere existence, as if technology or power were immediately synonymous with oppression. But both technology and power are relationships, not substances, and one could no sooner have a world without technology (or power) than one without social structure itself. 3. In Marx’s version, “Men may make their own history.” See The Eighteenth Brumaire of Louis Napoleon (1852), chap. 1. 4. “Force multiplier” can refer to military assets besides GPS; the term first appeared during World War II and became common after the mid-1970s. For GPS in the early 1990s, see GAO, “Global Positioning System: Production Should Be Limited Until Receiver Reliability Problems Are Resolved,” 20 Mar 1991 (available at dtic.mil), 4; Bradford W. Parkinson, Thomas Stansell, Ronald Beard, and Konstantine Gromov, “A History of Satellite Navigation,” Navigation (US) 42, no. 1 (1995): 152. 5. Even professional geographers relying on preexisting base maps would still spend years creating a single map sheet; see, for example, the map supplements included in the Annals of the Association of American Geographers from 1960 through the 1980s. 6. For example, see Jeff Robbins, “GPS Navigation . . . but What Is It Doing to Us?,” in Proceedings of the 2010 IEEE International Symposium on Technology and Society, ed. Katina Michael (Piscataway, NJ: IEEE, 2010). 7. Cartographically, these are included today under the umbrella of “thematic” mapping. See Susan Schulten, Mapping the Nation (Chicago: University of Chicago Press, 2012). Politically, these are about population in the Foucauldian sense; for these considerations in the context of territory, see Stuart Elden, “How Should We Do the History of Territory?,” Territory, Politics, Governance 1 (2013). 8. Most of my own experimentation has been solidly representational; see, for example, my website, www.radicalcartography.net. For other examples that engage more with GPS and spatial practice, see Michael Kimmelman, “G.P.S. Art,” New York Times, 14 Dec 2003; Nato Thompson, ed., Experimental Geography (New York: iCI, 2008); Laura Kurgan, Close Up at a Distance: Mapping, Technology, and Politics (New York: Zone, 2013). For the intersection of GPS tracking, consumer technology, and shifting jurisprudence in the US, see again chap. 6, nn. 96, 102.

Notes to Pages 293–298

375

I ND E X

Page numbers in italic refer to figures. Abler, Ronald, 281 Abyssinia, 319n44. See also Ethiopia Adams, Cyrus, 36 Adams, Oscar, 140, 141, 143–44, 339n52 aerial photography, 83, 85, 221, 231, 336n13 Aerosat satellite system, 367n26, 368n37 Aerospace Corporation, 274, 369n41, 370n48, 370n50 Africa: airway schemes and, 210–11; American antennas for radionavigation and, 264, 264, 266; cadastral coordinate systems and, 160, 343n98; Cape-to-Cairo line and, 211; cartographic partition of, 39; colonial possessions in, 158, 159, 160; distribution of IMW mapmaking responsibility and, 38–39, 39, 319n41, 331n77; ellipsoids and, 194, 194, 195; grid systems and, 132, 157–61, 159, 342nn93–94; missing sheets from IMW and, 67; provisionality in IMW maps and, 60, 325n100; US-offered mass calculations for, 192; UTM and, 169, 183, 184, 189–90, 192, 196, 344n9 Agnew, John, 13 AGS. See American Geographical Society (AGS) Alaska, 72, 76, 83, 191, 224, 243, 245, 348n65 Albania, 189, 350n85 Algeria, 224, 330n65 Alpha (Альфа) radionavigation, 364n5 Althusser, Louis, 321n65 American Airlines, 77 American Geographical Society (AGS): archival imperialism of, 58; Latin America sheets for the US Army Air Force and, 83;

Map of Hispanic America and, 57–62, 57, 58, 62, 63, 324n96, 325n98; Mexico map by, Gallery 6, 82; provisional mapping and, 324n88, 324n96; relative reliability diagrams and, 60, 60, 62; World Land Use Survey and, 331n75 AMS. See US Army Map Service (AMS) Antarctica: Antarctic Treaty (1959) and, 8; IMW and, 65, 67; missing from world maps, 77, 79; US-published map of, 325n2 Arctic Ocean, 9, 10, 312n20 Arctic region, 69–70 Argentina, 45, 264, 291, 318n27, 348–49n70, 366n23 Argos satellite system, 367n27, 368n37 artillery targeting. See military targeting Ascension, 292 Asia: central Asia as geographical pivot area and, 326n9; distribution of IMW mapmaking responsibility and, 319n41, 319n46; UTM and, 183, 184 AT&T, 235 Australia: distribution of mapmaking responsibility and, 86, 88, 319n41, 331n77; ellipsoids and, 195; IMW dues and, 89; negotiations with US over radionavigation and, 366n23; no maps for unsurveyed terrain and, 60; postwar IMW work and, 112; postwar proliferation of navigation technologies and, 244; radionavigation stations in, 224, 242, 264, 291; UTM and, 350n82 Austria, 134, 147, 340n63, 344n15, 355n28

377

aviation: air-age maps and, 73–74, 77, 79, 88–89; airmail and, 213; air traffic control and, 249, 268, 363n101; airways as highways and rivers and, 352n6; depiction of long-range airways and, 78, 79; digital and analog computers and, 246, 362n89; families of charts and, 103, 105; in-flight briefing systems and, 246, 248; flying “on the beam” and, 211, 353n10; freedom of the air in international law and, 77, 327n19; grids on maps and, 346n37; ground installations supporting, 210, 352n6; IMW and, 92–93; indicators of elevation and, 106; infrastructure versus superstructure and, 352n7; Jet Navigation Chart and, 105; landing systems and, 361n78; level of information pilots require and, 103–5, 104; map-display systems and, 246, 247, 248; mileage for comparative purposes in, 353n9; mismatches between adjacent national triangulation networks and, 221; modernization of GPS satellites and, 290; Operational Navigation Chart, 105; pilot autonomy versus air traffic efficiency and, 248, 354n21; radionavigation and, 210–13, 215–18; railroads of the air and, 210–11, 211, 212–13, 214; rapid innovation in mapmaking and, 102–3; reporting norms in, 354n19; roller maps and, 247, 363n94; sun and stars for navigation and, 174; WAC as map for pilots and, 89, 99, 105, 106; women excluded from, 354n18. See also International Civil Aviation Organization (ICAO); International Commission for Air Navigation (ICAN); World Aeronautical Chart (WAC) Bahamas, the, 242–43 Baltic Sea, 235 Barthes, Roland, 13 Bartholomew, John George, 318n25 BeiDou Navigation Satellite System. See Compass satellite system Belgium: Belgium-Germany border and, 350n89; grid systems and, 134, 139, 337n23; International Ellipsoid and, 151, 348n67; triangulation of, 153 Bengal, 311n17 Berthaut, Henri, 32, 318n23, 319n46 Boeing Company, 244, 267 Boelcke, Siegfried, 132 Bolivia, 61, 343n1 Bonsdorff, Ilmari, 156, 342n86 borders: AGS Map of Hispanic America in boundary disputes and, 61; boundary crossing versus boundary fixing and, 251; cartographic demarcation of elevation

378

Index

and, 45; debordering and rebordering and, 13, 14; GPS and, 285, 296; grid systems and, 145, 162; maritime, 14, 235, 251; 1919 Paris Peace Conference and, 27; radionavigation and, 209; rectangular coordinates for stabilization of, 350n89; representational mapping and, 15; state-centered border regimes and, 13; survey discrepancies along, 130, 154, 172; symbols for indefinite or disputed, 45; US-Mexico, 145; UTM and, 177–78, 197; World War II and, 167 Borges, Jorge Luis, 123 boundaries. See borders Bowie, William: Bowie method and, 156, 157, 342n89; International Ellipsoid and, 154; linear model of, 148; recalculation of the European triangulation and, 155–58, 185–88, 187, 191, 347n53; on two phases of geodesy, 147 Bowman, Isaiah, 59, 62, 64 Boxer Rebellion, 318n23 Braden, Spruille, 62 Brazil: AMS surveying and, 191; distribution of ICAO mapmaking responsibility and, 331n84; IMW and, 56–57, 58, 110, 324n90; provisional mapping and, 324n96; radionavigation stations in, 291; radiosurveying and, 234 Bright, Charles, 12 British Guiana, 232 British Petroleum Company, 234 Bruchmüller, Georg, 337n29 Brückner, Eduard, 319n37 Bulgaria, 161, 291 Burnett, D. Graham, 145 Cambodia, 192, 330n72 Canada: Aerosat project and, 368n37; distribution of mapmaking responsibility and, 86, 319n41, 331n77; grid systems and, 150, 341n71, 351n95; Inuit use of GPS in, 285; no maps for unsurveyed terrain and, 60; North American triangulation systems and, 154; postwar IMW work and, 112; radionavigation stations in, 224, 242–43, 245, 291; radiosurveying and, 232, 234; transcontinental airways and, 210; UTM and, 176, 196 Carroll, Lewis, 123 carte de France (Carte de Cassini), 127–30, 128, 129, 340n67 carte de l’État major, 125–27, 126, 127, 130, 340n67 Carte du ciel, 29 Carte Internationale du Tapis Végétal, 94–96, 330n65, 330n73 cartography: art versus science and, 35; authoritative representation and, 26,

27, 28–29; base-map ideal and, 26–27, 48–49, 94, 100, 102, 110–15; as branch of mapping sciences, 16; cartographic literacy and, 1; cartographic one-upmanship and, 32; cartographic perfection and, 60; cartographic virtues of points and, 281–82; construction versus reflection of reality and, 13; critical paradigm in, 68–69, 115–16, 334–35n134; depicting relief and, 35; failed goal of territorial unity in, 190; fragmentation of French cartography and, 149, 340n67; general versus specialist readers and, 102, 103–5, 107; general versus specific maps and, 92; Greenwich meridian and, 29, 33, 317–18n22; historical roots of, 311n16; historiographical shift in, 114, 334–35n134; ideology and, 321n65; IMW and new era in, 26, 35; international cartographic associations and, 317n21; international endorsement and, 38; limitations of, 310–11n14; making and remaking of territory and, 298–99; map as framework of points and, 123; map design research and, 102–3, 107, 332n89; mapmaking production speed and, 83; merger of mapmaking and navigation and, 257; neutrality and, 41–42, 48, 67, 80, 93, 114, 320n51; 1950s crisis of representation in, 14; number of international mapping projects and, 49, 321n69; Paris meridian and, 32, 317–18n22; paternalism and, 46; pilots as research subjects for, 102–3; provisional mapping and, 55–56, 59–61; relative reliability diagrams and, 60, 60; retreat from worldwide mapping and, 94–96, 99–100, 102; as a science, 102–3, 113; state sponsorship and, 48; surveys and, 323–24n86; territoriality and, 6–7, 113–16; thematic maps and, 49, 321n67, 375n7; view-fromabove experience of, 18; World War II as geographically transformative and, 69–70, 79, 88–89, 326n6. See also journalistic and popular mapping; mapping sciences; maps; military mapping Cassini de Thury, César-François (Cassini III), 127–30, 128–29, 132, 138, 336n17, 336n20 Castells, Manuel, 312n21 Ceylon, 330n65, 330n73. See also Sri Lanka Chad, 330n65 Chayka (Чайка) radionavigation, 245, 249, 363n102 Chile, 42, 194, 318n27 China: Compass satellite system and, 256, 291–92; distribution of IMW mapmaking responsibility and, 39–40, 46, 319n41; International Ellipsoid and, 161;

no maps for unsurveyed terrain and, 60; place-names in the IMW and, 45; radionavigation stations in, 224, 243, 245, 291; Soviet grid system and, 198; UTM and, 196 Chirac, Jacques, 287 Chovitz, Bernard, 349n78, 351n99 Churchill, Winston, 218, 326n5 Civil Works Administration, 142 Clinton, Bill, 277, 284 Close, Charles, 33, 35, 37, 47, 54, 314n40, 318n25 Cold War, 4–5, 164, 165, 166, 196, 249, 277–78 Compass satellite system, 256, 291–92 computer technology: aircraft computers and, 346, 362n89, 363n91; computers integrated with maps and, 250–51; development of radionavigation and, 209; early lack of enthusiasm about, 249; integrated navigation and, 248; map-display systems and, 246, 247, 248; satellite navigation systems and, 261, 262–63; UNIVAC machines and, 192, 194; UTM and, 187, 191–92, 347n53 Consol radionavigation: integrated navigation and, 246; postwar expansion and use of, 239–40, 240, 243, 248, 361n81, 362–63n90; during World War II, 230. See also Sonne radionavigation Costa Rica, 349n76 Courtier, André, 132, 148, 337n25, 337n27, 340n65 Crawford, Osbert, 52, 54, 94–95 Crimea, 8, 311n17 Crone, Gerald, 111, 114, 115 Crosthwait, Herbert Leland, 318n25 Cuba, 112, 348–49n70 Cubic Corporation, 366n25 Cyprus, 96 Czechoslovakia, 344n14. See also Slovakia Daston, Lorraine, 30 datums (geodetic): consolidation of, 347n47, 349n80, 351n94; defined and named, 341n78, 341–42n81; first world datum and, 349n78; for North America, 342n85; for South America, 349n80; World Geodetic System (WGS) and, 371–72n74, 372n75. See also geodesy Davis, William Morris, 318n25 Decca Company (gramophones and records), 221 Decca Navigator Company: coastal marine and helicopter markets and, 242; global radionavigation system by, 244, 258; integrated navigation and map-display systems and, 239, 246, 248; offshore survey systems and, 232, 234

Index

379

Decca radionavigation: competition with American systems and, 241–42, 241, 242, 244; expansion after World War II and, 240–41, 240, 242; integrated navigation and, 246, 248; radiosurveying and, 231, 234, 266–67, 359n66; during World War II, 221, 224, 228, 229. See also Decca Navigator Company Dectra radionavigation, 246, 362n82 de Graaff Hunter, James, 156, 157, 342n90 deflection of the vertical, 341n78, 351–52n100 Delrac radionavigation, 362n82, 364–65n6 Denman, Roderick, 354n21 Denmark, 161, 189, 222, 242, 245, 291, 324n96, 344n15, 348n67 Derrida, Jacques, 13 Diego Garcia, 292 Dippy, Roger, 225 direction finding (D/F), 210, 215–16, 216, 218, 230, 241 Dobson, Jerome, 287–88 Dodge, Martin, 13 Domesday Book, 6, 310n13 Dominican Republic, 45, 285 Doppler navigation, 244–46 Doumergue, Gaston, 40 Draco radionavigation, 364–65n6 Eastern Air Lines, 248 Easton, Roger, 274–75, 369n42, 369nn44–45 Eckert, Max, 35, 37, 49, 59, 103 Edney, Matthew, 113–14, 334–35n134, 348n66 Eggert, Otto, 342n86, 342n90 Egypt: aeronautical maps and, 60; colonial surveys and, 172; color of deserts on the IMW and, 320n58; distribution of mapmaking responsibility and, 39, 88, 99–100; domestic grid for, 139; International Ellipsoid and, 161; radionavigation stations in, 245, 291; UTM and, 343n1 Einthoven, Willem, 38 electronic grids and coordinates: origins of, 15, 16, 208, 218, 224–25; politics of, 230, 298; rhetoric of, 205–6, 225, 228, 235, 250; territoriality and, 4, 14, 229, 235, 251; uses of, 228, 230, 231, 285 Electronic Position Indicator (EPI), 234, 359n66 Electronics Research Center, 267. See also NASA Elektra radionavigation, 356n34 ellipsoids (geodetic). See geodesy Emergency Relief Administration, 142 Engineering Society of Rio de Janeiro, 56–57, 57 Equatorial Guinea, 318n24

380

Index

Eritrea, 197 Estonia, 223, 291, 324n96, 344n14, 350n85 Ethiopia, 39, 197, 232, 234, 318n24. See also Abyssinia Europe: Aerosat project and, 368n37; area navigation systems for sea and air in, 209, 210, 215–17; in development of radionavigation, 209–10; distribution of IMW mapmaking responsibility for, 38; Eurocentric understanding of civilization and, 45–47, 55, 320–21n59; international geological map of, 321n69; pilot autonomy versus air traffic efficiency and, 354n21; radionavigation stations in, 222–24, 242–43, 245, 264, 291; recalculation of triangulation of, 155–58, 161–62, 185–89, 187, 191, 281, 347n53; UTM and, 181, 185–89, 190–91, 195–96, 346n43; UTM coordinate mismatch in, 348n67 European Coal and Steel Community, 69 European Union, 11, 29. See also Galileo satellite system Federal Aviation Administration, 267, 276–77, 283, 291, 366–67n26 Fels, John, 320n51 Fiji, 372n75 Finland, 242, 291, 324n96 Fischer, Irene, 193, 194–95, 194, 349n78 Fisher, Peter, 287–88 fishing industry, 9, 14, 235, 243, 248, 251, 285 Fortune magazine, Gallery 5, 70, 73, 74–75, 74–75, 77, 80, 88 Foucault, Michel, 13, 19, 374n101, 375n7 France: aeronautical maps and, Gallery 3, 51, 52; Argos project and, 367n27, 368n37; attempted coordinated cadastre in, 348n66; aviation reporting norms in, 354n19; carte de France and, 127–30, 128, 129, 340n67; carte de l’État major and, 125–27, 126, 127, 130, 340n67; cartographic one-upmanship and, 32; as center of air-age map, 327n16; colonial grid systems of, 196; direction-finding technology during World War I and, 215; early gridded maps and, 126, 336n13; firing framework for, 136; fragmentation of French cartography and, 149, 340n67; funding for railroads in, 352n7; ICAO and, 331n84; IMW and, 32–34, 39, 40, 60, 110, 318n27, 319n41, 319n46, 324n96; International Ellipsoid and, 151, 161; military mapping and, 130, 132, 134, 139, 318n23, 327–28n27, 337n23; negotiations with US over radionavigation and, 366n23; Paris meridian and, 317–18n22; postwar proliferation of navigation tech-

nologies and, 244; private companies’ radiosurveying and, 232, 234; radionavigation stations in, 222–23, 242–43, 245, 291; satellite systems and, 267, 367n27; South Pacific nuclear sites and, 366n21; surveying during World War I and, 138, 338n38; tax cadastre of, 148, 336n13; triangulation of, 153, 156; trig list for, 136; UTM and, 343n1; volume of maps printed during World War I and, 338n32; World War I command and, 337n28 French Institute at Pondicherry, India, 95 Galileo satellite system, 256, 287, 291–92, 374n98, 374n105 Galison, Peter, 30 Gannett, Henry, 33 Gauss, Carl Friedrich, 132, 337n26. See also grid systems; projections Gaussen, Henri, 95, 96 Gee radionavigation: coverage and stability of, 221, 222, 228–29, 229; as electronic grid, 205–8, 221, 228; invention and principles of, 218, 224, 225, 272; jamming and political neutrality of, 229–30; postwar status of, 230, 240, 360n75 Gee-H radiosurveying, 221 General Bathymetric Chart of the Oceans (GEBCO), 321n69, 323n77 General Electric Company, 267, 270 geodesy: as branch of mapping sciences, 16; British geodetic efforts in India and, 348n66; Clarke ellipsoid and, 340n67, 342n85; deflection of the vertical and, 341n78, 351–52n100; Delambre/Plessis ellipsoid and, 340n67; ellipsoid for North America and, 194, 194, 342n85; GPS and, 281, 314–15n41; gravity measurements and, 194–95, 349n79; grid systems and, 124; history of, 314n38; Hough Ellipsoid and, 193, 194–95, 349n78; international attempts to address questions of, 146–47; International Geodetic Association and, 314n33; major world datums and, 349n80; multilateral arrangements and, 315n42; Pierre Tardi’s international grid for Africa and, 160; practical and scientific phases of, 147, 150, 155, 156–58, 161; regional ellipsoids and, 195; satellites and earth’s gravity field and, 259; size and shape of the earth and, 151, 152, 153; South America and, 194–95; spheroid versus ellipsoid and, 341n74; triangulation and, 154, 193–94, 193; trilateration and, 232, 233–34, 235; UTM and, 166, 183–84, 183, 195; World Geodetic System and, 195, 232. See also datums (geodetic); International Ellip-

soid; International Union of Geodesy and Geophysics (IUGG); mapping sciences; triangulation geo-epistemology: authoritative representation and, 115–16; base maps and, 49; coordinate-based systems and, 228, 257; decline of universalism and, 67–69; demise of authoritative representation and, 66; geographic knowledge and, 2; GPS and, 2, 280, 285–86, 293; of grids, 123, 124, 138; human context of, 200; ideal of a general map and, 89; IMW and, 54; instrumental sensibility in, 67; mapmakers’ interests and, 320n51; of paper maps, 116; politics of, 20; radionavigation systems and, 206, 231; representation and, 13, 26, 41, 293, 295; territory and, 7, 208; tools of, 298; truth and, 293, 295; twentieth-century shift in, 14–15, 166. See also geographic knowledge Geographic Information Systems (GIS), 18, 315n42, 372n76 geographic knowledge: accuracy and, 2, 309n3; consummatio strategy and, 145; control over production of, 4; dispersal of via grid systems, 140; GPS and accessibility of, 297, 375n8; grids versus maps and, 297; IMW and, 63, 64; paper maps and artillery grids and, 134; prewar focus on territorial control and, 295; scale of questions about, 1–2; territorial states and, 297; timing of twentieth-century shift in, 15. See also geo-epistemology GEOREF system, 179, 346n35 Germany: beam systems and, 217–19, 219, 224, 230; Belgium-Germany border and, 350n89; British radionavigation stations in, 220, 222; captured map data from, 186, 188, 189; cartographic oneupmanship and, 32; Cassini coordinates and, 336n20; conscription of captured German geodesists and, 185, 186–87, 187, 198, 347n50; direction-finding technology during World War I and, 215; disregard of IMW grid and, 324n96; distribution of IMW mapmaking responsibility and, 57, 319n41; electronic coordinate systems during World War II and, 229–30; fragmented World War I command and, 337n28; grid systems and aiming of artillery and, Gallery 9, 130, 131, 132, 134; IMW as base map and, 110; increased wartime map coverage and, 80–81; informal cooperation for radionavigation and, 249; interwar military grid for, 139; IUGG and, 147, 340n63; jamming of electronic signals during World War II and, 230, 358n57; location

Index

381

Germany (continued) of Allied positions during World War I and, 138; military mapping and, 84; predicted fire technique and, 134, 337n29; radionavigation during World War II and, 217, 239–40; Radio Range and, 355n28; scale factor in grid systems of, 346n43; standardization in the IMW and, 33; survey discrepancies on western border of, 172; trig beacons and, 137; UTM and, 164, 177; volume of maps printed during World War I and, 338n32; World War II grids and, 170–71, 173, 344n14, 344n15; World War II supply lines and, 73. See also West Germany Getting, Ivan, 369n43 Geyer, Michael, 12 Gigas, Erwin, 186, 347n50 GIS. See Geographic Information Systems (GIS) globalization and globalism: as an American project, 5, 12, 115, 295–96; commercial promotion of navigation technologies and, 240–41; compared to other forms of universalism, 256, 257, 270; erosion of the territorial state and, 11–13; exclusively economic understanding of, 313n24; as geographic universalism, 257, 258, 271; GPS and single global coordinate system, 280–81; GPS as globally uniform mathematical system and, 290; GPS orbits and, 253, 254; history and geopolitics of the word global and, 14–15, 314n34; local intensification and, 280, 282; mapping sciences and, 295; national/international versus global/ regional space and, 15, 88; overexuberant globalization theory and, 312n21; regionalism and, 15, 27, 68, 71, 77–80, 85, 88–89, 93, 99, 110, 123, 165–66, 178–79, 181, 192–93, 195, 249, 251, 267, 269, 271, 275, 281, 290, 291; versus sovereignty, 13; telecommunications cables and, 12; versus territoriality, 298; trade-offs in, 267; United Nations emblem and, 77, 78; universal receivers versus universal transmitters and, 249–50; US command of the commons and, 282; US global consciousness and, 70–71; US military mapping and, 85, 232; UTM and, 165–66, 178, 184, 195; Wendell Willkie’s one world concept and, 71–72, 75, 75, 77, 80; World War II and, 69–70, 326nn5–6. See also internationalism; universalism Global Mapping initiative, 112 Global Navigation Satellite System, 292 Global Positioning System. See GPS globes, 77, 78, 80

382

Index

GLONASS (ГЛОНАСС) satellite system, 256, 291–92 Gloran radionavigation, 360n77 Goode, J. Paul, 326–27n13 Google Earth, 197, 351n92 GPS: accuracy of, 253, 278; aircraft landing systems and, 361n78; atomic clocks and, 272, 274, 368n39, 369n42; borderlessness and, 296; budget battles and, 275–76, 277, 370n50, 370n56; civilianization of, 256, 257, 276–77, 283, 290; commercial power and, 374n102; constellation of satellites for, 254, 273, 275, 276, 279; design and launch of, 271–78; as dual-use military-civilian system, 277–78, 370n58; early opposition to, 276–77, 368n39; as force multiplier, 297, 375n4; functional deployment of, 273, 273; geodesy and, 281, 314–15n41; geo-epistemology and, 280, 285–86, 293; goals of, 3, 5; go-for-broke approach to, 274–75, 277, 278, 370n52; ground installations supporting, 275, 291, 292; Gulf War debut of, 1, 254, 273, 273, 282, 372n79; historical questions about, 207–8; humanitarian relief and, 290; internationalization of, 290–92, 292, 295; intersection with other mapping initiatives and, 17; inventors of, 274–75, 369nn42–43; limitations on users’ perspective and, 248; local and regional augmentation systems for, 290, 291, 374n105; location-based services and, 253, 364n1; versus maps, 2–3, 166, 285–86, 293; military culture and, 374n101; military origins of, 253–56, 286–87, 288–90, 298; military uses of, 1, 275–76, 282–83, 287–90, 289, 372n77; modernization of satellites of, 290; multiple uses of, 283–84, 286–87, 288; multipurpose nature of, 256, 257, 278; neutrality of, 287–90; origins of, 273–74, 276, 289–90; outages and, 370n55; passivity of, 253, 254, 278, 292; patterns of everyday life and, 297–98; politics of, 286–93; predecessors to, 239, 256–58, 273–76, 369nn41–42, 369nn44–45, 370n48, 370n50; prehistory of, 231; proposed user fee for, 276, 277; as public utility or infrastructure, 278, 284; reasons for success of, 253, 257, 271–72, 274; relationship between user, landscape, and authority and, 2; RNAV navigation rules and, 363n97; satellite orbits of, 254, 369n44; Selective Availability policy and, 276, 277, 290, 291; signal structure for, 274, 369n43, 370n58; size, quality, and cost of receivers for, 278–79, 279,

283, 371n69; as solution to technological fragmentation, 272–73, 278, 369n40; specifications for civilian survey receivers for, 280, 371n71; surveillance and, 287–88, 373n96; surveying and, 280–81, 286; targeted marketing and, 374n101; technical literature on, 364n2; territoriality and, 282–83, 285–86; total-station surveying equipment and, 371n72; tracking and, 253, 255, 286, 288, 290, 373n96; transnationalization of space before, 210; transportation applications for, 283; ubiquity of, 278–79, 284, 286; universalism of, 208, 255–60, 278, 290–92; US technological globalism and, 253, 254; UTM and, 197, 206–7; WGS 84 datum and, 281, 371–72n74; as worldwide common grid, 205–6. See also electronic grids and coordinates Gramsci, Antonio, 321n65 graticule. See latitude and longitude Great Britain. See United Kingdom Greco-Turkish War, 318n23 Greece, 189, 223 Greenland, 65, 67, 224, 231, 243, 245, 331n77 Greenwich meridian, 29, 33, 170, 171, 317–18n22, 351–52n100 grid systems: abbreviation of coordinates and, 174, 179; ad hoc nature of wartime grids and, 167–70, 171; advantages of for surveyors, 144–45, 339n54; for Africa, 157–61, 159, 342nn93–94, 342–43n96; versus air-force and naval coordinate systems, 173–75, 174–77; carte de France and, 128–30, 129; circular grid zones and, 149–50, 149; conformal projections and, 133, 134; correction equations and, 348n63; curvature of the earth and, 119, 121; domestic politics of, 139–46; experience of the grid and, 134–35, 138–39; flat Euclidean plane and, 119, 127, 128–29, 199; Gauss Grid for Germany and, 170–71; Gauss-Krüger grid (проекция Гаусса- Крюгера) and, 344n14, 351n94; Gauss projection and, 132, 141, 159, 171, 337n28, 340nn67–68; geo-epistemology of, 123, 124, 138; global space and, 198–201; global versus national systems and, 190–91; graphic treatment of, 335n4; Greenwich meridian and, 170, 171; as hybrid between map and world, 123, 138–39; impact of on mapping, 121, 123; imperial versus metric units and, 335n5; International Ellipsoid and, 154, 160; international politics of, 146–48; junctions between neighboring grids and, Gallery 9, 130, 131, 132; Lambert projection and, 132, 133,

141, 141, 148, 168–69, 337nn27–28, 339n52, 340nn66–68, 341n71; versus latitude and longitude, 16–17, 119, 121, 128, 133, 199–200, 314n40, 350n87, 351n97, 351–52n100, 351n97; mapping of Cold War enemies’ spheres and, 165; mathematics in invention of, 125, 130, 132, 337n25; Military Grid Reference System and, 351n92; multiple uses of, 121, 139, 163, 338n43; Oblique Mercator projection and, 344n8, 351n95; origins of, 127–30, 336n18; paper maps and, 167, 173, 199, 200–201, 297; permanence of grid coordinates and, 145, 146; polyconic projection and, 170, 344n12; professional ambitions of creators of, 140, 144, 162; recalculation of the European Triangulation and, 158; Rectified Skew Orthomorphic projection and, 348n65; Roussilhe’s international grids and, 148–51, 149, 154; scale and, 123, 124, 130, 138; shooting across grid and survey system boundaries and, 171–72; size and shape of the earth and, 151, 153; Soviet grid systems and, 164–65, 165, 170, 184–85, 196, 198, 200, 344n14, 351n94; spatial logic of radionavigation and, 208, 250–51; state power and, 145–46, 148; stereographic projection and, 148–50, 149, 161, 340n66, 343n99, 344n8, 345–46n34, 351n95; système Lambert and, 337n27; Transverse Mercator projection and, 141, 141, 148, 151, 158, 159, 160, 164, 168–70, 198, 344n8, 350n84; triangulation and, 153, 155–58; United States National Grid and, 350n84; usable width of a grid and, 344n8; useful versus true coordinates and, 200, 201; UTM versus Soviet system and, 165, 198; World Polyconic Grid and, 170, 174, 177, 344n12, 346n39, 349n72; World War I and, 119, 124–27, 130–32, 133, 134, 335n1, 337n23; World War II and, Gallery 10, 166–76, 168, 174–77, 344n12, 344nn14–16, 345n24. See also electronic grids and coordinates; State Plane Coordinate System; système Lambert; Universal Transverse Mercator (UTM) grid Grotius, Hugo, 213, 215 Guadeloupe, 372n75 Guatemala, 232 Guier, William, 259 Gulf War, 1, 254, 273, 282, 309n2, 372n79 Guyana. See British Guiana Habenicht, Hermann, 317n19 Haiti, 287, 288 Harley, J. B., 113–14, 321n65, 334–35n134

Index

383

Harrison, Richard Edes: hemispheres and, 327n16; on impossibility of authoritative projection, 80; “One World, One War” map of, 77, 78; perspective views and, 75, 76, 77–78, 327n25; regionalism and, 88; seafloor mapping and, 235, 238 Harvey, David, 13 Havana Convention of 1928, 322n71 Hawaii, 170, 224, 243, 245, 264, 291–92 Hayford, John, 152, 154, 161, 341–42n81, 349n72 Hedin, Sven, 57 Heezen, Bruce, 235 Helsinki Accords (1975), 8 Henrikson, Alan, 70–71, 77 Hertz, Heinrich, 354n22 Hi-Fix radiosurveying, 234, 359n66 Hinks, Arthur, 26, 59, 61, 79, 151, 324n88 Hiran radiosurveying, 232 Hobsbawm, Eric, 326n6 Hough, Floyd: career of, 185–86, 347n49; first presentation of UTM and, 195; Hough Ellipsoid and, 193, 194–95, 349n78; installation of UTM and, 185–86, 188, 189–90, 192, 348n61; on layered territory and UTM, 196; national grid systems and, 190; postwar search for German map data and, 186, 187, 347n50; retirement of, 195; on unsuitable uses for UTM, 196 Hudson River valley, Gallery 1, 23, 24 Hughes Aircraft Company, 366n25 Hungary, 112, 139, 318n27, 320n55, 344nn14–15 Hurault, Louis, 92–93 Hyper-Fix radiosurveying, 234 IBM, 366n25 ICAN. See International Commission for Air Navigation (ICAN) ICAO. See International Civil Aviation Organization (ICAO) Iceland, 9, 224, 243, 245, 291, 343n1 Imhof, Eduard, 319n38 imperialism and colonization: AGS and, 58; American empire and, 12–13; dissolution of empires and, 11; IMW and, 38–39, 39, 41–42, 57–59, 62–64; poorly surveyed terrain and, 60–61; provisional mapping and, 60, 61 IMW. See International Map of the World India: British geodetic efforts in, 348n66; Carte Internationale du Tapis Végétal and, 330n65, 330n73; IMW dues and, 89; problems in geodesy and, 341n78; radionavigation stations in, 224, 242, 245, 291; triangulation in Southeast Asia and, 192. See also Bengal; Kashmir; Punjab Indonesia, 350n82. See also Sumatra

384

Index

inertial navigation, 244–46 Inter-American Geodetic Survey, 191, 348–49n70 International Cartographic Association, 317n21 International Civil Aviation Organization (ICAO): aeronautical map standards of, 68; aircraft landing systems and, 361n78; arrangement of chart specifications and, 108; civilianization of GPS and, 277; clashes over postwar radionavigation technologies and, 244, 246; computers discussed at meetings of, 363n91; distribution of mapmaking responsibility and, 99–100, 101, 331n84; Future Air Navigation Systems conferences and, 372–73n85; GPS and, 283, 291–92; ICAN and, 90; IMW and, 109; international cooperation for radionavigation and, 258; level of standardization and, 99, 102, 111, 331n77; listing of long-distance aids by, 248; mapping sciences and, 17; member obligations and, 90, 329n56; political universalism and, 271; popularity of ICAO maps and, 105; readers of maps and, 94, 102, 107; satellite navigation systems and, 267, 271; standardization of radionavigation and, 239, 241–42, 360n73; US low-altitude airway network and, 241; WAC and, 66–67, 90, 91, 101, 107, 108; wartime versus postwar mapmakers and, 99, 331n76 International Commission for Air Navigation (ICAN): aeronautical maps and, Gallery 3–4, 49–50, 51, 52, 53, 60, 322n70; color to represent elevation and, Gallery 4, 52, 53; D/F navigation and, 216; establishment and membership of, 49–50, 322n70, 322n71; Havana Convention of 1928 and, 322n71; ICAO and, 90; IMW and, Gallery 4, 52, 53, 60, 323n78; maps versus charts and, 52; prohibitions on women aviators and, 354n18; regulations of, 354–55n25; scale and, 50, 52, 322n72 International Criminal Court, 11 Internationale Erdmessung, 147 International Ellipsoid: Africa and, 194, 194; for domestic purposes in Europe, 348n67; IUGG and, 151, 153–55, 160, 161–62, 343n100; USSR and, 188, 347n56; UTM and, 183, 184, 186, 188 International Geodetic Association, 314n33 International Geographical Congress: IMW and, 23, 30, 32–34, 59–60, 317–18n22; International Geographical Union and, 89; International Map of the Roman Empire and, 52 International Geographical Union: base-map ideal and, 110–11; graphics approved by,

96; IMW versus WAC and, 89, 92–93; mapping sciences and, 17; 1962 IMW conference and, 110–13; proposals for civilian mapping projects and, 90, 94–95, 329n57; statistical and thematic mapping and, 90 International Geological Congress, 49, 321n68 International Hydrographic Bureau, 52, 323n77 internationalism: American pragmatism and, 264, 270–71; American upstaging of, 66, 81, 188, 295–96; failures of, 27–28, 65–67, 94, 96, 146–47, 210, 239; versus globalism, 14–15, 27, 69; GPS and, 290–92; grids and, 124, 146–48, 161; of IMW, 25, 27, 29, 41–42, 44, 45–46, 47, 55, 59, 295, 329n53; of mapping projects, 112, 321n69; radionavigation and, 210, 213, 215–16, 239–42, 246; UTM and, 189–91; WAC and, 329n55. See also globalization and globalism; universalism International Map of the Roman Empire, 54, 94–96, 323n81, 323n84, 330n63, 330–31n74 International Map of the World: aeronautical maps and, 49–50, 52; from age of exploration to age of scientific synthesis and, 25; AGS Map of Hispanic America and, 57–64, 57, 58, 60, 62, 63, 324n96, 325n98; annual reports for, 46, 47, 93, 334n124; assembly of multiple sheets of, 62–63, 62, 63; authority of, 28–29, 34–35, 38, 40–41, 48, 54–55, 113; as base map, Gallery 4, 26–28, 48–52, 53, 54, 66, 68, 89–90, 92–94, 110–11, 112–13; cartographic one-upmanship and, 32; Central Bureau of, 35, 52, 54, 59–60, 89, 323n78, 323n81, 325n100, 329n53; changing cartographic norms and, 65–67; color to represent elevation and, Gallery 2, Gallery 4, 35–38, 45, 52, 53, 55–56, 56, 319n37, 320n56, 354–55n25; as continuation of past ambitions in cartography, 316–17n13; continued IMW-compliant mapmaking and, 325n2; countries adhering to, 25, 315n2; coverage of, 67; decline of universalism and, 69, 113, 115; declining interest in, 27, 65–67, 68, 70, 88–89; derivative maps and, 48, 52, 54; D/F navigation and, 216, 354–55n25; disagreement about elements of, 30, 32, 317n19; distribution of mapmaking responsibility and, 38–41, 42, 46–47, 56–58, 56, 64, 88, 319n41, 319n44, 319n46; dual visual and political representation and, 27, 28–29; economic development aid and, 67, 112; end of,

68–69, 113; Eurocentric understanding of civilization and, 45–47, 55; experts involved in, 314n40; features shown and symbols chosen for, 42, 43–44, 320n53, 320n58; French official title of, 315n1; frontispiece to annual reports of, 46, 47; general versus specialist readers and, 102, 109; goals of, 29, 30, 32, 41–42, 46, 64, 111–12, 114; grid for sheet divisions of, Gallery 4, 30, 33, 53, 91, 317n15, 324n96; grid systems and, 150, 158, 159; as guide to changes in cartography, 28; incompatibilities between sheets of, 32–33, 38; influence of in cartography, 27; International Commission for Air Navigation (ICAN) and, Gallery 4, 53, 323n78; internationalism and, 25, 27, 29, 41–42, 44, 45–46, 47, 55, 59, 295, 329n53; International Map of the Roman Empire and, 54, 323n84; intersection with other mapping initiatives and, 17, 315n42; League of Nations and, 92; lessons of, 112, 113; limited use of, 27, 316n8; logic of representation and, 26; mapping projects challenging the supremacy of, 66–67; military mapping and, 81, 109; neutrality and, 41–42, 47, 48, 54–55, 114; new era in cartography and, 26; 1909 London conference on, 33–34, 36–38, 37, 40–44, 48; 1913 Paris conference on, 34–35, 36–39, 37, 40–44, 46, 318n27; 1919 Paris Peace Conference and, 27, 28, 58, 61; 1928 conference on, 320n58; 1962 UN conference to revise specifications of, 68, 110–13; number of sheets in, 33, 328n31; origins of, 16, 23, 25, 28, 314n33; pace of production and, 66; patronage for, 32, 48, 317–18n22; place-names in, 44–45, 320n55; plan for Berlin conference on, 319n44; politics in production and use of maps and, 57–59, 62–64; printing of, 316n12, 317–18n22; projections for, 32, 33, 317n14, 318n23; provisional mapping and, 55–61, 325n100; relative reliability diagrams for, 60, 60; scale of, 25, 28–30, 32, 40–41, 42, 49–50, 112, 317n19, 320n53; Series 1301 and, 109, 112; slow start for, 30, 32; Soviet Union’s relationship to, 316n7, 324n87; standardization and, 23, 25, 29–39, 36–37, 42, 44, 84, 111–13, 318n25, 318n29; success versus failure of, 27–28, 65–66; timing of publication and, 56–57, 56; trade-off between the meter and Greenwich and, 33, 317–18n22; trustworthiness of, 26; United Nations and, 65, 92, 93; universalism and, 255, 271; unsurveyed areas in, 328n32; US

Index

385

International Map of the World (continued) upstaging of, 81, 188–89; UTM and, 163, 177, 178, 179, 181; “view from nowhere” and, 41–42, 47, 64, 320n52; WAC versus, Gallery 6, 68, 81, 82, 83, 89–90, 91, 92–93, 110–12, 328n32; widespread support for, 30, 32; World War I and, 28, 34; World War II and, 19–20, 27, 64, 65, 89 International Meeting on Radio Aids to Marine Navigation, 360n77 International Radio Conference (1947), 360n76 International Telecommunications Union (ITU), 239 International Union of Geodesy and Geophysics (IUGG): cadastral coordinate systems and, 343n98; countries excluded from, 147, 340n63; debates over grids and, 124; Henri Roussilhe’s international grids and, 148–51, 149, 161; International Ellipsoid and, 151, 153–55, 160, 161–62, 343n100; international standardization of coordinate systems and, 147; national and international projects and, 147–48, 161–62; Pierre Tardi’s international grid for Africa and, 158–61, 177; presenters at meetings of, 342–43n96; structure of, 147; tables for Transverse Mercator projection and, 151; UTM and, 186, 189; William Bowie’s recalculation of the European triangulation and, 155–58, 161–62 Internet, 197, 284 Iran, 188, 232, 344n14. See also Persia Iraq, 96, 282–83, 344n14, 347n57 Ireland, 89, 291, 324n96 Israel, 96, 197, 330–31n74 Italy: aeronautical maps and, 52; IMW and, 39, 56, 60, 110, 318n27, 319n44, 325n100; International Ellipsoid and, 348n67; military mapping and, 134, 327–28n27; radionavigation stations in, 222, 245; WAC map of, 106; World Land Use Survey and, Gallery 7, 98, 99 ITT Corporation, 217, 258, 361n79 IUGG. See International Union of Geodesy and Geophysics (IUGG) Japan: Global Mapping initiative and, 112; IMW and, 39, 46, 56, 112, 318n27; increased wartime map coverage and, 80–81; lack of US aeronautical maps for, 81; military mapping and, 84; negotiations with US over radionavigation and, 366n23; Pacific ambitions of, 79; Pearl Harbor and, 81; radionavigation during World War II and, 217; radionavigation stations in, 242–43, 245, 264, 291; trans-

386

Index

literation for maps and, 320n58; World Land Use Survey and, Gallery 7, 98, 99 Jet Navigation Chart, 105 John Deere, 291, 374n105 Johns Hopkins Applied Physics Laboratory, 259 Johns Hopkins University, 267 Joint Navigation Satellite Committee, 366–67n26 Jones, Reginald, 230, 357n45 Jordan, 197. See also Trans-Jordan journalistic and popular mapping: air-age maps and, 73, 74, 77, 327n16; aviation and, 78; directional captions and, 76, 78; general versus specialist readers and, 102; geopolitics of Halford Mackinder and, 326n9; global consciousness and, 70–71; globalization and regionalism and, 80; ICAO and, 105; making oceans seem less vast and, 75, 77, 327n15; maps centered on cities and, 77, 327n16; military mapping and, 70, 88; national/ international versus global/regional space and, 88; North Pole–centered maps and, Gallery 5, 72, 73–75, 74, 77; oil and supply routes during World War II and, 73; reader in subject position and, 78; readership of maps and, 94; Richard Edes Harrison’s perspective views and, 75, 76; US connections to major war zones and, 75; world atlases and, 77–78, 79, 327n21, 327n25 jurisdiction: cartography and, 3–4, 68; geographic and nongeographic bases of, 6; grids and, 123, 146, 162; oil and gas and, 14, 312n20; radiosurveying and, 235; versus territory, 5–6, 7, 15–16, 295; Treaty of Westphalia and, 6; universal jurisdiction and, 6, 11. See also sovereignty; territory and the territorial state Karussell radionavigation, 356n34 Kashmir, 8 Keltie, John Scott, 318n25 Kenya, 287, 342n93 Khristov, Vladimir K., 342–43n96 Kiebitz, Franz, 353n12 Kissam, Philip, 143–46, 314n40, 340n59 Kitchin, Rob, 13 Knickebein radionavigation, 218, 219, 356n32, 356n34 Korea. See South Korea Korean Airlines, Soviet attack on flight of, 277 Korean War, 85, 176, 231, 347n59 Korzybski, Alfred, 113 Kosovo, 11 Kurgan, Laura, 284

Kuwait, 282, 291, 311n17 Kwajalein, 292 Ladd, John, 188 La Guardia, Fiorello, 74–75 Lallemand, Charles, 322n70, 340n67 Lambert, Johann Heinrich, 132. See also grid systems; projections; système Lambert Laos, 192 Latin America, 191–92, 348n64, 348–49n70. See also Map of Hispanic America latitude and longitude: aviation and, 173–74, 174, 354n19; as badly designed grid, 199, 351n98; carte de l’État major and, 126, 127; Cassini’s carte de France and, 128–29; deficiencies of, 199; electronic grids as replacement for, 228; GEOREF and, 346n35; GPS and, 206; versus grid systems, 119, 121, 123, 126, 199–200, 350n87, 351n97, 351–52n100; IMW and, 23, 25, 30, 91, 106; international boundaries and, 145; marine navigation and, 121, 173–74; as natural versus artificial system, 199–200; origins of, 17, 199; radiosurveying and, 250; Roussilhe’s scheme and, 150; survey monuments and, 154, 158; système Lambert and, 133; triangulation and, 153; UTM and, 16, 160, 165–66, 178–79, 179, 181, 196–97, 200; WAC and, 81, 86, 91; WGS 84 datum and, 371–72n74 Latour, Bruno, 313n31 Latvia, 344n14, 350n85 League of Nations, 7, 61, 69, 72, 92 Lebanon, 161, 172, 343n99 Lee, Laurence Patrick, 342–43n96 Lefebvre, Henri, 13 Liberia, 264, 266, 366nn22–23 Libya, 60, 222, 224 Life magazine, 62, 62, 63, 70, 72, 73 Lippmann, Walter, 327n15 Lithuania, 291, 344nn14–15, 350n85 longitude and latitude. See latitude and longitude Loomis, Alfred, 357n46 Lorac radiosurveying, 359n66 Loran radionavigation: compared to GPS, 272; cooperation between US and other countries and, 245, 249, 363n102; coverage of, 224, 231, 243, 245, 360n74; Electronic Position Indicator (EPI) and, 359n66; expansion after World War II and, 231, 239–41, 243–44, 243, 245, 358n59; frequency allocations and, 360n76; Gloran system and, 360n77; grid metaphor and, 225–28, 352n3; integrated navigation and, 246, 362n89, 362–63n90; Loran-C compared to

Loran-A and, 244, 245, 362n82, 362n84; Low-Frequency (LF) Loran and later systems, 364–65n6; origins of, 357n46; during World War II, 221, 224–25, 224, 226–27, 228–29, 229, 258 Lüddecke, Richard, 317n19 Luxembourg, 343n1 Macau, 372n75 Mackinder, Halford, 326n9, 327n15 MacLeod, Malcolm, 158, 160 Madagascar, 132, 148, 160, 330n65 Magnavox, 261, 263, 266 Maier, Charles, 11–12 Malawi, 112. See also Nyasaland Malaya, 190, 348n65, 351n95 Malaysia, 291. See also Malaya map firing, 125–26, 130, 132, 134, 138, 167, 335n11 Map of Hispanic America, 57–64, 57, 58, 62, 63, 324n96, 325n98 mapping sciences: agency versus intentionality of mapmakers and, 19; civilianization of military developments in, 20; globalism and regionalism and, 295; grids versus graticule of latitude and longitude and, 16, 17, 314n40; practical versus scientific projects and, 295; project versus tools in historical analysis and, 18–19; representational maps versus pointillist approach to space and, 209; separation of the territorial and nonterritorial and, 161; three branches of, 16, 17–18, 314n37; undersung mapmakers and engineers and, 18–19; unexpected uses of tools and, 19 map projections. See projections maps: aeronautical maps and, 49–50; agricultural management and, 7; carte de l’État major and, 125–27, 126, 127, 130, 335–36n12; classroom wall maps and, 105; commonly used symbols on, 43–44; computers integrated with, 250–51; deformation of paper and, 135; diminishing importance of paper and, 134–35, 138; first appearance of national borders on, 7, 310–11n14; geo-epistemology of, 116; geographic history of the state and, 6–7; geographic truth versus functional tasks and, 27, 28; goal of, 3, 5; versus GPS, 3, 5, 166, 285–86, 293; versus grid systems, 123, 167, 199, 200–201, 297; malleability of, 15; of the ocean floor, 235, 238; orientation of north on, 70, 76, 80; versus pointillist approach to space, 209; printing of grids on, 163, 173; radionavigation and, 217, 355n27; roller maps and, 247, 363n94; taxation and, 7;

Index

387

maps (continued) territoriality and, 3–4, 113–14, 115; time required to produce maps and, 375n5; universal scientific truth and, 14; use of color on, 29, 37; Western Hemisphere and, 77. See also cartography; journalistic and popular mapping; military mapping Marconi, Guglielmo, 354n22 Martin, Ellis, 47 Marx, Karl, 297 Massachusetts Institute of Technology, 225, 267 mathematics: Bowie method and, 156, 157, 342n89; calculational efficiency of grids and, 171, 345n24; carte de France and, 128; circular grid zones and, 149–50; conformal grids and, 134; conformal projections and, 148; correction equations for grid systems and, 348n63; GPS as globally uniform mathematical system and, 290; grid junctions and, 172; grids’ geographic limits and, 167, 168; grid systems versus latitude and longitude and, 199; in invention of grid systems, 125, 130, 132, 337n25; land geography versus air and sea geography and, 175; mass calculation before computers and, 191–92; in military mapping, 135, 137, 138; Pierre Tardi’s international grid for Africa and, 160–61, 342–43n96; survey junctions and, 172; Transverse Mercator projection and, 160; triangulation and, 132, 156; UTM and, 163–64, 177, 178, 181, 184, 189–90, 346n44; WGS 72 datum and, 372n75; WGS 84 datum and, 281, 371–72n74, 372n75; Works Progress Administration projects and, 349n72 Mauritius, 372n75 Maury, Jean, 342n86 McKim, Mead & White, 83 Mediterranean, 235 Mexico: AGS map of, Gallery 6, 82; Carte Internationale du Tapis Végétal and, 330n65; Inter-American Geodetic Survey and, 348–49n70; North American triangulation systems and, 154; relative reliability in map of, 60; UTM and, 196 Micro-H radiosurveying, 221 Middle East, 73, 76, 188, 196, 234 military mapping: administrative regionalism and, 85, 88; aerial photography and, 83, 336n13; aiming of artillery and, 125–26, 130, 132, 133, 134–35, 335n11, 337n29; assembly-line techniques and, 83–84, 107; functionally specific maps and, 107, 109; general versus specialist readers and, 102, 103–5, 107, 109; IMW

388

Index

and, 32, 34, 66, 67, 81, 93, 318n23; increased map coverage, 80–81; indicators of elevation and, 332–33n106; information overload and, 103, 107; journalistic and popular mapping and, 70; mapmaking production speed and, 83–85; as a postwar project of state power, 296, 375n1; print quantities and, 1, 309n2, 338n32; regional division of responsibility and, 85–86, 86–87; scale and, 81, 85, 109, 126, 327–28n27; statistical and thematic mapping and, 90; surveying and, 135, 137, 138; thrust-line method and, 345n25; training in map reading and, 333n107, 338n32; US Joint Army-Navy reference system and, 176–77; US role in UTM map production and, 188–89; World War II as geographically transformative and, 88–89. See also cartography; maps; military targeting military targeting: ad hoc nature of wartime grids and, 167–70, 171; aiming of artillery and, 125–26, 130, 132, 133, 134–35, 335n11, 337n29; blind-bombing systems and, 217, 221, 231, 233; data capture and grid confusion and, 172–73; German intersecting-beam systems and, 218–19, 219; GPS and, 253, 254–56, 275–76, 285, 287, 289, 290; gravity measurements and, 194; gridded battlefield and, 228, 250; Identification-Friend-or-Foe equipment and, 356n35, 356–57n39; nonEuclidean and latitude-and-longitude reference systems for air and naval forces and, 173–75, 174–77, 178–79; Norden bombsight and, 220; northings and eastings and, 345–46n34, 351n93; nuclear and intercontinental ballistic missiles and, 164, 190, 259, 264, 275; Oboe system and, 219–20, 220; plotting boards and, 134, 135, 138, 338n34; as a postwar project of state power, 296, 375n1; power of grid systems and, 198–99; precision distance measurement and, 218–20, 220; precision revolution and, 282; radionavigation and, 209; radio trilateration and, 234; shooting across grid and survey system boundaries and, 171–72; Shoran system and, 219–20, 231; speed of calculations and, 181, 346n40; surveying and, 135, 137, 138; trig lists and, 135, 136, 137, 138, 188; UTM and, 181, 185, 190, 192; Wotan II system and, 356n36; Y-Gerät system and, 219, 219, 356n36. See also military mapping Moldova, 350n85 Mongolia, 198, 350n85 Montevideo Convention (1933), 8

morality of technological change, 296 Morocco, 318n24 Namibia, 255 Napoleonic Wars, 326n5 NASA, 257–58, 267–71, 276, 366–67nn25–26, 368n33 National Academy of Sciences, 267 National Research Council, 267, 277 nation-states: as constructed collective identity, 311n15; ideal of, 7, 8, 11; maps as anchors of nationalism and, 7. See also territory and the territorial state NATO: expansion of Loran systems and, 243, 243; radionavigation and, 239, 244; regionalism and, 69; UTM and, 164, 188, 191, 343n1 Navaglobe radionavigation, 362n82 Navarho radionavigation, 362n82 navigation: as branch of mapping sciences, 16; car navigation and, 283, 372n84; GPS transportation applications and, 283, 372n84; integrated navigation and, 239, 246, 248, 250, 271; merger of mapmaking with, 257; multilateral arrangements and, 315n42; possible use of satellites for, 259; sun and stars and, 174, 200; territorial state versus permeable boundaries and, 208; total navigational capability and, 248; useful versus true coordinates and, 200. See also electronic grids and coordinates; GPS; mapping sciences; military targeting; radionavigation (nonsatellite); satellites and satellite navigation systems neoliberalism, 11, 13 Nepal, 330n73 Netherlands, the, 139, 161, 223, 242, 291 Newton, Isaac, 151, 341n74 New Zealand, 86, 100, 112, 261, 366n20 Niger, 330n65 Nigeria, 234, 242, 318n24 Nixon, Richard, 349n76 Nørlund, Niels, 156, 157, 342n90 North America: airway schemes and, 211; consolidation of triangulation in, 154, 155; Pan-American Railway and, 210–11; UTM and, 183–84, 183, 193–95, 194 North Atlantic Treaty Organization. See NATO North Rhodesia, 342n93 North Sea, 235, 267 Norway: antinuclear activism and, 261; domestic grid for, 139; IMW as base map and, 110; negotiations with US over radionavigation and, 366n23; radionavigation stations in, 223, 242–43, 245, 264, 291; territorial claims to the

sea and, 235; UTM and, 181, 346n41, 351n93 nuclear weapons, 259–61, 262, 264, 267, 275, 366n21 Nyasaland, 342n93. See also Malawi Oboe navigation and radiosurveying, 219–20, 220, 221, 231, 272 oceans: as barriers or connectors, 71, 75, 77; as metaphor, 77, 210, 213, 215; national claims to, 9, 9–10, 14, 235, 251; radionavigation coverage in, 224, 242–43, 245; radiosurveying and, 232, 234–35; satellite coverage over, 268, 269; visualized as land, 235, 236–38. See also UN Convention on the Law of the Sea (UNCLOS) Office of Naval Research, 102, 104 oil and gas industry, 235, 250, 266–67, 312n20 O’Keefe, John: career of, 314n40; on grid systems, 191, 199–200, 339n54; satellites and, 259; UTM and, 17, 176–77, 181, 185, 314n40, 347n47 Omega Platform Location Equipment, 269, 368n32 Omega radionavigation: compared with other global systems, 257–58, 267; Decca’s Delrac and, 365n7; decommissioning of, 267; differential GPS and, 290; integrated navigation and, 267, 271; lattice charts for, 260, 260; military and civilian uses of, 260–61; origins and design of, 258, 272, 364–65n6, 365n13; politics of ground installations for, 261, 264, 264–65, 266, 366nn22–23; satellites and, 269; Soviet counterpart to, 364n5 Operational Navigation Chart, 105, 332n100 Oreskes, Naomi, 341–42n81 Outer Space Treaty (1967), 9 Page Communications Engineering, 366n25 Pakistan, 242, 330n73 Palestine, 172 Panama, 191, 264, 291, 343n1, 372n77 Pan American Institute of Geography and History, 195 Pan American system, 69 Paraguay, 61, 348–49n70 Paris meridian, 32, 317–18n22 Paris Peace Conference of 1919, 27, 28, 58, 61 Parkinson, Bradford, 274–76, 282, 284, 290, 369nn43–45, 370n56 Parus (Парус) satellite system, 364n5 Penck, Albrecht: authority of the IMW and, 34–35, 38; goals of the IMW and, 29, 32, 64; incompatibilities between sheets of the IMW and, 38; origins of the IMW

Index

389

Penck, Albrecht (continued) and, 23, 25, 26, 28; printing of the IMW and, 316n12, 317–18n22; on provisional and definitive representations, 55, 323–24n86; significance of IMW for, 25; slow start for the IMW and, 30, 32; standardization in the IMW and, 36, 318n25 Percy, Charles, 277 Perrier, Georges, 153, 154, 155, 158, 161 Persia, 318n23. See also Iran Persian Gulf, 242, 291, 372n77 Peucker, Karl, 37–38, 45, 319n38 Philco, 366n25 Philippines, 170, 224, 243, 343n1 Pierce, John, 258, 260, 364–65n6, 365n13 plate tectonics, 235 Poland, 161, 223, 230, 291, 344n14–15 popular mapping. See journalistic and popular mapping Portugal, 100, 155, 161, 291, 318n27, 343n100 Posen, Barry, 282 Prince Albert (Monaco), 52 projections: air-age maps and, 73–74; authoritative, 79–80; Bonne projection and, 126, 127, 130, 133, 340n67; carte de l’État major and, 127; Cassini projection and, 128–29, 130, 170; conformal, 132, 133, 148, 337n24; distortions created by, 127, 130, 133, 337n25; experimental world maps and, 79; in French cartography, 149, 340n67; Gauss projection and, 132, 141, 159, 171, 337n28, 340nn66–68; grid systems and, 124, 128–29; for the IMW, 32, 33, 317n14, 318n23; IMW versus WAC and, 90; Lambert projection and, 132, 133, 141, 141, 148, 168–69, 337nn27–28, 339n52, 340nn66–68, 344n8; Mercator projection and, 70, 71, 74, 79–80, 141, 164, 322n72, 326–27n13; national versus international systems of, 160; North Pole–centered, Gallery 5, 72, 73–75, 74, 75, 77, 78, 79–80, 88; Oblique Mercator, 79, 344n8; polyconic, 170, 344n12; polyhedric, 317n14; Rectified Skew Orthomorphic, 348n65; stereographic, 148–50, 149, 340nn65–66, 343n99, 344n8, 345–46n34; Tissot’s projections and, 337n25, 340n65; Transverse Mercator, 141, 141, 148, 151, 158, 159, 160, 164, 168–70, 342–43n96, 344nn8–9, 350n84; tronconique, 317n14. See also grid systems Project 621B, 273–76, 369nn41–42, 369nn44–45, 370n48, 370n50 Prony, Gaspard Riche de, 348n66 Puissant, Louis, 129, 336n18 Punjab, 311n17

390

Index

Radio Mailles radionavigation, 362n82 radionavigation (nonsatellite): air traffic control and, 215, 249, 251, 354n21; Alpha (Альфа) system and, 364n5; black boxes for, Gallery 11, 205, 206, 207; Chayka (Чайка) system and, 245, 249, 363n102; civilianization of, 244, 251; Consol system and, 230, 239–40, 240, 243, 246, 248, 361n81, 362–63n90; coverage of systems and, 222–24, 242–43, 245; Decca systems and, 221, 224, 228, 229, 231–32, 234, 240–41, 240, 242, 244, 246, 248, 266–67, 359n66, 362n82; Dectra system and, 246, 362n82; Delrac system and, 362n82, 364–65n6; DIAN system and, 246; direction-finding (D/F) technology and, 210, 215–16, 216, 218, 230, 241; distance measurement and, 221, 231, 272, 359n66; distinction between shortand long-range navigation and, 361n78; DME system and, 361n79; Doppler navigation and, 244–46; Draco system and, 364–65n6; early inventions in, 353n12, 354n22; Elektra system and, 356n34; epistemic concerns of, 206, 231; Gee system and, 206–8, 218, 221, 222, 224, 225, 228–29, 229, 230, 240, 272, 360n75; German intersecting-beam systems and, 217–18; Gloran system and, 360n77; goals of navigation and, 248–49; government supply of commercial equipment for, 240–41, 360nn75–76; GPS and, 256–57, 278; grids and, 205–9, 218, 224–25, 228–31; ground installations supporting, 210, 211–12, 213, 214, 215–16, 216, 218–19, 226, 228–30, 229, 261, 264, 264, 266; historical transition in, 20; hyperbolic, 224–25, 225–27, 258, 357n44, 359n66; Identification-Friend-or-Foe equipment and, 356n35, 356–57n39; IMW and, 216, 354–55n25; inertial navigation and, 244–46; integrated navigation and, 239, 246, 248–50, 258, 267; international standardization and, 239, 241–42, 244, 246; jamming of signals and, 230, 358n57; Karussell system and, 356n34; Knickebein system and, 218, 219, 356n32, 356n34; lattice charts and, Gallery 12, 205, 207, 224–25, 227, 228–30, 250–51, 260, 260; Loran systems and, 221, 224–25, 224, 226–27, 228–29, 229, 231, 239–41, 243–44, 243, 245, 246, 249, 258, 272, 357n46, 358n59, 359n66, 360n74, 360n76, 362n82, 362n84, 362–63n90, 363n102, 364–65n6; mapdisplay systems and, 246, 247, 248; military targeting and, 209, 217–21, 224, 225; Morse code and, 211–12; multiple

uses of, 209, 234–35, 238; Navaglobe system and, 362n82; Navarho system and, 362n82; navigational problems with, 354n20; Oboe system and, 219–20, 220, 272; Omega system and, 257–58, 260–61, 260, 264, 264–65, 266, 267, 269, 271–72, 290, 364–65nn5–7, 365n13, 366nn22–23; point-to-point versus area systems and, 208–13, 215–18, 239, 241, 248, 251, 352n4, 355n28; postwar proliferation of technologies and, 238–39, 243–44; postwar US-UK clashes over, 239–44, 242, 246; precision distance measurement and, 218–19, 221, 357n40; Radio Mailles system and, 362n82; Radio Range system and, 210–13, 212, 214, 217–18, 219, 230, 239, 241, 353n12, 355nn27–29, 360n76; radiosurveying and, 218, 221, 231–32; Radux system and, 362n82, 364–65n6; RDF and, 352n5; resistance to, 357n45; RNAV navigation rules and, 248, 363n97; RSDN-20 (РСДН- 20) system and, 364n5; self-contained, 245–46, 362n87; Shoran system and, 221, 231–32, 272; Sonne system and, 221, 223, 224–25, 228, 229, 230, 357n44; TACAN system and, 361n79; talking down pilots and, 354n19; track-guidance versus area systems and, 208–9, 210–13, 215–18, 352n4; U-boats and, 228, 230; in US versus Europe, 210, 215–17, 354n21; VOR (VHF Omni-Range) system and, 213, 239, 241, 241–42, 248, 251, 357n44, 361n79, 364n106; VORTAC system and, 241, 361n79; during World War II, Gallery 11, 205, 206–7, 217, 222–24; Wotan I system and, 356n32; Wotan II system and, 356n36; X-Gerät system and, 218, 219, 356n32; Y-Gerät system and, 219, 219, 356n36; Zyklop system and, 356n34. See also GPS; satellites and satellite navigation systems Radio Range radionavigation: compared with direction-finding technology, 210, 215, 217, 230, 241; design and principles of, 211–13, 212, 217, 353n12, 355n27; German intersecting-beam systems and, 218, 219; in Germany and Austria, 355n28; postwar US offers of equipment for, 360n76; as predecessor to VOR, 217, 239; railroad metaphor and, 210–13, 214; US airway network and, 213, 214; in World War II, 217 radiosurveying: aerial photography and, 221, 231; civilianization of, 251; Electronic Position Indicator (EPI) and, 234, 359n66; Gee system and, 356–57n39; Gee-H system and, 221; GPS and, 280–81,

371–72n74; high-accuracy geodetic measurement and, 231–32, 233; Hiran system and, 232; Hyper-Fix system and, 234; Lorac system and, 359n66; Micro-H system and, 221, 231; Oboe system and, 221, 231; offshore surveying and, 232, 234–35, 236–37, 250; radionavigation and, 218, 221, 231–32; Rana system and, 359n66; Raydist system and, 234, 237, 359n66; Rebecca-H system and, 221, 231, 356–57n39; Sea-Fix system and, 234, 359n66; seafloor mapping and, 235, 238; Shiran system and, 232; Shoran system and 221, 231–32, 233, 236, 359n66; Toran system and, 359n66; Transit system and, 266–67, 281, 372n75; trilateration and, 232, 233–34, 235 Radux radionavigation, 362n82, 364–65n6 railroads: GPS and, 283; as metaphor in aviation, 210–13, 211, 214, 352n7; shown on maps, Gallery 3–4, Gallery 7, 36, 42, 50, 51, 53, 83, 104; Trans-Siberian, 186 Rana radiosurveying, 359n66 RAND Corporation, 277 Rand McNally, 71, 74, 77, 79, 327n25 Raydist radiosurveying, 234, 237, 359n66 RCA (Radio Corporation of America), 217, 221, 267, 368n36 Reagan, Ronald, 277 Rebecca-H radiosurveying, 221, 356–57n39 regions and regionalism: administrative regionalism and, 85, 88; air-age maps and, 77–80; versus base-map ideal, 100, 110; division of Allied mapping responsibility and, 85–86, 86–87; GPS augmentation systems and, 290, 291; grid coordinates and, 123; meaning of regional and, vii; national/international versus global/ regional space and, 15, 68, 88; postwar radionavigation and radiosurveying and, 239, 249, 251; in prewar Rand McNally atlases, 327n25; recalculation of the European triangulation and, 281; regional coherence of maps and, 67; regional ellipsoids and, 195; satellites and, 267, 268, 269–71, 269, 270, 275; sphericity in maps and, 78–80; twentieth-century shifts in geographic terms and, 15; universalism and, 110–11; US global consciousness and, 71; UTM and, 165–66, 178, 181, 191–93; in world atlases, 78; World War II and new global order and, 69 Renner, George, 326–27n13 representation: authoritative, 26–29, 38, 64, 66, 113, 115–16, 295; as both visual and political, 27–29, 64; color to represent elevation and, 35–38; geo-epistemology and, 26, 41, 293, 295; grid systems and,

Index

391

representation (continued) 124, 127, 139; impossibility of neutrality and, 41, 80; IMW and, 26–29, 35; legibility and centralization and, 2, 309n4; mapping sciences and, 18; versus mimesis, 113, 114, 334n130; origins of GPS and, 15; versus pointillist approach to space, 209; pointillist GPS mapping and, 281; political boundaries and, 15; postrepresentational approach and, 13–14; versus presentation, 3, 293; from provisional to definitive, 55–56, 323–24n86; as tool-making practice, 114; as transformative, 113; truthfulness of maps and, 2–3; twentieth-century cartography and, 13; as waning organizing framework, 68 Réunion, 264 Rio Treaty, 191 Robinson, Arthur, 112–13, 114, 115, 332n89 Roman Empire. See International Map of the Roman Empire Romania, 161, 291, 324n96 Roosevelt, Teddy, 32 Rosén, Karl, 150 Roussilhe, Henri: career of, 148; International Ellipsoid and, 154; international grids of, 148–51, 149, 161, 340n68; Pierre Tardi and, 158; Tissot’s projections and, 340n65 Royal Geographical Society of London, 56, 57, 324n88 RSDN-20 (РСДН- 20) radionavigation, 364n5 Russia: cartographic one-upmanship and, 32; distribution of IMW mapmaking responsibility and, 39, 46, 319n41; GLONASS (ГЛОНАСС) system and, 291; grids during World War I and, 337n30; radionavigation stations in, 291; TransSiberian Railway and, 186. See also USSR Russo-Japanese War, 125, 318n23 Sahara, 184, 231, 324n96 Samoa, 372n75 Sanceau, V. E. H., 314n40 Sassen, Saskia, 13 satellites and satellite navigation systems: as active versus passive, 268, 367n30; Aerosat project and, 368n37; Argos system and, 368n37; civilianization of, 266–68, 268, 269, 270, 368n33; commercial proposals and, 368n36; Compass system and, 256, 291–92; computer technology and, 261, 262–63; diversity among proposals for, 267, 367n28; Doppler effect and, 259–60; early lack of enthusiasm about, 249; first test satellite launch and, 259, 365n9; Galileo system

392

Index

and, 256, 287, 291–92, 374n98, 374n105; geosychronous orbits and, 269–70, 275, 291, 368n32, 368n34, 368n36; GLONASS (ГЛОНАСС) system and, 256, 291–92; ground installations supporting, 261, 266, 268, 268–70, 270, 275, 291, 292; international cost sharing and, 271, 368n37; modernization of GPS and, 290; Parus (Парус) system and, 364n5; passivity of GPS and, 253; Project 621B and, 273–76, 369nn41–42, 369nn44–45, 370n48, 370n50; redundancy and, 367n28; regional versus global coverage and, 267, 269–71, 269–70, 275; satellite mapping and, 314–15n41; Timation system and, 273–76, 369nn41–42, 369n44, 370n48, 370n50; traffic control and, 268–71, 268; Transit system and, 257–61, 259, 262–63, 266–67, 266, 271, 275–76, 281, 364n5, 365n9, 365n12, 365n14, 372n75; Tsikada (Цикада) system and, 364n5; Tsiklon (Циклон) system and, 364n5; two-way communication and, 268–69, 367–68n31; types of orbits and, 368n34; weather balloons and, 269, 367n27, 368n32. See also GPS Saudi Arabia, 245, 291 scale: aeronautical charts and, 49–50, 52, 104, 105, 322n72, 322–23n74, 346n37; carte de l’État major and, 126, 126; comparison of, 31; coordination between land, air, and naval forces and, 167, 172–74; General Bathymetric Chart of the Oceans and, 323n77; grid systems and, 123, 124, 130, 138; of the IMW, 25, 28–30, 32, 33, 40–41, 42, 49–50, 112, 320n53; International Geological Congress and, 49, 321n68; Jet Navigation Chart and, 105; map at a scale of 1:1 and, 123, 138; in map of telecommunications cables, 12; meaning of small and large scale and, vii; military mapping and, 81, 109, 126, 327–28n27; of questions about geographic knowledge, 1–2; US role in allied military mapping and, 85, 189; UTM and, 177–79, 181; WAC and, 66, 81; World Land Use Survey and, Gallery 7, 98, 330n72; World Population Map and, 97 Scheller, Otto, 353n12 Schrader, Franz, 46, 318n25 Schulten, Susan, 71 Scott, James, 140 Sea-Fix radiosurveying, 234, 359n66 Second Boer War, 335n11 Seeley, Stuart, 221, 357n40 Seven Years’ War, 326n5 Seychelles, the, 372n75

Shell Oil Company, 191, 234 ship navigation: direction-finding technology and, 215; distinction between shortand long-range navigation and, 361n78; latitude and departures in, 345–46n34; Omega system and, 260–61, 260; radionavigation and, 210; satellite navigation systems and, 269; systems planning approach and, 248. See also navigation; radionavigation (nonsatellite); satellites and satellite navigation systems Shiran radiosurveying, 232 Shokalsky, Yuly, 318n25 Shoran navigation and radiosurveying, 221, 231–32, 233, 236, 272, 359n66 Siam, 50, 139, 161. See also Thailand Sierra Leone, 11 Slipchenko, Vladimir, 282 Slovakia, 344n15. See also Czechoslovakia Smith, Neil, 12–13, 64 Soldner, Johann Georg von, 129, 336n20 Sonne radionavigation, 221, 223, 224–25, 228, 229, 230, 357n44. See also Consol radionavigation South Africa, 86, 88, 242, 291, 342n94, 343n1 South America: airway schemes and, 211; distribution of IMW mapmaking responsibility and, 319n41; ellipsoids for, 194–95, 194; imperialism in AGS work and, 58–59, 58; Pan-American Railway and, 210–11; territorial claims to the ocean and, 235; UTM and, 183, 184, 189–91, 195 Southeast Asia, 182, 192, 196, 221, 229, 349n75 Southern Hemisphere, 79 South Korea, 291 South Vietnam. See Vietnam sovereignty: over airspace, 8–9, 9, 213, 215; authority of a neutral map and, 93; cartography and, 3–4, 93, 100–101; function-based “sovereignty” and, 312n22; geographic and nongeographic bases of, 6; versus globalization, 13; grids and, 146–47, 150; IMW mapping and, 39, 44, 61; military mapping and, 84, 88; Montevideo Convention (1933) and, 8; radionavigation and, 210; territorial waters and, 9, 9–10, 235; versus territory, 4, 5–6, 7, 15–16, 295; Treaty of Westphalia and, 6, 310n12. See also jurisdiction; territory and the territorial state Soviet Union. See Russia; USSR Spain, 96, 110, 189, 223, 230, 242–43, 245, 291, 318n27, 324n96, 330–31n74 Spanish-American War, 318n23 Sputnik, 258, 259 Sri Lanka, 291. See also Ceylon

Stanford University, 267 State Plane Coordinate System: design and installation of, 140–41, 141, 338–39nn45–46, 339n52, 346n43; as professionalization project, 140, 143–44; relationship to law, state power, and territoriality and, 142–43, 145–46, 339n48; as replacement for city systems, 139; survey monuments and, 142, 143, 144–45; UTM and, 190, 196, 350n84 Stocks, Theodor, 323n82 Sudan, 89, 96, 191, 350n82 Sumatra, 330n65 Surinam, 348–49n70 surveying: advantages of grid systems for, 144–45, 339n54; Bowie method and, 157; carte de l’État major and, 130; Cassini’s carte de France and, 128–29, 130, 336n17; correction equations and, 348n63; eclipse of traditional methods of, 195; field surveys during war and, 135, 137, 138; geodetic datums and, 371–72n74, 372n75; GPS and, 280–81, 286; grid coordinates and national survey systems and, 183–84; of inaccessible places, 231; inertia in, 339n55; installation of UTM and, 185; Inter-American Geodetic Survey and, 191, 348–49n70; origins of rectangular coordinates in, 336n18; precision limit of, 339n46; survey monuments and, 124, 134–35, 137, 138, 142, 143, 144–45, 163, 172, 188, 340n59; TransSiberian Railway and, 186; US-Mexico border and, 145; US unemploymentrelief projects of, 142, 143; UTM and, 164, 189, 196. See also radiosurveying; triangulation Survey of Egypt, 36 Survey of India, 36 Sweden, 242, 291, 318n27 Switzerland, 33, 130, 139 Syme, George, 339n52 Syria, 161, 343n99, 347n57 système Lambert: advantages of, 132, 134; development of, 125, 148; Lambert projection for, 132, 133, 141, 337n27, 346n43; military grids and, 139; Nord de Guerre zone of World War II and, Gallery 10, 169, 169; scale and, 138; survey points and, 136 Tabula Imperii Romani. See International Map of the Roman Empire TACAN radionavigation, 361n79 Tanganyika, 342n93 Tanzania. See Tanganyika Tardi, Pierre, 157–61, 159, 186, 342nn87–88 targeting. See military targeting

Index

393

technology: agency of users in history of, 315n45; morality of technological change and, 296; patterns of everyday life and, 297–98; power relations and, 297, 375n2; state and nonstate status of, 14, 313–14n32; territoriality and, 297 Technology Audit Corporation, 366n25 Telecommunications Research Establishment (TRE), 225, 228, 360n75 Tennessee Valley Authority, 142 territory and the territorial state: before and after 1945, 7–8, 8, 100, 102, 295; bounded areas versus disconnected points and, 256; definitions of territory and the state and, 5–6, 310n9, 310n10; deterritorialization and reterritorialization and, 13; dissolution of empires and, 11; electronic coordinates and, 4, 14, 229, 235; Exclusive Economic Zones and, 9, 9–10; geo-epistemology and, 7, 208; geographic knowledge and, 297; globalization and, 11–13, 14, 298; GPS and, 282–83, 285–86; grid systems and, 123, 139, 145–46, 148; IMW and, 38–42, 55; as main site of mapping activity, 19, 26, 28–29, 46–47, 48; making and remaking of, 298–99; maps and, 4; maps as tools and, 18, 113–16; national borders’ first appearance on maps and, 7, 310–11n14; national/international versus global/regional space and, 15; navigation systems and, 208, 210; new spatial technologies and, 297; non-national territories and, 115; partition versus unification and, 7–8, 8, 311n17; patronage for the IMW and, 32; pointillism and, 15, 297; preservation of boundaries and, 7, 8; radionavigation and state autonomy and, 210; sovereignty over airspace and, 8–9, 9; standardization in the IMW and, 33–34; state versus capitalism and, 14; territorial waters and, 9, 9–10, 235; territory versus network and, 13, 15–16; territory versus sovereignty and jurisdiction and, 4, 5–6, 7, 15–16, 295; traditional versus nontraditional, 15, 314n36; Treaty of Westphalia and, 3, 6, 310n12; two- versus threedimensional, 8–9, 9; UTM and, 123–24, 165–66, 184–85, 190. See also borders; nation-states Thailand, 192. See also Siam Tharp, Marie, 235 Thompson, E. H., 342–43n96 Thrower, Norman, 26 Tillo, Aleksey, 30, 317n21 Timation satellite system, 273–76, 369nn41–42, 369n44, 370n48, 370n50 Time magazine, 70

394

Index

Tissot, Nicolas Auguste, 337n25, 340n65 Toran radiosurveying, 359n66 Transit satellite system: compared to other global systems, 257–58, 267; decommissioning of, 267; design and principles of, 259, 259, 365n9, 365n14; equipment for, 261, 262–63, 365n12; GPS and, 274, 275–76; ground installations for, 261, 266; military and civilian uses of, 260–61, 266–67; integrated navigation and, 267, 271; radiosurveying and, 281, 372n75; Soviet counterparts to, 364n5 Trans-Jordan, 172. See also Jordan Trans-Siberian Railway, 186 TRE. See Telecommunications Research Establishment (TRE) Treaty of Westphalia, 6, 310n12 triangulation: GPS and, 280–81; grid systems and, 142, 172, 339n46, 342n93, 343n98; junctions between neighboring countries and, 153–54, 153, 221; latitude and longitude coordinates and, 153–54, 153, 351–52n100; recalculation and consolidation of, 154–58, 157, 185–92, 187; rectangular coordinates and, 128, 132; size and shape of the earth and, 154, 193–94, 193; trilateration and, 232, 233; UTM and, 346n44, 347n56. See also geodesy; surveying trilateration, 232, 233–34, 235, 280 trimetrogon aerial photography, 83, 85 Trinidad, 264, 264, 266, 366n20 Truman, Harry S., 235 TRW Inc., 262, 366n25, 368n36 Tsikada (Цикада) satellite system, 364n5 Tsiklon (Циклон) satellite system, 364n5 Tunisia, 222, 224, 330n65 Turkey, 245, 291, 344n14 Turnbull, David, 113 Tuvalu, 372n75 Uganda, 192, 342n93 Ukraine, 291. See also Crimea UN Convention on the Law of the Sea (UNCLOS), 9, 9, 235, 312n19, 360n71 United Kingdom: African colonial surveying and, 191; aircraft landing systems and, 361n78; American antennas in Africa and, 266; aviation in international law and, 327n19; aviation reporting norms in, 354n19; British Admiralty charts and, 321n69; conformal map projections and, 337n24; disregard of IMW grid and, 324n96; division of mapmaking responsibility and, 38–40, 39, 56, 85–86, 86, 88, 331n84; East African possessions of, 158, 160; geodetic efforts of in India, 348n66; German intersecting-beam

systems and, 218; government supply of commercial navigation equipment and, 240–41, 360n75; Greenwich meridian and, 317–18n22; grid systems and aiming of artillery and, 134, 337n23; imperialism in IMW work and, 57–58, 59; increased wartime map coverage and, 80–81; International Ellipsoid and, 154, 161; interwar military grid for, 139; jamming of electronic signals during World War II and, 358n57; Korean War and, 347n59; location of German positions during World War I and, 138; 1913 Paris IMW conference and, 318n27; pace of IMW production and, 66; place-names in the IMW and, 44; postal service in, 215; postwar administration of the IMW and, 329n53; postwar IMW work and, 112; postwar radionavigation technologies and, 239–44, 240, 242, 246, 249; precision distance measurement and, 219–21, 220; predicted fire technique and, 134, 337n29; private companies’ radiosurveying and, 232, 234; radionavigation stations in, 222, 224, 242–43, 264, 291; reconstitution of the IMW and, 93; resistance to radionavigation and, 357n45; RNAV navigation rules and, 248; Second Boer War and, 335n11; standardization in the IMW and, 33–34, 110–11; territorial claims to the sea and, 235; trig beacons and trig lists and, 137; versus US radionavigation proposals, 209, 239–44, 246; UTM and, 176, 188, 192; volume of maps printed during World War I and, 338n32; World War I direction-finding technology and, 215; World War II bombing performance of, 225; World War II electronic coordinate systems and, 229–30; World War II grids of, Gallery 10, 168–71, 168, 173–74, 188, 196, 200, 345n29; World War II radionavigation and, Gallery 11, 205, 206, 207, 217 United Nations: ICAO and, 90; IMW and, 65, 92, 93, 111, 112; norms regarding territorial expansion and, 7; official emblem of, 77, 78; support for thematic mapping and, 330n67. See also UN Convention on the Law of the Sea (UNCLOS) United States: Aerosat project and, 368n37; aircraft landing systems and, 361n78; airway system of, 213, 214, 241; Allied division of mapping responsibility and, 85–86, 86, 88; Argos system and, 368n37; aviation in international law and, 327n19; command of the commons by, 282; diplomatic negotiations over antennas and, 261, 264, 266, 366n21; global

consciousness and, 70–71; Gloran system and, 360n77; government supply of commercial navigation equipment and, 241, 360n76; GPS and, 14, 209–10, 253, 254, 296, 375n108; ground installations for satellite navigation and, 261, 266, 275, 291, 292; IMW and, 32–33, 57–59, 63–64, 66, 93, 163, 319n41, 329n53; increased wartime map coverage and, 80–81; International Ellipsoid and, 151, 154; interwar military grid of, 140, 170, 344n12; Korean War and, 347n59; mapping policy in, 49, 321n66; maps of Latin America and, 61–62; Mercator projection and, 70, 71; military globalism of, 85, 232, 253, 254; military mapping and, 84; Monroe Doctrine and, 70; 1913 Paris IMW conference and, 318n27; nonterritorial hegemony of, 12–13; pilot autonomy versus air traffic efficiency in, 354n21; point-to-point navigation systems and, 209, 210, 213, 215–17; pragmatic internationalism and, 270–71; precision distance measurement and, 219–20; radionavigation and radiosurveying and, 208, 210–13, 214, 217, 219–21, 224, 226–27, 232, 234, 234, 239–44, 241, 242, 246, 249, 363n102; radionavigation stations in, 224, 242–43, 245, 264, 291; references for coordinates in, 345n33; RNAV navigation rules and, 248, 363n97; State Plane Coordinate System in, 139–44, 141, 190, 196, 338–39nn45–46, 339n48, 339n52, 346n43, 350n84; trade-off between the meter and Greenwich and, 317–18n22; transcontinental airways and, 210, 214, 352–53n8; unemployment-relief surveying in, 142, 143; United States National Grid and, 350n84; UTM and, 163–66, 184–95, 200; VOR phaseout and, 364n106; WAC as international base map and, 329n55; World War II and, Gallery 5, Gallery 10, 74, 168–71, 168, 173, 174–77, 344n12, 345n24 universalism: administrative, 256, 257, 271–78; of authorship and readership, 84, 94, 95, 102; cartography and, 67–69, 113, 115–16, 201; functional, 256, 257, 258, 267–68, 271; geographic, 256, 257, 258, 271; before GPS, 257–60, 264, 266–71; of GPS, 255–57, 290–93; of IMW, 25–26, 47, 63–64, 92–93, 110–12, 255; political, 256, 257, 270–71, 292; types of, 256, 257, 270; receivers versus transmitters and, 249–50; US military mapping and, 84; of UTM, 184–85, 189, 198, 200–201, 255. See also globalization and globalism; internationalism

Index

395

Universal Transverse Mercator (UTM) grid: adoption of, 164–65, 165, 343n1; AMS goals and, 185, 347n48; beneficiaries of, 190, 192; cadastral surveying and, 190, 196, 348n63; captured German data and geodesists and, 185, 186–87, 187, 188–89, 198; changes to, 181, 184, 198, 346n41, 351n93; civilianization of, 195–98; computers in calculations for, 187, 191–92, 347n53; datum adjustments and, 347n47; degrees versus miles and, 181, 346n39; design of, 163–67, 164, 176–84, 180, 182–83, 346n43; eastings and northings and, 345–46n34; ellipsoids used for, 183–84, 183, 186, 188, 194–95, 194, 346n44; errors at edges of grid zones and, 181, 182, 346n43; Eurocentrism of, 181, 346n43; GEOREF coordinates and, 346n35; Google Earth and Wikipedia and, 197; GPS and, 197, 206–7; IMW and, 163, 177, 178, 179, 181; on IMW maps, 333n111; installation of, 164, 165–66, 184–92; international boundaries and, 183–86, 196; Korean War and, 176; versus latitude and longitude, 166, 200; local, regional, and global considerations and, 178–79, 181, 184, 191–95, 295; on map of coastal Maryland, 122; Military Grid Reference System and, 178, 345–46n34; military targeting and, 185, 190, 192; multiple uses of, 121, 164, 184, 197, 198; origins of, 16, 17, 121; other grid systems and, 177–79, 179–80, 181, 183–86, 190–91, 251, 345n32; overlapping zones and, 181, 346n39; polar areas and, 164, 345–46n34, 346n41; purposeful delays in, 188, 347n57; radiosurveying and, 231–32; reasons for success of, 198; repetition of codes in, 179, 346n36; size of UTM squares and, 197; State Plane Coordinate System and, 190; territorial implications of, 123–24, 165–66, 184–85, 190; universalism of, 195–98, 255; WGS 84 datum and, 371–72n74; William Bowie’s recalculation of the European Triangulation and, 185, 186–88, 187, 347n53; World War II and, 19–20, 170 University of Michigan, 267 Uruguay, 348–49n70 US Aeronautical Chart Service, 83, 85, 104–5, 328n37 US Air Force, 103–5, 193–95, 199–200, 273–76, 369nn41–42, 369nn44–45, 370n48, 370n50. See also US Army Air Force US Army, 121, 140, 181, 196, 206. See also US Army Map Service (AMS) US Army Air Corps, 213 US Army Air Force, 81, 83, 90

396

Index

US Army Air Service, 213 US Army Corps of Engineers, 142 US Army Map Service (AMS): administrative reshuffling and renaming and, 328n37; ellipsoids created by, 194–95, 194; InterAmerican Geodetic Survey and, 191, 348–49n70; intercontinental survey connections and, 193, 193–94, 232, 234, 235; mapping sciences and, 17; mapping where it saw fit, 347n59; pace of production and, 83–85, 109; recalculation of the European triangulation and, 185–89, 187, 281; specialized map users and, 107; upstaging of interwar domestic grids by, 200; UTM and, 163–64, 164, 181, 183–84, 183, 186, 188–89, 191–92, 198, 347n48; World War II grids and, 172, 345n24 US Coast and Geodetic Survey, 140, 155, 187, 191–92, 321n66, 349n75. See also US National Geodetic Survey US Coast Guard, 240, 291 US Department of Defense: administrative universalism of GPS and, 257; budget issues and, 276; civilianization of GPS and, 276, 277; conflicts with NASA and, 269, 368n33; design and launch of GPS and, 272–73, 273, 274, 275–78, 369n40; formation of, 192; need for global survey system and, 192; sponsorship of GPS and, 16, 254, 273; universalist qualities of, 272; World Geodetic System and, 195 US Department of Transportation, 277 US Geological Survey: base maps and, 321n66, 321n68; grids shown on maps by, 335n4; maps by, Gallery 1, 24, 31, 122; UTM and, 196 US National Bureau of Standards, 212, 213 US National Geodetic Survey, 350n84 US Navy: geographic universalism of, 257, 258; versus NASA regarding satellite navigation systems, 271; Omega system and, 257, 258, 260–61, 264, 266; postwar proliferation of navigation technologies and, 244; seafloor mapping and, 235; Timation system and, 273–76, 369nn41–42, 369n44, 370n48, 370n50; Transit system and, 257–60, 259, 261, 262, 266; versus US Air Force in origins of GPS, 273–74, 276; UTM and, 179, 179, 199; World War II reference systems and, 173, 174–75, 176–77. See also Loran radionavigation USSR: availability of survey data from, 186; datum consolidation and, 351n94; development of GPS and, 275; distribution of ICAO mapmaking responsibility and, 101; domestic grid for, 139, 170; GLONASS (ГЛОНАСС) system and, 206,

256, 291–92; IMW and, 316n7, 324n87; international cooperation for radionavigation and, 249, 363n102; International Ellipsoid and, 155, 188, 347n56; military mapping and, 327–28n27; no maps for unsurveyed terrain and, 60; radionavigation stations in, 245; radionavigation and satellite systems with American counterparts and, 245, 249, 258, 363n102, 364n5; scale factor in grid systems of, 346n43; shooting down of Korean Airlines flight and, 277; Soviet Unified Reference System, 164–65, 165, 184–85, 196, 198, 200, 351n94; transliteration of place names and, 333n111; UTM and, 177, 183–85, 188, 196, 345n32; UTM coordinate mismatch in, 348n67; World War II and, 73, 170, 171, 344n14. See also Russia UTM. See Universal Transverse Mercator (UTM) grid Vanuatu, 372n75 Vidal de la Blache, Paul, 40 Vietnam, 1, 181, 182, 192, 242, 245, 344n6 VOR (VHF Omni-Range) radionavigation: origins and design of, 213, 357n44; pairing with DME and TACAN systems, 361n79; phaseout of, 251, 364n106; postwar expansion and debates over, 239, 241, 241–42; RNAV navigation rules and, 248 VORTAC radionavigation, 361n79 WAC. See World Aeronautical Chart (WAC) Wagner, Hermann, 30 Warsaw Pact, 69 Washington Island, Wisconsin, 31 Watson, Charles Moore, 318n25 Watson-Watt, Robert, 205, 230, 240–41, 356n35, 358n59 Weber, Max, 5–6, 14, 15, 310n10 Weiffenbach, George, 259 Weigel, Kasper, 342nn86–87 Western Sahara, 8, 311n17 West Germany, 242, 331n84, 343n1. See also Germany Westinghouse Electric Corporation, 269, 366n25 WGS. See World Geodetic System (WGS) Wikipedia, 197, 351n92 Willkie, Wendell, 71–72, 75, 77, 80 Winterbotham, Harold, 59, 314n40, 338n32, 342n93 Wold, Karsten, 342n90 Wood, Denis, 113–14, 320n51 Works Progress Administration, 142, 349n72 World Aeronautical Chart (WAC): coverage of, 81, 101; distribution of mapping re-

sponsibility and, 88, 99–100, 101; fading importance of, 109; general versus specialist readers and, 103–5, 107; graphic uniformity and, 85; ICAO and, 66–67, 90, 91, 107, 108; IMW versus, Gallery 6, 68, 81, 82, 83, 89–90, 91, 92–93, 110–12, 328n32; indicators of elevation and, 106; initiation of, 66–67; as international base map, 90, 329n55; as map for pilots, 89, 99, 105; Mexico map of, Gallery 6, 82; Operational Navigation Chart and, 105, 332n100; UTM strategy and, 188–89 World Geodetic System (WGS), 195, 232, 281, 314n41, 371–72n74, 372n75 World Land Use Survey, Gallery 7, 94–96, 98, 330n72, 331n75 World Population Map, 94–96, 97 World’s Fair (1938), 62 World War I: aiming of artillery in, 130, 132, 133, 134–35, 137, 138, 167, 337n29; Allied versus German surveying during, 138; Allies’ use of système Lambert and, 139; Battle of Amiens and, Gallery 8, 119, 120; Battle of Cambrai and, 337n29; Battle of the Marne and, 125; Bruchmüller’s attack on Riga in, 337n29; directionfinding technology during, 215; German Spring Offensive and, 134; grid systems in maps for, Gallery 8–9, 119, 120, 121, 124–27, 130, 131, 132, 133, 134, 335n1, 337n23; IMW and, 28, 34; mismatch between neighboring countries’ coordinates and, 130, 154; quantity of maps printed during, 1, 338n32; rate of fire during, 138, 338n35; Russian mapping and, 337n30; as truly global war, 326n5; versus World War II, 326n6 World War II: AGS Map of Hispanic America and, 61; Allies’ advance from Normandy and, 228; Battle of Britain and, 219; coordination across geographic scales and, 172–74; data captured during, 172, 229–30; division of mapping responsibility and, 85–86, 86–87; electronic grids during, 221, 222–24, 224–25, 226–27, 228–30, 229; force multipliers and, 375n4; as geographically transformative, 69–70, 79, 88–89, 115, 167–68, 175–76, 185–86, 326n6; German beam system during, 217–19, 219, 224, 230; Germany and oil supplies in, 73; grid systems during, Gallery 10, 163, 166–76, 168–69, 174–77, 342n92, 344n12, 344nn14–15, 345n24; ground installations for radionavigation during, 228–30; IMW and, 20, 27, 64, 65, 89; jamming of electronic signals during, 230, 358n57; Pearl Harbor and, 81; precision distance

Index

397

World War II (continued) measurement and, 218–20, 220; quantity of maps printed during, 1; radionavigation during, Gallery 11, 205, 206–7, 217; shooting across grid and survey system boundaries and, 171–72; surprise bombing of Coventry in, 218; targeting systems during, 217–21, 224; Tizard Mission during, 357n46; as truly global war, Gallery 5, 69–70, 74, 168, 326n5; UTM and, 20, 177; versus World War I, 326n6

398

Index

Wotan I radionavigation, 356n32 Wotan II radionavigation, 356n36 X-Gerät radionavigation, 218, 219, 356n32 Y-Gerät radionavigation, 219, 219, 356n36 Yugoslavia, 324n96 Zambia. See North Rhodesia Zyklop radionavigation, 356n34