Shaping Romance: Interpretation, Truth, and Closure in Twelfth-Century French Fictions 9781512801057

Table of contents :
Contents
Acknowledgments
Introduction
1. Truth in Disguise: The Voice of Renarration in the Folie Tristan d'Oxford
The Mediation of Signs
Iseut's Process of Interpretation
Tristan's Process of Interpretation and Recognition
The Sound of Voice in a Written Text
The Voice of the Oxford Poet in Intertextual Dialogue
The Corporeality of Truth
2. Putting Off the Ending: Thomas and the Legend of Tristan and Iseut
Doubling, the Key to Experience, Knowledge, and Identity
Two Final Visions of Death
The Narrator's Double View
The Epilogue
Thomas and the Tristan Legend
3. A Case for mise en abyme: Chrétien's Chevalier de la Charrete
The Tournament at Noauz: mise en abyme
Intratextual Recalls
The Process of Recognition: The Herald
The Process of Recognition: The Queen
Delay and Characterization of the Lovers
Authors, Story, Public, Patroness
Yvain and the Charrete: Patterns of Multiplying Stories
Tristan, Lancelot, and the Measure of an Arthurian Ideal
The Fictionality of Romance
4. The Interplay of Gender and Genres in Partonopeu de Blois
The Narrator's Double Stance
History and Romance
Beauty and Birth
Crisscrossing of Gender and Genres
The Interplay of Male and Female Power, Male and Female Beauty
Invisibility, Power, and Knowledge
Deception, Judgment, and the Role of Gender
The Mobility of Categories
The Poetics of Continuation
5. Textual Identity and the Name of a Collection: Marie de France's Lais
Lai and Romance
Selection and Substitution: Guigemar
Recognition and Eliduc
Marie's Textual Identity: Names and Titles
Marie's Fusion of Voices
Orality and Writing, Orality in Writing
Closing and Opening the Collection
Conclusion
Notes
Bibliography of Works Cited
Index
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
W
Y
Z

Citation preview

Shaping Romance

University of Pennsylvania Press MIDDLE AGES SERIES Edited by Edward Peters

Henry Charles Lea Professor of Medieval History University of Pennsylvania

Shaping Romance Interpretation, Truth, and Closure in Twelfth-Century French Fictions Matilda Tomaryn Bruckner

University of Pennsylvania Press Philadelphia

This publication was assisted by a grant from the Trustees of Boston College. Portions of Chapters i and 2 first appeared as "The Folie Tristan d'Oxford: Speaking Voice, Written Text," Tristania 7 (Autumn 1981-Spring 1982): 47-59 and "The Representation of the Lovers' Death: Thomas' Tristan as Open Text," Tristania 9 (Autumn 1983-Spring 1984): 49—61. Reprinted by permission. A portion of Chapter 5 first appeared as "Strategies of Naming in Marie de France's Lais: At the Crossroads of Genre and Gender," Neophilolqgus 75 (1991): 31-40. Reprinted by permission.

Copyright © 1993 by the University of Pennsylvania Press All rights reserved Library of Congress Cataloging-in-Publication Data Bruckner, Matilda Tomaryn. Shaping romance : interpretation, truth, and closure in twelfth-century French fictions / Matilda Tomaryn Bruckner. p. cm. — (Middle Ages series) Includes bibliographical references and index. ISBN 0-8122-3169-4 i. French poetry—To 1500—History and criticism. 2. Tristan (Legendary character)—Romances—History and criticism. 3. Narrative poetry, French—History and criticism. 4. Romances—History and criticism. 5. Fictions, Theory of. 6. Rhetoric, Medieval. 7. Closure (Rhetoric). I. Title. II. Series. PQI78.B78 1993 841'.030901—dc20 93-2069 CIP

For my family

This page intentionally left blank

Contents

Acknowledgments

ix

Introduction

1

1. Truth in Disguise: The Voice of Renarration in the Folie Tristan d'Oxford

12

The Mediation of Signs Iseut's Process of Interpretation Tristan's Process of Interpretation and Recognition The Sound of Voice in a Written Text The Voice of the Oxford Poet in Intertextual Dialogue The Corporeality of Truth

13 17 19 25 30 34

2. Putting Off the Ending: Thomas and the Legend of Tristan and Iseut

37

Doubling, the Key to Experience, Knowledge, and Identity Two Final Visions of Death The Narrator's Double View The Epilogue Thomas and the Tristan Legend

38 41 45 50 53

3. A Case for mise en abyme: Chrétien's Chevalier de la Charrete

60

The Tournament at Noauz: mise en abyme Intratextual Recalls The Process of Recognition: The Herald The Process of Recognition: The Queen Delay and Characterization of the Lovers Authors, Story, Public, Patroness

61 65 70 73 78 84

viii

Contents

Yvain and the Charrete: Patterns of Multiplying Stories Tristan, Lancelot, and the Measure of an Arthurian Ideal The Fictionality of Romance 4. The Interplay of Gender and Genres in Partonopeu de Blois

90 94 104

109 10 113 117 120

The Narrator's Double Stance History and Romance Beauty and Birth Crisscrossing of Gender and Genres The Interplay of Male and Female Power, Male and Female Beauty Invisibility, Power, and Knowledge Deception, Judgment, and the Role of Gender The Mobility of Categories The Poetics of Continuation

126 133 138 144 151

5. Textual Identity and the Name of a Collection: Marie de France's Lais

157

Lai and Romance Selection and Substitution: Guigemar Recognition and Eliduc Marie's Textual Identity: Names and Titles Marie's Fusion of Voices Orality and Writing, Orality in Writing Closing and Opening the Collection

163 170 177 183 189 199

Conclusion Notes Bibliography of Works Cited Index

207 227 267 285

157

Acknowledgments I acknowledge with pleasure the debts of gratitude owed to colleagues and friends who have helped me in writing this book. First and foremost, to Peter Haidu, Donald Maddox, and Nancy Regalado, who read and gave extensive comments on earlier drafts of the five chapters, I offer deepest and heartfelt thanks for their many suggestions accompanied by hard questions, occasional disagreements, indispensable insights, and much appreciated encouragement. Roberta Krueger, Nancy Jones, and Sarah White were most generous in their support and suggestions on various parts of the manuscript. Kevin Brownlee and Rachel Jacoff played Virgil to my Dante by initiating me into the mysteries of seeking a publisher for the book, whose final revisions were significantly improved by the probing commentary of the two readers chosen by the press, Keith Busby and Douglas Kelly. Nancy Regalado's suggestions for the final revision of the title, Introduction, and Conclusion were as essential as her earlier ones to make sure that I put out front what belonged in the beginning instead of hiding it until the end. I would also like to thank the institutions that have made it possible for me to complete this book: the American Council of Learned Societies for a fellowship in the 1985—86 academic year, and Boston College for its generous sabbatical in 1990-91 and its continuing support through the process of publication. I am most grateful to the University of Pennsylvania Press for accepting the book for publication and offer a special thank you to Jerome E. Singerman for his support and guidance in the various stages of preparing the manuscript. I would also like to thank Victoria Jordan for her research assistance and Lisa Port White for her help in preparing the index. Finally I would like to thank my family for being there: my mother for her unflagging assistance in helping me manage the intricate demands of family and professional life, my sons for their patience and impatience, my husband for his understanding and support of my passionate attachment to the complexities of life and literature.

This page intentionally left blank

Introduction

This is a book about shapes and shaping in twelfth-century French romance fictions. I use the term "romance fictions" here to include both romances as normally designated and the short stories, contes and lais, which use the same kinds of materials and operate in the same context for medieval writers and their public. Both types of narrative are clearly perceived as fictions, unbounded by certain principles of reality and yet related in some way to the lived experiences of their public.1 If romances and stories frequently overlap in the medieval context, they nevertheless retain a kind of distinctness that helps the modern reader appreciate better the specific character of each, as well as the considerable play in their variations, which link as much as they distinguish the two types of fictions. I have analyzed the very particular and complex patterns that give shape to five different works, each one chosen for its distinctly different shape within the range of twelfth-century narratives. They are all roughly datable from the latter half of the twelfth century, the period when romance has reached beyond its initial form in the romans antiques (the Old French "translations" from Latin epics) and has begun to assume the many forms, sizes, and shapes that will be played out and transformed in succeeding centuries. Surely one of the most impressive aspects of the twelfth-century Renaissance is the explosion of literary works in the vernacular, and that explosion is nowhere more noticeable than in the popular appeal of romance, with its gift for experimentation in form and its taste for a wide array of materials combined through intergeneric mixing of all sorts.2 What I offer in the following chapters is by no means an exhaustive or systematic survey of twelfth-century romance, but rather a small sample of works that crystallize some of the most brilliant moments of romance writing in the period. Each of the chapters represents an experiment in close reading, in which I analyze the many-faceted and subtle patterns of five romance fictions, in order to show the depth of complexity woven into their designs, their resistance to any monolithic or reductive conception of what romances may do or be.3 If I were to characterize in a nutshell the main critical issue that animates the book throughout, I would point to the role played by repetition and its variations, the values and

2 Introduction meanings we assign to them in medieval textuality. But in each chapter my first allegiance is textual rather than theoretical: the specific coherence of each text chosen determines the direction and results of my analysis and interpretation. Of course, the intertextual nature of medieval writing requires modern readers, as well as the medieval public, to ground our reception and understanding of individual texts within the interplay set up with the surrounding literary traditions, sometimes with specific intertexts evoked implicitly or explicitly, sometimes with the general models associated with a given genre. That perspective informs the individual readings of each chapter and establishes a kind of implicit dialogue between the chapters as well, insofar as they explore texts which respond to each other or to common stories and traditions. Although each work is analyzed with respect to its own coherence, similar problems and questions reappear in varying guises in each of the five chapters, as I explore the problem of signs and their multiple interpretations from multiple points of view, the role of language and the transposition from action to discourse, from oral tale to written romance, the mise en question of categories as they play through gender and genres, ending up more often than not as non-disjunctive oppositions, the linking of truth and fiction, history and romance.4 The twelfth-century fascination, not to say obsession, with the Tristan story plays an important role in this exploration, as do the connections between experience, knowledge, power, and identity, the split between author and fictionalized narrator, and the interplay between formal shaping and semantic possibilities, as author/narrators constantly play with closing or continuing stories and their meanings. As a reader I am no doubt particularly drawn to certain aspects of these texts as indicated by the set of questions raised, features of romance writing that make it such a complex and protean genre. Their recurrence from text to text suggests that such characteristics, as well as the issues they entail, play an especially compelling role, not only in these particular romance fictions but in the rich field of romance in the twelfth and later centuries. If the main thrust of my analyses aims to demonstrate how the complexity of these narratives emerges in multiple ways, not easily systematized, common features of romance nevertheless appear across the range of their differences, recognizable despite or perhaps even because of their variability.5 In the Conclusion, I shift my focus by moving from the specifics of the five texts to some of the general features they share, in order to sketch out critical models sufficiently flexible to make sense of twelfth-

Introduction

3

century French romance. Although the selection of texts analyzed is small, I hope to raise questions and propose hypotheses that invite testing by others working on these and other texts to verify their promise for individual romance fictions or romance as genre. Much has been said in recent years about the alterity of the Middle Ages, a concept that usefully warns us against assuming we can read medieval literature or understand medieval culture as if they were no different from our own contemporary forms, ideas, and experiences.6 In Hans Robert Jauss's discussions, alterity is also a way of recognizing and channeling our modern response to medieval texts, a way to ask why they continue to interest and teach us, even in a world so radically different from that of twelfth-century authors and public. Jauss's comments suggest that we medievalists have a responsibility to reach a wider audience, not only among scholars and critics but among general readers. We need to catch their attention and introduce them to the delights, as well as the profound lessons, of medieval literature for a modern public, just as we need to deepen our understanding of medieval texts thanks to the insights of contemporary culture.7 Without trying to make medieval literature fit into any Procrustean bed of theory, we can take advantage of the tools and concepts available in modern critical discourses, as well as adjust and nuance the findings of modern theory where it does not adequately take into account or represent medieval narrative.8 When I speak of delights, the pleasures of romance come readily to mind. It strikes me as particularly important to remind the twentieth century that not everyone living in the Middle Ages was a monk or a priest, and even the religious seemed to derive a good deal of enjoyment from the likes of Tristan, Arthur, and Perceval, if we judge by the complaints from some of the more severe among them.9 All too often we still see the Middle Ages through the eyes of ignorance and nostalgia, as we project a golden age of plenitude in the past when everyone knew his place in society, when the Church ruled people's lives, their minds, and their faith.10 The medieval period appears in this view as a monolith completely dominated by a single ideology, but we should remember that the so-called Middle Ages cover hundreds of years, during which Church power and doctrines changed considerably, indeed varied considerably even within a given time and certainly across the widely dispersed geographical limits of medieval Christendom. Romances, vernacular lyric and fabliaux, feudal institutions and values, as well as the history of individual lives and communities, give ample testimony to the existence of concerns and values,

4 Introduction ideas and desires that conflicted or competed with, or simply coexisted alongside of, the views and teachings of the Church.11 Romance offers not only the delights of good stories, wonderfully told, filled with the extraordinary exploits of fabulous knights and beautiful ladies, but an important and privileged place to look deeper into different periods of the Middle Ages, to see how such alternate discourses are pursued, what issues they raise and explore. Romance may seem to some as too frivolous, too fictional, to offer a real view of social history. Maurice Keen considers and rejects romances as a source of information for his inquiry into the social reality behind that evocative word, "chivalry3': "An ideal of knighthood culled from what appears so often to be essentially a literature of escape is scarcely a promising model for a social historian to make much of.5'12 Yet it may not be romance's obviously fictional trappings that pose the problem, so much as the way we read them to see where history and fiction intersect.13 In some of the chapters that follow, I shall argue for seeing twelfth-century romance as a kind of mise en abyme of contemporary feudal society, where the play of fiction allows the expression—however indirect—of fundamental conflicts within and between different value systems or ways of thinking, operating primarily in the secular domains of twelfth-century society. Romance fictions thus allow a kind of free space for experiments—in forms and ideas—that may function as a way to redirect and change the society it mirrors. Neither simply mimetic, nor dynamically cut off from real life, romance is an integral part of the dialectic of history.14 If this is a book about shaping romance, its own shape necessarily reflects some of the formal and semantic characteristics of such arrangements. The first two chapters set up a kind of "point counterpoint"15 in terms of a particular problematic. Each focuses on a different version of the Tristan legend among the many available in verse narrative of the twelfth century: the first chapter, on one of two extant Folly episodes, the Folie Tristan d'Oxford', the second, on Thomas's Tristan^ or rather what is left of it in eight fragments recorded in manuscript. Both of these works pose the typical romance problem of signs and their interpretation, but each one offers a radically different response to the multiplicity of signs. The process of recognition set up between Tristan and Iseut in the Folly's single episode, which brings them together for a brief reunion during Tristan's exile, introduces voice as an expression of being. Orality is transposed and inscribed in a written narrative that nowhere explicitly recognizes its own textuality. Tristan's voice, first disguised and then revealed to

Introduction

5

Iseut, ultimately establishes a kind of transparence that incorporates the lovers' truth and puts an end to the inexhaustibility of signs, the antics of Tristan as fool, and the renarrations that move the lovers' story into discourse and legend. The poet's own voice written into his narrative plays with his public outside the text with the same resourceful trickery that Tristan demonstrates in court and chamber. Through the manipulations of Tristan and the anonymous poet, the Folly episode reaches a complete resolution and offers a sense of closure, on both narrative and semantic levels, to a degree that is rarely available in romance writing. On the contrary and somewhat more typically, Thomas's romance ends conclusively with the description of the lovers' deaths followed by the author's epilogue, and yet it continues to reverberate with non-resolution. My analysis focuses on the representation of Tristan's and Iseut's deaths and the conflicting values assigned to their final union, in order to determine the role of doubling, as it occurs at every level of Thomas's text. Behind the desire to duplicate the other's experience, to share the other's knowledge, lies the belief that sameness establishes the lovers' identity as a couple and difference separates them, as it introduces ambiguity, multiplicity, and uncertainty. Paradoxically, doubling furnishes the basis for both unity and multiplication within the couple, which expands to the foursome constituted by Tristan and King Mark, Iseut la blonde and Iseut as Blanches Mains. The inexhaustible quest for unity and identity in the couple, reflected in the combination of Tristan's and Iseut's own double views of their union in death, remains unresolved in part through the narrator's own refusal to evaluate them, since he claims to have no experience like theirs and therefore no knowledge to judge them, despite the long interventions in which he takes on the appearance of omniscience and displays his skills as psychologist of love. Like the critical tradition that alternately sees Thomas as condemning or condoning love, the narrator's stance vis-a-vis the story he tells remains double, engaged and disengaged. The juxtaposition between the final description of the lovers' embrace in death and the epilogue's closing remarks on the truth of Thomas's version and the character of his public, shifts the questions of ermnce and novelerie into the domain of storytelling, the intertextual dialogue between Thomas and other versions of the Tristan legend. Unlike the clear and satisfying resolution offered by the Folie Tristan ^Oxford, Thomas's romance leaves its public engaged in the same kind of open-ended process that characterized the lovers' own search for unity through identity and difference. The object of our pursuit and the materials evaluated may dif-

6 Introduction fer, but the process itself remains constant and constantly deferred to a future endpoint. The process of evaluation and interpretation connects our world to that of the characters, as it is both distanced and mediated by the role of the fictional narrator. If Thomas's open-ended ending appears to be more typical of romance than the firm closure of the Oxford Folly, the detailed analysis of each of the five works included here will demonstrate that in every case we can see complex variations on both resolution and non-resolution, as we focus on different levels and aspects of these romance fictions. The pattern of "point counterpoint" operates inside as well as between texts. Once the two opening chapters have set up a dialogue between different versions of the Tristan legend and different kinds of romance writing, the placement of a chapter on Chretien de Troyes's Chevalier de la Charrete at the center of this book is doubly appropriate. Chretien's romances have frequently been used as models for all twelfth-century romance, or even all medieval romance, and while this privileging of Chretien's corpus certainly leads to some misconceptions about his own works, as well as the wide variety represented within the catch-all category we name "romance," it does recognize Chretien's quite remarkable and verifiable influence on subsequent romances, the amazing breadth and depth of his richly provocative experiments in romance writing. If Chretien's romances have become a standard by which we measure twelfth-century romance, what he offers as "the standard twelfth-century romance" is by no means standard, if that term implies something humdrum, the mere repetition of a traditional story, lacking in invention. One could indeed make the argument that Chretien's romances are so complex as to defy our use of them as norm, whether literary or social. And yet there is some advantage to be gained in taking the most complex case as one's model, if for no other reason than to make it as hard as possible to sink into abusive generalizations or simplistic and reductive abstractions. In the quest for complexity, Chretien's corpus offers an embarrassment of riches, but surely we need not be too embarrassed if the choice falls on Lancelot. Not only does this romance fall somewhere in the middle of Chretien's romance production, not only does it offer a profound recasting of the Tristan story, one of the central obsessions of the twelfth century and the starting point for this study of romance; the Chevalier de la Charrete places at its own center, at least logically and metaphorically, a mise en abyme of itself as romance, a replication and refraction that ultimately ask us to read the fictions of romance as a mise en abyme of contem-

Introduction

7

porary feudal society, an exploration of the problems and issues raised in the transposition of an Arthurian ideal into the competing and conflicting value systems of the romance public's own world. Through the representation of Lancelot as a touchstone who reveals the paradoxical and contradictory norms of his society, the Charrete explicitly poses the question of romance as "Case."16 Moreover, the frame of Prologue and Epilogue sets up a series of parallels and duplications between the characters in the story and the figures involved with romance production, such that the problem of signs and their disjunctions bridges the worlds of fiction and lived experience: the diegetic and extradiegetic worlds mirror each other and lead us further into the serious games of medieval textuality. The chapter itself begins with the intratextual, when it analyzes the tournament episode as a mise en abyme of the entire Charrete\ it then focuses on the extratextual, as it is inscribed in the relations between authors and patron, story and narrators; and finally it follows a series of intertextual patterns set up first within Chretien's corpus and then between the Charrete and the Tristan legend. As always in romance, the issue of closure is many-leveled, especially as it is complicated here by the play of intertextuality and the inscription of mouvance within the production of the Charrete as a romance already containing a continuation. With Godefroi's epilogue, the romance both refuses and invites further continuation through the unresolved issues implicitly and explicitly raised. As a complete romance with an added continuation specifically invited by the first part's epilogue, the anonymous Partonopeu de Blois acknowledges and goes beyond the challenge posed by Chretien's Charrete. Although not well known today, Partonopeu is, like the Charrete, one of the seminal works for medieval romance, not just in France, but throughout Europe, including England and Scandinavia. It deserves a wider readership today, despite its somewhat intimidating length. Not only its ample proportions but its focus on the intersection of history and romance, Latin learning and vernacular public and pleasures, recall the romans antiques and their interest in translatio and amplificatio. Indeed, Partonopeu grows by blending together a wide variety of elements from different genres, as if to invent a series of experiments in fusion that cross over boundaries, on the level of formal shaping as well as on that of meaning and its possibilities. At the heart of these reinventions is the double stance of the narrator, which combines lyric engagement and romance distance: Partonopeu's mirroring of diegetic and extradiegetic worlds passes through the character of the narrator, whose own love story parallels and contrasts with that of his

8 Introduction characters. As a rewriting (at least in part) of the Cupid and Psyche story, with the male and female roles reversed, Partonopeu explores the interplay between gender and genres as categories in contemporary literature and culture. In particular it offers an analysis of beauty as a counterpart to the "fils a vilain," the base-born person raised out of his proper place in society: beauty, at first associated with noble birth and lineage (that of Partonopeu himself), ultimately joins inheritance and individual performance in non-disjunctive opposition; both are clearly necessary to the well-being of feudal society. The hiding of Melior's own beauty behind the taboo of invisibility in the first part of the romance, as well as her deceptions and manipulations in the second part, lead us to explore the important nexus of power, knowledge, and identity, as Partonopeu replays, with suitable shifts and variations, the story of Adam and Eve in the Garden of Eden. The romance public, as well as the characters, are confronted by a series of puzzles, where the difference between good and bad examples must be recognized and acted upon; ultimately we may learn that these categories and others are less stable, more problematic, than we first recognized. The mobilization of categories, whether social or political, literary or philosophical, cultural or generic, invites us to meditate on choices and examples, to think as experimentally as the author/narrator's own example demonstrates. This is particularly compelling, at least in terms of his experiments with the form of romance, in what we might call the romancer's "poetics of continuation," which, along with the Perceval Continuations, anticipate and prepare for the enormous continuations of romance that constitute the prose cycles of the thirteenth century. It may seem odd to associate Marie de France's Lais with those same cycles of stories filled with multiple heroes and quests, yet I would argue that Marie's collection of twelve lais does indeed operate as a kind of miniromance cycle. Marie is known—and well known to twentieth-century readers—as a miniaturist first and foremost: I would not underplay her skills in the art of brevitas, but I would point out how her arrangement of /##, with the General Prologue as introduction and the first and last stories as boundary markers, not only takes on the dimensions of romance but introduces in its transposition of oral tales into written discourse the same issues and problems we see elsewhere associated with romance writing: the relation between Latin and vernacular, oral and written traditions, the interpretation of signs and the role of renarration as they are combined in the process of recognition, the reinvention of the Tristan story and the displacements of gender and genre, the doubling of diegetic and extra-

Introduction

9

diegetic worlds. Marie accomplishes this mutual integration of author/ narrator and text, in part through the pattern of names and titles elaborated across the collection of twelve lais, in part through her blending of voices, as the characters speak through Marie's own distinctive voice as oral storyteller and writing author. As female clerk, Marie combines in her Lais both the authority of lived experience associated with orality and the textual authority of a written tradition. She demonstrates in her own way the intersection and interdependence of categories that we discover elsewhere in romance's tendency to undermine any disjunctive oppositions. The shape of the Lais as a collection that integrates both the repeated discontinuities of the short tales and the continuity of the assemblage assured by the framing prologues and epilogues raises with particular insistence the issue of closure, the impetus for continuing or closing a given narrative program or search for meaning. As such, it seems especially appropriate to situate an analysis of Marie's Lais as the closing chapter of this book. Her short narratives recall (through their particular variations) the interplay with romance set up in the opening chapter, where the single episode of the Oxford Folly played with and against the different romance versions of Tristan and Iseut. Together the first and last chapters form a useful frame within which to consider the middle chapters specifically focused on three different romance texts. Romance as genre characteristically plays with and across the borders of what is and is not romance. The Conclusion takes up for general discussion two of the most significant critical issues that recur throughout the chapters: the interplay between the diegetic and extradiegetic worlds achieved through the triangle of relationships linking author/narrator, story, and public, and the question of closure, whether rhetorical, narrative, or semantic. As we go around the three poles of the triangle connecting author/narrator, story, and public, it becomes apparent that the same issues arise in each position. Through positive and negative parallels, whether based on structure or content, these three "actors" in the drama of romance are separated from each other, while at the same time they are mutually implicated in the same processes or problems. This mirroring depends first on the relationship between the author and his or her fictional double, the narrator, and secondly on the varied interplay between those two figures and the mediation one or both of them provide(s) to disengage and engage romance story and public. We might usefully remember here the opening remarks of the knight Calogrenant, who retells his (mis)adventure of seven years past and thus

io Introduction initiates Yvain's own story in the Chevalier au Lion. Here is a narrator figure who supplies the missing prologue in Chretien's romance, as he invites his listening public to attend his tale with open ears and understanding heart: "Cuers et oroilles m'aportez, car parole est tote perdue s'ele n'est de cuer entandue. . . . Et qui or me voldra entandre, cuer et oroilles me doit randre, car ne vuel pas parler de songe, ne de fable, ne de man^onge." (w. 150-53,169-72) ("Bring me your hearts and your ears, for speech is lost if it is not heard by the heart. . . . The one who wants to hear/understand me must give me heart and ears, because I do not want to speak of dreams or fables or lies.") Of course Calogrenant is located squarely in the fiction of Chretien's Train, but his recit not only offers a mise en abyme of the entire romance, it makes explicit the same interplay between fable and what is not mangonge that characterizes not only this romance but twelfth-century romance in general.17 In so doing, Calogrenant calls for his public's attention beyond the superficial level of an immediate enjoyment, associated with the ears and the passive act of listening, and enjoins his listeners to wake up their hearts, to take and enclose and keep ("prendre, et anclorre, et retenir," v. 164) the words heard and submit them to a further act of entandement constituted by an understanding deep in the heart.18 His invitation, which parallels that of many romance narrators, suggests that within the heart of fiction itself there are matters of vital concern to the romance public, if only it knows how to take romance fictions properly to heart. A fictional narrator like Calogrenant emphasizes both the distancing set up in the contract romance establishes between authors and their public, who exchange a good story from elsewhere and long ago, but also the rapprochement through bridges that crisscross the spaces between each of the positions around the triangle. The publics inside and outside the fiction encounter similar problems and are thus compelled to follow the same process of entandement, from passive listening to active recognitions and interpretations.

Introduction

n

If I were forced to formulate a single goal for this study of twelfthcentury romance fictions, I might express it as a desire to prevent modern readers from assuming that the Middle Ages produced only, or even most characteristically, closed texts. The enormously popular romance genre, with all its complexities, ambiguities, and irreducible differences, is surely eloquent testimony to the contrary. But that testimony cannot speak if it is not read, cannot be heard if it is approached by readers whose presuppositions prevent them from hearing. Hence my desire to demonstrate here the sophistication of romance fictions, which may demand an equal sophistication from its public, whether we judge its complexities in literary or intellectual, cultural, or philosophical terms.19 The second part of the Conclusion thus brings together and explores further the issues posed by the question of closure in romance fictions. Medieval textuality requires the modern reader to rethink the notions of fragment and whole, continuity and discontinuity, as it locates the play between closure and openendedness on a multi-dimensional continuum operating on many different levels of text and context, form and meaning. The order of chapters presented here obviously has little to do with the five works' chronological order of composition. Although Partonopeu de Elois was undoubtedly written after Chretien's romances, and the Folie d^Oxford after Thomas's Tristan, there is no way to determine with any reliability the order of precedence among Chretien, Thomas, and Marie.20 For us in the twentieth century there may be no order of precedence, or even preference, among these great moderns of the twelfth-century Renaissance, but whoever would learn about the possibilities and the rich complexities of romance fictions in that period must lend an attentive ear to the intertextual dialogue among them, anonymous and named authors who set up among their texts models for reshaping romance, in short forms and long, that will continue to reverberate for centuries after them. Lending attentive ears—and eyes—that is the project to be pursued in the next five chapters.

i. Truth in Disguise: The Voice of Renarration in the Folie Tristan d^Oxford

In every extant version of the Folly episode—except the Folie Tristan #Oxford—Iseut recognizes Tristan when he shows her the ring she herself gave him at an earlier moment of parting.1 In the Oxford version, showing the ring fails to produce recognition because Tristan continues to speak in a disguised voice. Iseut finally recognizes the fool as her lover only when Tristan returns to his own speaking voice—at which point the Oxford Folly returns to share in the joyousness of the lovers' interlude that completes all versions of the episode. Given this radical difference in the detail of Tristan's disguise and (final) revelation, we need to begin studying the Oxford version's particular coherence by asking why the change in Tristan's voice plays such a vital role first in concealing and then finally in revealing his true identity. What are the necessary links between voice and truth, truth and narration in the Folie Tristan d'Oxford* As I attempt to answer that question we will discover how voice operates as a problem for Tristan and Iseut within the story of the text and as a problem of representation posed for and by the author of the text. In the introduction to his edition of the Oxford Folie, Ernest Hoepffner has already remarked that the change in the ending is carefully prepared by the poet as early as v. 212, when the description of Tristan's disguise includes a comment on his ability to change his voice. Hoepffner also accounts for Tristan's tactic of delay as a way to test Iseut's love for him.2 Indeed the Queen's identity as "Isolt PAmeruse" is certainly as problematic for Tristan in this version as the identity of "Tristran PAmerus" is for Iseut, and requires a similar, though not identical, process of discovery, as we shall see shortly.3 What Hoepffner and other interpreters of the Folie d'Oxford have so far failed to remark, however, even when they notice Tristan's voice change, is that the change in the ending of the Oxford version is not simply an added twist to the plot, but rather a change in the entire process

Renarration in the Folk Tristan d'Oxford 13 whereby Tristan and Iseut come to recognize each other. In the Folie Tristan de Berne, for example, Tristan does not intentionally trick Iseut once he meets her in private. His effort to reveal himself appears to be blocked primarily by Iseut's own injured feelings, the intense humiliation she continues to feel from Tristan's "foolishness" in front of the court.4 On the contrary, in the Oxford Folie Tristan intentionally continues to speak in a changed voice, even in the privacy of Iseut's chamber. His explicit claims to reveal himself are (at least partially) lies, since implicitly he continues to conceal himself by preserving the impenetrability of his disguise. Once Tristan actually puts on his disguise, the narrator assures him complete control over his identity, since no one in the world looking at him or listening to him speak can recognize the fool as Tristan (w. 217—20). The early descriptions of Tristan's disguise suggest that voice operates as one sign among many others that Tristan manipulates in order to fool Mark and his court or to test Iseut's love. What we learn, however, in the course of the Oxford version, is that voice is not a sign of something else, but the thing itself. As such, voice gives transparent expression to Tristan's being ("Pestre de li," v. 46). Voice allows direct apprehension of the truth, immediate recognition of Tristan's true identity, while signs—whether visual, verbal, or gestural—remain mediators requiring endless interpretation to check their ambiguity, verify their accuracy. Once Tristan disguises (falsifies) his true voice, he cuts off any direct relationship between narration and reality. He separates the sound of a word from its content, creates a tension between signifiant and signifie that gives truth the appearance of lies. This is no simple reversal of true and false: Tristan scrambles these categories such that we along with his immediate public must give up the certainties of simplicity for the greater truths of complexity.

The Mediation of Signs As he tells stories, invented for the occasion or transposed from the past, the fool requires first Mark and his court, then Brengain and the Queen to interpret the signs that multiply throughout his performance. The object of interpretation seems to be clear: see through his lies, arrive at truth, certainty, and transparence. But the obstacles to truth—dissimulation, separation, time, subjectivity, and ambiguity—multiply as rapidly as the signs themselves. As Tristan retells his and Iseut's story in public and then

14

Chapter i

in private, he plays on the difference between public knowledge open to the court and private memories known only to himself, Iseut, and Brengain. Tristan uses renarrations simultaneously to associate and dissociate past and present, according to his listener. For Mark and his barons Tristan's claims to be "Tantris," to associate the fool and Tantris's exploits, lead to ironic dissociation: their unproblematic laughter repeatedly underlines the success of Tristan's disguise.5 For Iseut, the apparent dissociation between present fool and past experience could lead to private association: the absent Tristan might secretly be present in the "fol naif" (v. 411). Of course Iseut does remember her and Tristan's past; she does admit—at least to Brengain—the truth of every word in Tristan's renarrations, but she will not believe the truth of his claim to be "Tristran k'amer vus solt" ("Tristan who used to love you," v. 714), as long as the voice speaking to her remains untrue, disguised—and that despite the presence of those objects from her past, Husdent, the ring, and Tristan himself. In fact, the more truthful his renarrations, the more Tristan repeats their private memories of the past, the less Iseut can believe that the fool and Tristan are one and the same person.6 Howard Bloch has pointed out in Beroul's Tristan how repetition of the narrative, especially from multiple points of view—that is, the very technique given special form in the Folie d'Oxford by Tristan's voice change—introduces subjectivity and interiority into the problem of interpreting reality, replaces "knowable phenomena waiting to be placed within a predetermined historical or metaphysical perspective" with "objects and events whose symbolic meaning is dependent upon a single knower, the context of knowledge, and is liable to exhaustless interpretation."7 A world where signs and reality have become disparate precisely because certain techniques of narration open gaps between them is the very world in which Tristan—and the Oxford poet—place those listeners denied access to the lover's real voice. Bloch's formulation of the problem stresses its negative aspect: the limitations of subjective reality lead to unlimited attempts to interpret it. We need eventually to explore how voice, true voice, restores us to a world of transparent being in the Folie d^Oxford, but first let us explore that fallen world of the sign. Iseut herself steers us to the heart of the problem: "les ensengnes crei" (v. 957), "I believe signs," she tells the fool when he offers to show her the ring that she herself gave Tristan to use as a sign between them. That seems straightforward enough, but when Iseut sees the ring itself, what she believes is not that the fool is Tristan, but rather that Tristan is dead

Renarration in the Folk Tristan d^Oxford 15 because someone else has the ring. Her conclusion is obviously wrong— or is it? One thing is certain: the verb crei ("I believe") is as important here as the ensengnes, since the signs themselves require an interpretation that depends on belief, as much as it does on reason, memory, and perception. Each of these faculties has its own domain, its own tools and procedures. Together they operate hierarchically from the lowest level of perception to the highest level of belief. Memory is a key intermediary, fed from below by the realm of sense experience, nourished from above by the realm of intellect. Reason itself disposes of systems of logic and proof in order to interpret the signs received through the five senses, but its decisions remain subject to the operations of memory and belief. Belief at the crown of the hierarchy remains the ultimate arbiter, since it (like voice) deals directly with being, total and integral. If signs may lead toward the truth, they may equally lead away from it: words become lies, actions deceit, objects tokens of magic, and appearance disguise. Tristan uses signs to send contradictory messages. While the verbal content of his renarrations, Husdent, and the ring seek to trigger memory, the fool's antics, his crazy appearance, and most particularly the sound of his disguised voice send along different, incompatible signals to trouble and confuse (whether for protection in front of Mark's court or for verification of Iseut's faithfulness). But, of course, this is not the first time Tristan has manipulated the ambiguity of signs. His and Iseut's story is built on a series of episodes that play constantly on conflicting messages, conflicting interpretations of signs. If we follow that play through the course of Tristan's renarrations, we can better understand how the present case differs so essentially from the past. A significant number of the episodes Tristan retells in private concern efforts—whether by Mark, spies, or the lovers themselves—to prove or disprove the truth of the couple's adulterous love. Tracks in the snow that allow the seneschal to follow Tristan's path to the Queen's chamber, sheets that Mark finds bloodied in both Tristan's bed and the Queen's, these are all physical traces of events that seem to prove the couple's guilt. Iseut and Tristan themselves have used such signs to forecast events: for example, the whittled sticks ("cospels," v. 784) that signal to Iseut her lover's arrival under the pine. Unfortunately for them, these private signs—"enseignes entre nus" (v. 785)—can also be discovered and correctly interpreted by outsiders. Such discoveries thus generate further manipulations of signs as the lovers perform elaborate charades for Mark hidden in the pine tree or later for the entire court at the scene of the trial. In Tristan's recital, the

i6

Chapter i

trial episode follows immediately.8 The proof for Iseut's innocence "guarie al jugement / Del serement e de la loi" ("guaranteed at the judgment of the compurgation oath," w. 832-33) is elaborately constructed and staged by the lovers. Tristan's summary highlights the essential elements: his disguise, Iseut's verbal instructions both before and during the event, and the action invented to make Iseut's oath perfectly (if partially) true: "Ben sai quai me dei'stes dune: K'od vus me laissasse chaeir. Ysolt amie, n'est go vair? Suef a la terre chai'stes E voz quissettes m'aiivristes, E m'i laissai chair dedenz, E go virent tutes les genz." (w. 824-30) ("I know what you said to me then, that I should let myself fall. Beloved Iseut, isn't this true? You fell gently to the ground and you opened your thighs and I let myself fall between them, and everyone saw this.") This real event, observed by everyone present, conveniently masks another real event from the past—which is thus disproved and redefined as untrue by Iseut's oath. The latest event renarrates the previous one: it gives a new, innocent referent for the apparently restrictive description in Iseut's oath (not repeated here by Tristan). It is true that Iseut has opened her thighs to only two men, but of course we know that the same words may mean different things in different contexts, that in fact they describe three events, not just the two publicly witnessed by the court. Iseut, mastermind of this particular trick, takes advantage of that ambiguity. She allows the unlawfulness of one (secret) action to be covered over by the exuberant innocence of another, whose verbal sign, under the right circumstances and in the mouth of a clever woman, can be identical.9 The ambiguity of signs that the lovers take advantage of here is not simply the product of their own machinations. It plays an important role in their lives and in their story even when no one seems to be actively interfering in the process of signification. We are reminded of this quite explicitly by another event repeated in Tristan's renarration. When Mark finds the lovers asleep in the cave, they are separated by a sword. Mark interprets the accidental configuration of their bodies—providential, in

Renarration in the Folie Tristan d'Oxford 17 Tristan's view (v. 881)—as a sign of their innocent relationship (v. 89194). It is this inherent ambiguity of signs that helps keep the story of Tristan and Iseut moving from episode to episode, as Mark fails to resolve his uncertainty about their love once and for all. If we return now to the present action of the Folly episode, what immediately strikes us is the change in the unity of the lovers. In the past they were always in league against everyone else, making sure that only they knew the truth of their love. Now Tristan has joined Mark in the realm of uncertainty and for the first time the lovers are separated not just physically, but emotionally. Where previously their tricks worked only to defend them from the outside world (while the lovers themselves remained "in the know"), now Tristan is tricking Iseut herself. Each member of the couple now occupies a different area of the unknown. By absolutely concealing his own identity, the invisibly present Tristan (therefore susbstantially absent as far as Iseut is concerned) tries to overhear her "true confessions," tries, in fact, to do exactly what Mark and his spies were always trying to do, prove or disprove the truth of Iseut's love.10 The difficulty of interpreting signs, heretofore never a problem for the lovers who could always intuitively follow each other's clues, has now enveloped Tristan and Iseut in its inexhaustibility, requiring each of the lovers to decipher the other's signs.11

Iseufs Process of Interpretation When Iseut retires to her private chamber, she and Brengain discuss in some detail how they should interpret the fool's performance and his claims to be Tristan. Both of them apply the tools of logic; they reason syllogistically, apparently from the same premises, but each comes up with different conclusions: 1. The fool has completely and accurately reported Iseut's past life. 2. Some intimate details of that past are known to only three people: Iseut, Tristan, and Brengain. 3a. Therefore, according to Iseut, the fool is an enchanter. or 3b. Therefore, according to Brengain, the fool is Tristan. Logically speaking, of course, Brengain's conclusion appears simpler, more convincing. But we should not dismiss Iseut's argument out of hand.

i8

Chapter i

We have seen her earlier in Tristan's renarration proceeding quite methodically to prove her suspicion that the man in the bath was none other than the murderer of her uncle, the Morholt. Iseut's problem is not how to reason logically, but rather how to coordinate all the signs requiring interpretation.12 In reporting Brengain's conclusion Iseut clarifies her own starting point, apparently overlooked or unheard by Brengain. The fool cannot be Tristan himself, " . . . kar cist est laiz, Hidus e mult cunterfaiz; E Tristran est tant aliniez, Bels horn, ben fait, mult ensenez: Ne serroit truvez en nul pai's Nul chevalier de greniur pris." (w. 577-82) (". . . because this one is ugly, hideous and misshapen, and Tristan is so well shaped, a handsome man, well made, well mannered: in no country could you find a more worthy knight.") Reason does not operate alone in the process of recognition. Here Iseut is responding—necessarily—to all the signs of Tristan's disguise—visual, aural, and gestural as well as verbal. In order to evaluate the fool's identity, his physical evidence must be compared, along with the content of his renarrations, to her memory of Tristan. Memory measures present perceptions against the image of Tristan that Iseut has built up in her mind over the years, memories of appearance, behavior, and shared experiences.13 The fool has repeatedly called on Iseut to look back to those memories: "Membrer vus dait" ("you must remember") and its variants operate as introduction and refrain throughout his renarrations. But the fool who frightened and humiliated her in Mark's court simply cannot fit into those memories, remembered in this context with intense pain: "Tur go ne crerai je uwan / K'igo sait mun ami Tristran'" ("Tor this reason I shall never believe that this is my beloved Tristan,'" w. 583-84). This conclusion itself reflects the hierarchy of faculties mentioned earlier, in which belief orders the results of intellect, memory, and sensory experience. Belief grasps the totality of a person's being. But Tristan's disguise disintegrates his totality; the fragments of Tristan and the fool cannot make a credible whole for Iseut, who continues to reason syllogistically:

Renarration in the Folie Tristan d'Oxford 19 1. She sees: the fool is ugly and villainous. 2. She remembers: Tristan is handsome, courtly, chivalric. 3. She believes: the fool is not Tristan. The fool as magician represents Iseut's effort to reconcile the contradiction between her overriding belief and the logic of Brengain's conclusion. Magic may account for the discrepancy between what the fool knows and who he is. Iseut's interpretation thus underlines the relative weight and priority given to competing signs. Her starting point remains a fundamental article of belief, reiterated, shaken, but never abandoned as long as Tristan continues to disguise his voice: the fool is not Tristan, because he does not resemble the Tristan she remembers. In the confusion of signs that argue for and against equating the fool and Tristan, no sign carries the force of conviction, as long as Tristan conceals the voice of his true self. Iseut thinks it would be crazy ("Folie serrait e engan," v. 845) to recognize him as Tristan when she believes him to be someone else ("ele vait e pense e creit / N'est pas Tristran, mais autre esteit," w. 845-46). The Husdent episode is built on the same conflicting signs and interpretations, but adds an important element of concrete presence lacking in Tristan's renarrations. Part of their past experience, the dog can actually bring that past into the present, when Brengain brings him into the presence of Iseut and the fool. What we see happening from Iseut's point of view, however, is an increase in her confusion and distress. Troubled thoughts, embarrassed and blushing silence, show Iseut refusing to let present signs overrule her belief in the past. When the fool tries again to link past and present by showing her the ring she herself gave Tristan as a sign of their mutual love and trust, Iseut thinks she is going crazy ("quidat desver," v. 962). This "craziness" is a sign that Iseut still refuses to abandon her first belief, a sign that her past image of "Tristran PAmerus" still cannot accommodate the "fol naif"—and a sign for Tristan himself that she has not abandoned her love for him despite the strains of time and separation.

Tristan's Process of Interpretation and Recognition Up to this point in the narrative Tristan has been intentionally multiplying the conflicting signs, true and false, of his own identity, ostensibly to convince Iseut about the truth of his claim to be himself, more covertly to

20

Chapter i

prove the truth or falsity of Iseut's love for him. Now he has reached the critical turning point in this episode. We along with Tristan have followed Iseut's attempts to interpret those signs and have discovered that they cannot lead her to the truth. We must now backtrack somewhat to see how, throughout this process, Tristan has been interpreting the involuntary signs of Iseut's love. In one of his opening monologues Tristan brags that he knows quite a bit about folie and sen: his decision to try out a fool's disguise is immediately conceived as a test of wisdom and folly. Tristan's own appearance of folly will be "grant veisdie" (v. 182), while the apparent good sense of those who take him for a fool will in fact reveal their own foolishness (w. 180—88). This is precisely what happens when the fool arrives at Mark's court and Tristan (as well as the narrator) obviously takes great pleasure in this aspect of his game. It is mainly in retrospect that we discover the second, more covert (but at the same time more important) aspect of Tristan's foolish game—that is, the test of Iseut's love, one of the main themes of the Oxford Folly. Tristan's uncertainty about her love appears only briefly in one of his early monologues: "Ne sai s'a mai parleriez" ("I don't know if you would speak to me," v. 174), he wonders when he arrives at Tintagel.14 Later when Brengain leads Tristan into Iseut's chamber, there is a moment of confusion for the two lovers when the fool tries to kiss Iseut and sees that she pulls back without quite knowing what to do (w. 677-88). At this moment the narrator describes both Tristan and Iseut with the same adjective: huntus(e) (ashamed, w. 682, 686). While the Queen is frequently observed in such distress throughout this confrontation, the only other time we see Tristan thus troubled is in the opening monologues of the Folie. This confusion is the prelude to Tristan's first complaint about Iseut, in which he accuses her of disdaining his love and compares her love to a fountain whose abundance has dried up (w. 690708). Can we speculate that Tristan experiences a moment of real doubt here, that this complaint is not simply a tactic in his fool's strategy (as his second complaint, w. 851—56, seems much more clearly to be)? We cannot, of course, resolve such a question: true and false do not always divide into the two neatly separate categories indicated in Tristan's monologues. Indeed, when Tristan attributes to "feintise" (w. 851-56, 934-38) Iseut's failure to acknowledge him as her lover, his choice of word is wonderfully ambivalent, since it links the problem of true or false to the question of dissimulation.15 As we have already pointed out, Tristan's problem is how to establish certainty, given the complexity and ambiva-

Renarration in the Folie Tristan d'Oxford 21 lence of signs. His knowledge of the past tells him that Iseut is capable of manipulating signs, even if he himself was not at that time the target of her deceit. Since Iseut is not openly disguised, since she is not concealing her true voice, the process of recognition cannot be exactly symmetrical for the two lovers. In order to recognize the truth of Iseut's love, Tristan focuses his attention on all the passive signs of Iseut's emotions. He registers with special care the involuntary movements of her body: "Ben aperceit k'ele ad irrur, / Kar el vis mue la culur" ("He certainly notices that she's disturbed, for her face changes color," w. 325—26—cf. w. 382, 673, 920, 939). This expression "muer culur" has been used before to describe part of Tristan's disguise, when he darkens his face with an herb (v. 216). Muer is the very verb associated earlier and later with Tristan's voice change (w. 212, 975). It is thus a verb that links the lovers, but also places them on opposing sides of intentionality: Tristan's changes are always voluntary, self-motivated, just as Iseut's changes are always involuntary reactions to her lover's tricks. The Queen remains essentially passive in this episode, registering by her mounting distress the effects of the fool's manipulations, allowing the surface changes of mood and expression to signal (involuntarily) the unchanging character of her love for Tristan. Gradually in the course of these two scenes Iseut has been driven to experience the very madness of love ("'il m'ad fet dever,'" v. 595) that drove the mad Tristan ("afolez," v. 175) to feign folly ("Feindre mei fol, faire folie," v. 181) and appear to the court a "fol naif" (v. 411).16 When Iseut recognizes Tristan's ring, she reaches the extreme point of folie referred to earlier. Her love-maddened sorrow now breaks out uncontrollably in gesture and lamentation: Si s'escreve dune a plurer; Ses poinz detort, quidat desver: "Lasse!" fait ele, "Mar nasqui! En fin ai perdu mun ami, Kar c^o sai je ben, s'il vif fust, K'autre hume cest anel n'eust. Mais or sai jo ben k'il est mort. Lasse! ja meis n'avrai confort." (w. 961-68) (Then she breaks out into crying, wrings her hands, think's she's going crazy. "Alas!" she says, "Woe that I was ever born! I have finally lost my beloved, for I know well that if he were alive, no other man

22

Chapter i would have this ring. But now I well know that he's dead. Alas! I'll never be comforted.")

Paradoxically, the (false) certainty that she twice expresses here—"sai je ben," w. 965, 967—in fact conveys to Tristan the certain truth about her love for him. At this point, then, Tristan's test—originally conceived to distinguish those who merely sccmfou from those who really are/ow (tfamour)—has been completely satisfied. Now he recognizes the truth of Iseut's love because it recalls (as it repeats) his own true feeling of love-crazy madness, the starting point for the entire Folly episode. Appropriately, Tristan pronounces Iseut true when she corresponds to the image he has of himself as a lover (that is, when she exhibits all the signs of love that he has personally experienced) and when her present image corresponds exactly to the image of her that he had in the past. The fusion of past and present figures explicitly in Tristan's descriptions of the Queen: "'Mult suliez estre enterine'" ("'You used to be so completely whole/true,'" v. 942). The complaint that a gap exists between the Iseut of Tristan's renarrations and the present Iseut who will not believe the fool becomes in the moment of recognition a declaration of praise: "'Bele estes e enterine'" ("'You are beautiful and true,'" v. 972). Tristan's choice of adjective here is particularly significant for the Folie d^Oxford.17 His visual observations of the Queen culminate when he hears the sound and the message of her lamentations. Tristan thus bases his conviction on the integrity of Iseut's response, on the wholeness of her being, the concord of every aspect of her self. Iseut, finally, is not a sign of something else; she is herself, intact, and that self is "Isolt 1'Ameruse." In a sense, Tristan is able to recognize Iseut by returning to his earlier knowledge of their shared love, to a time in the past when the couple had no doubts about the integrity of their relationship (e.g., w. 851-53, 857-58).18 Memory working through Tristan's renarration of the past, as well as with his perception of the present, restores Tristan's belief in Iseut and her love: "Mais quant Tristran plurer la vait, / Pite I'em prist, e go fu droit" ("But when Tristan saw her cry, he rightly took pity on her," w. 969-70). We realize now (or are reminded) that Iseut's refusal to recognize the fool as Tristan was well founded: "Tristran PAmerus" could not be present, not totally present, as long as he was not present in his own totality—that is, as long as the fool was speaking with someone else's voice. "Tristran PAmerus" implies an integrity of being and love (like

Renarration in the Folie Tristan d'Oxford 23 Iseut's adjective enterine] that falls into fragments as long as the fool claims to be Tristan, claims to speak truly in a false voice. "Des or ne m'en voil mes cuvrir, Cunuistre me frai e oir." Sa voiz muat, parlat a dreit. (w. 974—76) ("From now on I do not want to conceal myself. I'll let myself be recognized and heard." He changed his voice and spoke straight.) Once Tristan does speak in his own voice, he puts an end to the false route of inexhaustible interpretation; he stops producing ambiguous signs and speaks openly, "a dreit." Once Tristan restores his fragmented identity to wholeness, the effect is immediate: Iseut perceives at once that he is himself (v. 976). Tristan's own voice offers the final, in fact, the only convincing proof that he really is "Tristran 1'Amerus." Why has voice been thus privileged in the Oxford version—and only there, at least within the Tristan tradition as we have it!1 I can only begin to answer that question by exploring within the context of this Folie some of the qualities associated with sound and the spoken word. Perhaps it should not surprise us that, in a story largely composed of Tristan's retelling of a story, the spoken word should play such a preeminent role. After all, Mark's court appreciates with special gusto the verbal antics of the fool's performance (w. 313-14). Throughout the Folie d'Oxford—as in other versions of the Tristan story—the verbs of telling and listening play an important role in the narrative fabric. The special genius of the Oxford poet, however, privileges verbal clues at their most sensible level; it shifts the focus from the content of words to the sound of the voice that produces them. Tristan's renarration reminds us, as well as Iseut, that sound (especially musical sound) has already played a key role in their relationship. Remember that Tantris first comes to Iseut's attention because of his talents as a harper (w. 353-56). The transition back to straightforward truth in Tristan's upside-down speech is a reference to his musical talents (w. 521-24). And those very same talents win the Queen a second time for her husband Mark, when Tristan defeats Gamarien in the contest of rote against harp (w. 763-76). With the story of Husdent and how Tristan has trained him to hunt silently, we are reminded that sound gives away presence (w. 873-74). Iseut's radically different reactions to Tristan's two voices emphasizes

24

Chapter i

how sound plays an important role in our perception and understanding of multi-faceted human reality. Both within themselves and in the world around them, men and women must coordinate an abundance of factors in order to interpret signs and grasp being. I have already described the hierarchy of faculties—belief, intellect, memory, and perception—that orders the inner part of that process. The phenonmena of the outside world, too, follow certain patterns—whether in the way its constituent elements combine with each other (that is, form a particular context) or in the way our five senses interact with the appropriate sensory experiences. Voice sets words, visual images, and gestures in a particular context. We have seen how the context created by Tristan's false voice makes it impossible for Iseut to line up her present perception of the fool and the memory of Tristan she retains and believes. Voice is not just one sign among others, interchangeable with them. Voice can express directly and transparently the truth of inner being, while signs always mediate, call for interpretation. Compare the difference between visual perception, which operates at a distance, sharpens line, and distinguishes surfaces, and the way sound surrounds us, tells us "CI am in his presence . . . not in front of his presence.' Being in is what we experience in a world of sound." Walter Ong reminds us how intensely the spoken word conveys a sense of presence, such as we can obtain from no other sensory experience.19 As verbal signs cut off from their true sound, Tristan's renarrations fail to support the fool's claim to be "Tristran PAmerus"; interference from competing signs leads Iseut away from, instead of toward, the truth. Once Tristan speaks in his own voice, however, what he conveys to Iseut is no longer a reflecting surface, a sign of something else, but the truth of his interior self. The presence and presentness of his voice forge the vital link between past and present without which his renarrations, however truthful in content, cannot reunite the lovers.20 The Oxford poet uses the same word—with two different, but closely related meanings—to describe both Tristan's return to his real voice ("parlat a dreif) and the narrator's own guarantee that Tristan's pity for Iseut's tears was right ("e c;o fu droit"), the moment of recognition was appropriate.21 If voice has the power to convey certainty, as I have asserted above, then the poet's affirmation of rightness and truth at this point conveys simultaneously both the renewed presence of Tristan's voice and the author's role in representing that voice. We in the public are reminded indirectly that the Folie Tristan d'Oxford exists as a written text, and in a written text we lose precisely that element that conceals and then reveals

Renarration in the Folie Tristan d'Oxford 25 Tristan's identity—that is, the sound of his voice. Just as Tristan's disguise breaks apart the identity of his words by separating and opposing form and content, so the poet's written text goes one step further in the process of dissociation, breaking the signifier itself into written and oral signs, visual presence and oral absence. It is quite interesting to note in this respect that the narrator tells us about Tristan's power to change his voice while enumerating all the visual aspects of the disguise. We see Tristan put on the fisherman's clothes, cut his hair, dye his face, take a stick (w. 197-222)—and in the middle of all this we learn: "Tristran sout ben muer sa voiz" ("Tristan knew well how to change his voice," v. 212). Why do we learn this fact now when Tristan is all alone and obviously not yet speaking:1 The displacement may already signal the potential importance of that singular audible element. Tristan's intention to use his ability does appear indirectly when the narrator emphasizes the complete visual and aural impenetrability of the disguise (v. 220). When Tristan actually arrives at court, his voice change is not mentioned, will not be mentioned at all until he finally returns to his own speaking voice in the final scene (v. 975). Because of this significant omission, at least at the explicit level of the narrator's own words, we participate during the main body of the text in the characters' own concentration on the visual, verbal, and gestural aspects of the fool's performance. We, along with Mark and his barons, Iseut and Brengain, may become victims of Tristan's disguise, unless we remember that key oral element surreptitiously—silently—operating throughout the fool's performance, as effective in Iseut's chamber as in Mark's court.

The Sound of Voice in a Written Text Once we have finished reading the Folie from the viewpoint of a suspenseful first reading and have begun to explore the retrospective pleasure of the text, we may begin to suspect that the author has tried to trick us with the same artful cleverness that Tristan uses on his public, especially on the nearest and dearest member of that public.22 In order to verify that suspicion we need to consider what the Oxford poet might have had at his disposal to represent Tristan's disguised voice and how he chose to do so. Though visual disguises are frequent elements in contemporary plots, disguises that involve voice seem to be rather rare.23 The Renart stories, often comic variations of the same trickster materials that underlie the more serious treatment of Tristan narrators, include an episode in which Renart,

26

Chapter i

accidentally dyed yellow, decides to complete his disguise by changing his way of speaking: "il changera son langage."24 He does so by speaking French with an English accent—and thus gives the author of the written text ample opportunity to represent graphically his character's voice change.25 The Folie author, by contrast, cannot represent Tristan's voice change in the graphic inscription of the text. In fact, we have no knowledge of how Tristan changed his voice, only that he did so—or rather only that he could do so and stopped doing so at a certain point in the narrative. What the Oxford poet has given us instead of the sound change in Tristan's voice is a change in the type of stories Tristan tells—that is, something that can be recorded narratively in his text. When Tristan arrives at Mark's court he begins his performance with the doorman by inventing a foolish story: he has just come from a wedding celebration; the abbot of Mont St. Michel has married a fat abbess at "Bel Encumbre" (w. 229-40). From this highpoint of wild invention, Tristan descends through various mixtures of truth and falsehood (whether lies, dreams, deceptions, or inventions) until he finally returns to absolute and unalloyed truth. But that return to truth is two-staged, just as the Folly episode itself is divided into the public show in front of Mark's court and the private encounter in Iseut's chamber. In the first part, Tristan obviously mixes preposterous fabrication and covertly open truth-telling. His game is only occasionally obvious to his own public, but (in this part, at least) always obvious to the author's public. When Tristan claims to be the son of a whale nursed by a tiger or when he enumerates his hunting skills in a topsy-turvy fashion, the court can appreciate and enjoy his verbal skill. When he claims to be Tantris and thus associates himself with Tristan and Iseut's publicly known, true story, they laugh wholeheartedly at his fool's play, a reaction that Tristan reinforces by the use of his most obvious and active attribute as fool, his stick (w. 222, 238, 260, 268, 516, 528, 611). Those of us in the know, on the other hand, can see the real thrust of Tristan's true stories; we can even appreciate some of the hidden glimpses of truth that shine through his foolish inventions.26 Iseut, who occupies a position between ours and the court's as far as the distribution of knowledge is concerned, aggressively labels the truth as lies every time she knows the fool is telling her true story.27 She does so, that is, as long as the court is present; in private her reactions are more complex, as they are more interiorized: except for the momentary excitement when she anticipates showing Husdent to the fool and seeing the

Renarration in the Folie Tristan d^Oxford 27 ring she gave Tristan (w. 898-902, 957-58), the Queen primarily experiences and sometimes expresses out loud the growing distress and uncertainty already described above.28 We in the public can surely understand her defensive aggression and disbelief in the court scene, but once the fool enters her private chamber it may strike us (especially on the first time through the story) as decidedly bizarre that she cannot recognize Tristan; he so obviously tells the truth.29 The narrator implicitly reinforces just such a reaction by eliminating Tristan's foolish stories just at the moment when the fool leaves behind his stick, that is, as soon as he sees Brengain (w. 610—n). In this second stage of the return to truth, all Tristan's speeches are straightforward retellings of their true past (except for the mournful complaints of the suffering lover).30 The most obvious response to this narrative maneuver is to assume that Tristan is being more truthful than ever. We are indeed all the more inclined to believe in such increased truthfulness to the extent that we have been privileged by the narrative point of view to overhear Tristan's private monologues, to watch him put on his disguise and perform the part of the fool with such consummate skill and control. We know all along that it really is Tristan speaking. On the other hand, thanks to the scene between Iseut and Brengain, and to the continuing descriptions of Iseut's state of mind, we are also privy to the Queen's confidential thoughts: we know how sincerely she hesitates in disbelief, even if we cannot immediately understand what motivates her. While these insights contribute to our certainty that the final recognitions are as true as Iseut and Tristan finally believe them to be, they place us readers in an ambiguous position during most of the scene in Iseut's chamber: we know simultaneously more and less than the characters; we never doubt Tristan's identity because we never hear his voice change. On the contrary, narrative strategies of commission and omission seem designed to lull us into forgetting that voice change, so that we can be as surprised as Iseut within the story when the narrator finally announces Tristan's return to his own speaking voice. The poet thus understood impresses us with his cleverness: he is both a serious and playful trickster, fully worthy to represent as artfully foolish a protagonist as Tristan. Of course, my description of the relationship between author and reader so far ignores the specificity of the medieval context: written texts of the twelfth century were read aloud to a listening public. In an oral performance of the text, the voice of the narrator remains present, presents the voices of his characters. An oral reading of the Folie d^Oxford could

28

Chapter i

dramatize Tristan's game by supplying the sound of his voice change (and thus supplement the silence of the written text).31 Did the author count on such a performance to remind his audience of Tristan's impenetrable disguise or did he still expect to catch us off guard with the narrative maneuvers recorded in the movement of his text? However we answer that question, the situation in this Folly clearly leads us to explore the relation between oral story and written text. What are we actually getting when we hold the Folie d'Oxford as book or manuscript in our hands ? Here is a written text whose words record the traces of a now absent author, an author who has remained anonymous in the single extant manuscript now found at Oxford. His words represent the actions and, especially, report the spoken words of Tristan, Iseut, and others: two thirds of the text are in direct discourse—that is, 680 verses out of 998.32 Tristan's words, in turn, retell earlier words spoken, as well as events experienced by them. We can assume that the earlier episodes are present in the memory of the Folly author because they were already circulating in other versions of the Tristan legend, some oral, some written (and most lost to modern readers).33 We find ourselves constantly crossing over from the story of the characters to how that story is told, both within the plot of the Folly episode itself and in the narrator's realization of that episode. Mises en abyme multiply, as we discern previous episodes retold by the fool as miniature versions of the Folly episode, the Folly itself as a mise en abyme of the entire Tristan legend, and Tristan's retellings as a mirror of the author's own retelling.34 Just as Tristan uses a disguise along with his renarrations, the author uses anonymity to cover his presence as writing author in favor of the anonymous voice of the storyteller. Both hero and poet present themselves as storytellers whose individual voices are as important to the truth of their stories as the words chosen to represent them.35 In view of the oral and written traditions reflected in different extant versions of the Tristan legend, we can see the Oxford poet distinguishing himself by a desire for transparence, for the same sort of intimate, direct spoken contact with his public that Tristan finally establishes with Iseut at the end of the episode. Nowhere does the Oxford Folie refer to itself as a written text; it does not designate itself as conte or te', as we see in the case of Marie de France's Chiewefoil, her single, independent Tristan episode. The Folie d'Oxford has neither prologue nor epilogue to call attention to itself as a literary object or to name its author (though, of course, contemporary practice endorses such procedures).36 Nor does the Oxford Folly refer explicitly to any other versions of the Tristan story—unlike Beroul

Renarration in the Folie Tristan d^Oxford 29 and Thomas, who have disparaging remarks to make about their competitors.37 On the other hand, the order in which Tristan retells his adventures is so close to the order in which they appear in Thomas's particular version that Joseph Bedier has pictured the Oxford poet composing his own work as he turns over, one by one, the pages of a Thomas manuscript.38 Though I am emphasizing here the difference between the Folie d^Oxford and other versions, that difference must also be understood in the context of a shared tradition. From that point of view what is striking in the Oxford version's undeniable reference to Thomas's written version is the way it remains implicit—written into the order of renarrations, but unstated by the voice of the text. The Oxford poet thus indirectly acknowledges written tradition, but presents his own written version as much as possible as a story that tells itself. Why might the poet seek to obscure his written text in favor of the spoken word? While the private intentions of a medieval author may seem particularly unreachable (even if modern criticism had not already warned us about the "intentional fallacy"39), we can try to explore the cultural context in which such decisions are made. While intensely concerned with written tradition—based on the model of the Holy Scriptures, as well as that of the classical auctores—medieval writers and thinkers still operate in a culture of the spoken word; manuscript culture remains organized to a large extent according to the values and operations of an oral-aural society.40 If for the moment we simplify the complexities of this cultural context, two paradoxical systems appear in competition. A conservative thrust looks back to the greater reality and presence of the spoken word (cf. the voice of the singer whose epic formulae and traditional stories guarantee society's memory of the past); a modern thrust identifies written words with permanence and thus values the written text as a vital means to reinforce society's memory. Written words are more permanent, but less real; spoken words are more elusive, but more real.41 There is a price to pay for written permanence, at one remove from actuality. As Eugene Vance points out, the Christian Middle Ages were alive to the dangers of textuality, to the death threat posed by the opaque letter. The spirit that breathes life back into writing remains the spoken word: voice keeps a text alive, just as God's word creates and maintains the world, provides in the Bible a model of "ultimate presence, a voice spoken and heard in the duration of all time."42 It is against such a cultural background that we need to situate the Folie Tristan d^Oxford and its problematic existence in writing: how can the

3O

Chapter i

opaque, written text convey the transparent reality of the voice? The author of the Folie deals with paradox paradoxically. The values endorsed openly by the text remain those of the spoken word. We have seen in the plot of the Folie how Tristan's real voice is the spirit, the very corporeal spirit, that removes the block between past truth and present perception, and thus guarantees memory and continuity for the lovers. Yet the modernity of the Oxford poet is such that he has chosen to represent and recreate voice exclusively by means of the written text. He has not inserted himself as a character telling Tristan and Iseut's story: his narrator rarely speaks to us directly in his own "I"; any obvious narratorial voice (such as we can observe in Beroul, for example) is subordinated to the characters' own voices. Instead the Oxford author has given us a poem that tells its story, even as it conceals its identity as written text offered by a narrator. Its implicit allusion to another, specifically a written version of the Tristan legend, reminds us indirectly of just what has been suppressed in favor of the "transparent word." It may remind us as well of a problem that greatly concerned Thomas: the truth of any given version of the story (Douce, w. 835-84). Within the scope of medieval invention, the lovers' truth may be embodied in a number of different versions, each a product of human invention, told and retold from a different point of view, in a different voice. Each new version is true insofar as it recalls and breathes life into the whole body of the Tristan tradition. Even if we accept this plurality of true versions, we can still ask how a particular text verifies the truth of its own voice. For the Oxford poet it is the transparence of his written text that gives it the reality and presence of the spoken word. The author of the Folie d}Oxford revitalizes the Tristan tradition, the mythic past of the lovers' story, by incorporating it through his own distinctive voice: the concrete, written text traces that voice, lends its own presence to embody the absent author, the absent myth.43 If someone were reading the Oxford Folly out loud, the speaking voice would give a more lively presence to the narrator's, but such an oral performance is no longer absolutely necessary in this case, since the author has recorded the peculiar features of his voice in the written text itself.

The Voice of the Oxford Poet in Intertextual Dialogue How do we hear the author's voice:1 That question must be understood and answered here within the context of medieval literary practice. To start

Renarration in the Folie Tristan d^Oxford 31 at the broadest level of generality, we can say that the Oxford poet establishes his individuality by repeating and reinventing the already known, by remaining both faithful to and innovative within the bounds of received tradition. The intertextuality of the Tristan stories is such that, if we attempt to characterize a given text by picking out its salient features, we are inevitably focusing on elements present to a certain degree in the rest of the tradition: there is always a zone of overlapping features pulling the disparate individual texts back into the whole Tristan legend.44 Thus Ernest Hoepffner compares Thomas's London (Douce, w. 1379—91) to the Folios Tintagel (w. 99-140) and sees in such vivid descriptions one of the characteristics linking the two authors.45 Yet it seems to me that the narrative and rhetorical context of the two texts is such that these descriptions play a much more prominent role in the Oxford Folly than they do in Thomas's romance. The quick accumulation of eccentric details, many unnecessary for the plot—whether Tristan's dialogue with the mariners (w. 76-84), the description of his disguise (w. 189-224), or his encounter with the doorman (w. 225-47)—stand out within the fast-paced movement of the opening narrative, adding their own effect to the theme of presentness, alive presence, developed later with the lovers' recognitions. This effect is even more striking if we compare equivalent passages in the Berne Folly—for example, Tristan's welcome in Mark's country (Fb, w. 135-36; Fo, w. 248-58), the doorman (Fb, v. 149; Fo, w. 225-47), the fantasy house (Fb^ w. 164—67; Fo, w. 301—10). Comparisons between Tristan texts necessarily alternate between pointing out nuances of difference and charting the equally important resemblances. The elaborate character of the renarrations in the Oxford Folie emphatically calls attention to its links with other versions, especially, but not only, with Thomas's particular variation. When the author of the Folie d^Oxford invents a new "detail"— Tristan's voice change—he introduces the unexpected into a well-known tradition and thus expresses more directly the individuality of his own voice. The resulting change in thematic focus develops less directly, but more forcefully that individual voice. Compare Thomas's emphasis on human changeability, the suffering and errance of the lovers, with the Oxford Folly's optimistic reunion, the lovers' recovery of transparent being. Thomas's narrator intervenes frequently with his personal "I," comments at length on Tristan's wavering feelings for Iseut la blonde and Iseut aux Blanches Mains (e.g., Sneyd,1 w. 182-304) or on the relative sufferings of the four unhappy lovers (Turin, w. 71—151). Like Tristan himself in the

32

Chapter i

Hall of Statues, the narrator turns around and around the lovers' painful ermnce-, his manner recalls the movement of a persistent tongue that cannot stop touching a canker sore to see if the pain is still there. Significantly, the Oxford author omits key episodes from Thomas's version that support the emphasis on changeability: the second Iseut (mentioned twice in the Folie de Berne, w. 49—50 and 241), Tristan's statues, and the mistaken sails that lead to death in the end. Unlike Marie's Chievrefoil (w. 9-10), the Oxford Folly contains no explicit allusions to the lovers' fate after their reunion here and now. The Folly episode's success reconnects past and present, conceals for a moment the future deaths and separations.46 Of course, dying does appear verbally in the Folie ^Oxford—the metaphors of death that describe Tristan's love suffering (w. 5-24,169-72), Tristan's thoughts about the danger posed by Mark's hatred (w. 163-67), the Queen's lament upon seeing Tristan's ring (w. 964-66)—but these evocations turn out to be momentary fears and delusions in light of the lovers' final recognitions. In a sense, the Oxford poet seems to take Thomas's theme of human changeability as the starting point for Tristan's decision to test Iseut's love. We readers of the Folie no longer have textually before us the kind of reasoning that leads Tristan to marry the second Iseut; instead we have a text that resolves all the lovers' doubts and restores their integrity as a couple. Doubting Thomas has been determinedly recalled and reinvented by a new/old version of the Tristan legend.47 What about the Oxford Folios relation to the Berne version? How can we pinpoint the way these two texts tell the same story, yet appear so unlike each other?48 Consider first the nature of Tristan's disguise and its relation to his renarrations. If we juxtapose the Folie d^Oxford and the Folie de Berne, an intriguing chasse-croise appears in their treatment of transparence and opacity. To put it most succinctly, where Tristan's renarrations are more disordered and less intelligible, Tristan's disguise itself is more easily penetrated. Mark's barons murmur suspiciously about the fool's credibility (w. 248-49) and Brengain distinguishes Tristan's own shapely body beneath the fool's clothes, even before he identifies himself directly (w. 290-95).49 On the other hand, where Tristan's retellings are perfectly clear and detailed, arranged in chronological order, the fool's disguise is completely impenetrable. Indeed, that impenetrability vouchsafes the apparent clarity of Tristan's renarrations; his change of voice guarantees their failure to reveal him. If we consider the two Folies in relation to the larger Tristan tradition the same criss-cross pattern appears. In the Berne version, the incoherence of Tristan's story means that we in the audience, as well

Renarration in the Folie Tristan d'Oxford 33 as the participants, must know the tradition well in order to reconstruct or situate Tristan and Iseut's story. To that extent, the Folie de Berne is closely tied to other Tristan stories. Yet this incoherence is functionalized in Tristan's characterization as the effect of his love, whose pain has maddened him (w. 121-23). As an independent poetic text, the Berne Folly has its own "psychological" coherence, while the independence of the Folie d'Oxford results from its more artfully-constructed coherence. The order and clarity of its summaries allow the audience to follow Tristan's recital without having to know other variations of the story. Of course, such an uninformed audience may be unlikely in the Middle Ages, given the popularity of Tristan and Iseut. More importantly, a well-informed audience would perceive how it depends on and implicitly alludes to Thomas's romance, another quite individual retelling of the myth. That specific link leads Maurice Delbouille to describe the Folie d'Oxford as a mere digest of Thomas, minus the psychological reflexions.50 But if we see Tristan's renarrations as themselves the major focus of the Oxford Folly—a reaction easily understood, given the number of verses they occupy in the whole text—we have, I think, mistaken the real problem posed by the Oxford poet. We are well advised to remember that two poets can (and do constantly in the Middle Ages) tell the same story for very different reasons, with very different effects. Using one text as a necessary standard by which to judge the other may block our understanding of the differences. If we follow through with the comparison of the two Folies initiated above, we can see that the problem posed by the Berne Folly is that of recollection and reconstruction: we readers are witnesses and participants who must use our powers of memory, just as Tristan and Iseut themselves, to reassemble in the present the scattered fragments of the past. It is completely appropriate for the Berne Folly to base its recognition and closure on the presentation of Tristan's ring: once Tristan and Iseut have again exchanged dog and ring, they have recalled and brought together all the elements present at that moment in the past when they parted in the Morrois forest.51 Memory has completed its work in the reunion of the ending. In the Oxford Folly, however, memory is not sufficient for recognition to occur. The task required is not limited to remembering: we must also interpret. We must move from the invmisemblnnce of Tristan's performance to the truth that expresses and restores the lovers' identity, from the opaque signs of Tristan's disguise to the transparent voice of his true being. The author of the Oxford Folie first leads us and his characters into a

34

Chapter i

dense romance world where everything is a sign to be interpreted again and again, as we try to grasp a reality that continually splinters into subjectivity and ambiguity. But for Tristan and Iseut in the Folie d'Oxford there is finally an end to interpreting: the process of evaluating and verifying signs closes with the joyous celebration of their mutual recognitions. After a sometimes amusing, sometimes disturbing sojourn in the world of ambiguity, where signs remain opaque and memory fails to reconnect past and present, the lovers do manage to return to their original selves, to the truth of their mutual, unfailing love. If in the Tristan myth as a whole different characters (whether spies or Mark himself) claim to have different views of that truth, in the Oxford Folly we are offered the truth of Iseut and Tristan rediscovered and reaffirmed exclusively for themselves (and for the irreplaceable witness Brengain, the stand-in for each one of us who becomes privy to the lovers' secrets). The convincing presence of the human voice—whether Iseut's lamentations or Tristan's selfrevelation—finally rescues the lovers from a world of separation where known and knower remain disconnected, reunites them in a refuge of transparence that recalls and actualizes the crystal chamber imagined by Tristan himself.

The Corporeality of Truth With the Oxford poet's lovers we arrive at a conception of truth circumscribed within human dimensions, both rational and irrational, material and spiritual. I am tempted with the Tristan story to talk about the "corporeality" of truth: change the configuration of material signs that constitute a human being, leave out the sound of Tristan's real voice, and you can no longer coordinate past and present, memory, perception, reason and belief; restore the vital, infinitely fragile sign of a particular human voice and you can once again recognize the total person.52 The truth associated with Tristan and Iseut refuses to settle into a neatly divided, Platonic model with reality and appearance, spirit and form, on opposing sides. The mixed nature of truth asserts itself throughout the Folly episode, but perhaps most forcefully at the end when we suddenly realize that lying need not be a verbal change in words; a physical change in voice acts even more effectively to undermine and then reaffirm the truth of Tristan's renarrations. The lovers' truth does not, of course, coincide with the transcendental

Renarration in the Folie Tristan d^Oxford 35 truth of medieval Christian doctrine. Their reality remains steadfastly rooted in their most physical and emotional identity. Yet, in one sense, the limitations we have seen placed on their human truth seem perfectly orthodox within the larger Christian vision. In the great chain of being all levels of creation have their own degree of goodness, as long as they remain properly subordinated in the overall hierarchy.53 There is even at first view a certain resemblance between the process of recognition that characterizes the Oxford Folie and that which the Church Fathers starting with Augustine describe as the only way to attain true knowledge—that is, knowledge of God.54 There is a similar emphasis on prior knowledge, directly obtained, which can be found again in our memory with the help of certain signs, especially verbal signs. For Augustine, as for Tristan and the author of the Folie^ words are sensible signs that are powerful, but at the same time limited instruments of knowledge. Moreover, love plays an essential role in the recognition of God, just as it does in the lovers' recognitions of each other. But this analogy ultimately points out the irreconcilable difference between God and man or woman as the object of true knowledge. God's being is irrcducibly other; we human creatures necessarily remain in the mediating realm of signs when we try to apprehend that being. On the contrary, the Oxford Foliis ending emphasizes that human beings can know each other directly and completely in the transparence of their inner selves. Human voice can make Tristan and Iseut fully known and present to each other. Where recognition of God leads to union beyond reason and loss of self, human recognition defines the self in relation to the other.55 What is unorthodox finally in the lovers' vision of truth is the shift in emphasis and value, the horizontality of the vision focused on human complexity and limitation. Within the secular context of Tristan's romance world, God's interventions, when they do appear, come "down" in support of the human lovers (see w. 881 and 949). The lovers themselves do not seek transcendence in Christian love, but only in each others' most human arms. The Oxford Folly seems designed to verify, through the most rigorous of tests, the possibility of locating values within human limits. What could be more unstable, more subject to change than human life:1 And if adulterous love, unsanctioned by law and the marriage bond, can be seen as human instability par excellence, then what better arena to test the human capacity for true fidelity:5 Courtly love lyric and the romances inspired by it pose the same problem of fidelity and trust between unmarried lovers. Wherever human relationships are involved, misunderstanding and

36

Chapter i

ambiguity threaten the desire for Constance and love. But the happy ending we and the lovers are offered in the Folie d^Oxford asserts the truth of human fidelity, proven in strictly human terms. If the larger context of the Tristan story, as explicitly recalled in the renarrations, shows how limited the lovers' success is, how separation always returns to require another quest for union, such is the truth of their human condition, necessarily based on material factors as well as on inner being. Iseut and Tristan's love leads not to union, but to a series of minions. The Oxford Folly shares with twelfth-century romance in general a preoccupation with human complexities that integrates failure as well as achievement, a willingness to explore the dimensions of the human without resorting to other-worldly solutions. But the optimism of the Folie Tristan d^Oxford results, I think, from the transfer of closure from the domain of God into the realm of human experience: truth can put off its disguise, when it resumes its own voice to retell its own human story.

2. Putting Off the Ending: Thomas and the Legend of Tristan and Iseut

Thomas's version of the Tristan legend offers its audience a picture of Tristan and Iseut's love that invades the boundaries of our emotional lives. It sets in motion channels of positive and negative fascination whose polarities appear already in the medieval reception of Thomas's work, eagerly translated into different languages and forms, while its own identity as text was reduced to a collection of fragments. These same polarities reappear in the tradition of modern scholarship, which alternately classifies Thomas as the poet who transforms fatal love into a sublime religion of love or the moralist who warns us against the bankruptcy of a love linked to death.1 Such critical ambivalence is, I think, the necessary result of Thomas's choices as author of his own "true" version: he generates such ambivalence in part through the invention of a narrator whose point of view remains essentially unresolved, in part through his choice, amplification, and blocking of material from the diverse possibilities of the Tristan matter. Much of the disagreement about Thomas's version centers on differing interpretations of the ending: is it a success or a failure, endorsement of a union that transcends the boundaries of death or condemnation of a love that kills without redemption? Both kinds of interpretation operate on the same underlying assumption: if Thomas accents the harmony of union, he is "for" the lovers; if, on the contrary, he insists on the failure to achieve true union, he is "against" them. If for the moment we accept that assumption, we need to follow through by testing it against the romance's own criteria for judging union and disunion. Such guidelines are drawn up explicitly by the discourse of narrator and characters and implicitly by the characters' actions. All readers will agree that in Thomas's version doubling occurs at every level and in every facet of the romance, from stylistic doubles that characterize the construction of the octosyllables, to the doubling of characters and episodes.2

38

Chapter 2

Doubling, the Key to Experience, Knowledge, and Identity In Thomas's version, the lovers themselves are especially characterized by their obsessive desire to duplicate the experience of the other in the couple. If we would grasp the principle that underlies their drive to double each other, we might first consider an exchange between Tristan and his strange double, Tristan le Nain (who is not actually a dwarf, but perhaps a miniature version of the first Tristan).3 Though it is Tristan le Nain who does most of the talking, he functions here as the perfect double who expresses out loud and explicitly what the hero himself thinks. Caught in a love triangle that echoes that of the main characters, the second Tristan has come to seek help in a mission to save his beloved, carried off by Estult 1'Orgueilleux. Tristan at once agrees to help win the lady back, but proposes first a return to his castle, in order to prepare for an early morning departure. Tristan le Nain angrily concludes that he is not talking with "Tristran le Amerus," who, having experienced the pain of love, would know the other Tristan's suffering and would act without delay on his behalf: " . . . Par fei, amis W*estes cil que tant a pris! Jo sai que, si Tristan fuisset^ La dolur qu'ai sentisset^ Car Tristran si ad ame tant Qu'il set ben quel mal unt amant. Si Tristran oi'st ma dolur, II m'aidast a icest amur; . . . Qui que vus seiet^ baus amis, Unques ne amastes, go m'est avis. Se seitsez que fad amiste, De ma dolur eussez pite: Que unc ne sot que fad amur Ne put saver que est dolur, E vus, amis que ren amez, Ma dolur sentir ne poez; Se ma dolur pusset sentir^ Dune vuldriez od mei venir."4 (" . . . On my faith, friend, you are not the man who possesses such reknown! I know that if you were Tristan you would feel the sorrow

Thomas and the Legend of Tristan and Iseut

39

I suffer; for Tristan has loved so deeply that he well knows the ill by which lovers are afflicted. If Tristan heard of my sorrow he would aid me in this love-affair. . . . Whoever you are, fair friend, in my opinion you have never loved. If you knew what affection is, you would have pity on my grief. He that never knew love could not know sorrow either; and you, friend, who love no one, are unable to feel my sorrow. Were you able to feel my grief, you would be ready to accompany me." p. 340) Three ideas, three verbs, are repeatedly interwoven in Tristan le Nain's long outburst: if you know love, then you can/££/ another's pain, then you must be Tristan the Lover. All these motifs are moved into the negative by the second Tristan's disappointment in the face of delay. His persuasive oratory, however, provokes an affirmation of his experience and identity from "Tristran le Amerus" (D, w. 1012—15). Experience, knowledge, and identity are thus intertwined and reveal the underlying premise of Tristan's and Iseut's efforts to double each other.5 This is not simply a desire for empathy or sympathy; shared experience furnishes knowledge of self and other that leads to oneness, the identity of the couple as a unit. By blocking the connections between self and other, lack of the same experience leads conversely to "twoness"—ambiguity, multiplicity, and uncertainty. In the description of the couple's experience, three verbs recur like leitmotifs to explore and reiterate the nature of this connection: assaier (to try out), partir (to share or part), and dubler (to double). Each verb is, in fact, a set of double meanings. Given the couple's desire to fuse their two identities into a single, undifferentiated one, every time one of the lovers perceives a gap between self and other—be it across time, space, or sexual identity—he or she tries to try out the other's experience of joy or suffering. As Tristan reasons during his debate on whether or not to marry the second Iseut, how can he eliminate the difference between himself and the Queen if not by trying out—"Tar ovres, par faiz assaier*" (Sn1, v. 158)—by imitating her own situation >6 Assaier: the connection between experience and knowledge is embedded in the verb itself, at whose center appears an expression of knowledge: sai (cf. Tristan le Nain's use of saver, sentir, and estre, quoted above). In Old French the verb partir also expresses the lovers' desire for shared experience. Tristan wants to marry the second Iseut not because he hates the first, but because he wants to share (partir] the way she loves him and the king (Sn1, w. 180—82). Likewise, Iseut decides to wear a

40

Chapter 2

hairshirt in order to recreate in her own body the suffering Tristan has described to her: Partir volt a la penitence Pur go que Tristran veit languir, Ove sa dolur vult partir Si cum ele a Pamur partist Od Tristran que pur li languist, E partir vult ore Tristran A la dolur e a Fahan. (D, w. 746-52; emphasis added)7 (She wishes to share in the penance. Having seen Tristan languishing, she wishes to share in his sorrow. Just as she has shared love with Tristan who has languished for her, so she desires to share the sorrow and the hardship, p. 336) Rut partir also has another meaning in Old French: to part, divide or separate—that is, precisely what the lovers can never really do and yet must repeatedly do. As Iseut announces to Tristan the day they are discovered in the garden: "Bien vos doit menbrer de cest jor Que partistes a tel dolor . . . Ja n'avrai mais, amis, deport, . . . Quant doi partir de vostre amor; Nos cors partir ore convient, Mais Pamor nzpartira nient." (C, w. 41-43, 45, 48-50; emphasis added)8 ("You should well remember this day when you left with such sorrow . . . I shall never more have enjoyment, beloved, . . . when I must separate myself from your love. Our bodies must now part, but our love will remain undivided.55) The same verb partir expresses both the centripetal and centrifugal forces of their love, which bind them together as strongly as they drive them apart. Even more troubling—and dangerous—is the double meaning of dubler. At its best, doubling fuses the two members of the couple, eliminates the difference of now and then, here and there, him and her, and

Thomas and the Legend of Tristan and Iseut

41

thus acts as a source of joy and comfort for Tristan and Iseut. When, however, doubling escapes the bounds of the couple—as it inevitably does, through Tristan's errances—it becomes a source of new pain, further separation and renewed efforts to eliminate difference. Tristan's "duble Isolt" (Sn1, v. 472) unleashes double pain and doubles the number of sufferers:9 A sun mal [Tristan] quert tel vengement Dunt il doblera sun turment. (Sn1, w. 215-16; emphasis added) (For his suffering he seeks such vengeance that he will double his torment, p. 304) Duble paigne, doble dolur Ha dan Tristran por s'amor. (T1, w. 109—10; emphasis added)10 (Lord Tristan has double pain and double grief because of his love. P- 3i7) Iseut aux Blanches Mains, as much as she doubles the name and beauty of Iseut la blonde, remains other; the gap opened between Tristan and the Queen must be filled again—hence the statues of Brengain and Iseut that Tristan makes and visits secretly in the cave, in order to remember and retell their shared past (T1, w. 1—4). Tristan's projects of doubling exacerbate rather than satisfy the desire for unity: the verb dubler catches not only the sought-for duplication (reducing two to one) but also the doubling of sorrow and the multiplication of victims from two to four that leads ultimately to Iseut aux Blanches Mains's duplicity and the duble tref, the two sails, black and white, that signal life or death for the dying Tristan.11

Two Final Visions of Death Now that we have the criteria for judging successful union, based on the connection between experience, knowledge, and identity, and have been apprised of the dangers that accompany such efforts to double the other, we are ready to ask which of these possibilities, what kind of doubling, appears in the final scene. And in order to answer that question, we need

42

Chapter 2

to see the death scene through the separate, converging points of view offered by Tristan, Iseut, and the narrator. Gender is the essential irreducible difference in the couple that, in a sense, originates and fuels the lovers' drive toward unity. This is especially clear in the way each of the lovers approaches the moment of death. Tristan's last vision of their love is a double one, focused at once on Iseut's absence—which he mistakenly believes to be a voluntary refusal to come heal him—and the tie of pity and sorrow (D, w. 1765-68) that he believes will nevertheless link them beyond his own death. When Tristan pronounces Iseut's name three times (his own personal Trinity) and then dies on the fourth, he makes the Queen present, at least verbally, as an acoustical image of the couple's unity. Tristan's last words and his death freeze him in a moment of understanding and misunderstanding: his ermnce, the restless and often erroneous wandering of his thoughts, here and earlier, attributes a lack of constancy to Iseut that bears no relation to the truth of her actions and emotions, yet expresses the ambiguity of Tristan's own errant thoughts, as they draw him toward and away from Iseut. Of course, Iseut, like Tristan, does at times waver in her belief about the absolute unity and unanimity of the couple—and insofar as she does, we see that the narrator's comments on human ermnce as a general principle are indeed generally true. But there is an essential difference between Tristan and Iseut in this respect, which Thomas's text represents without acknowledging in an explicit way through the usual narrative interventions: Iseut's momentary wavering never wanders outside the bounds of the couple—and that despite the opportunities presented by a character like Cariadoc, her would-be lover, whose very unsuitability may be a reflection of how unlikely Iseut's infidelity to Tristan really is. While Thomas's commentary blames women more than men for their inconstancy, his story consistently reveals greater constancy in Iseut than in Tristan. "Cum veraie amie" (Sn2, v. 807), as a true beloved, Iseut always ends up reinforcing her tie to Tristan and the oneness of the couple. This is especially striking in the death scene, when she discovers Tristan's body. Her lament first describes the scene she would have shared with Tristan had she arrived in time to cure him, how she would have recalled all the joy and pain of their love (w. 790-99). But since she has come too late, they can at least die together: "ensemble poissum dune murrir" (v. 801). Iseut announces she will take comfort from the same drink (beivre, v. 805) as Tristan—and thus carries out her vision of their death, as she imagined

Thomas and the Legend of Tristan and Iseut

43

it earlier in a monologue during the storm at sea. Her lament finished, she places her body next to Tristan's in a way that recalls his careful placement of statues in the cave. Like Tristan, she returns to the episode of the garden for a model to duplicate; but while he chose to represent Iseut giving him a ring at the moment of parting (T1, w. 35-40), Iseut chooses rather the earlier moment of unity, when their two bodies lay side by side. In the Cambridge fragment, which gives us the first extant scene in Thomas's version, the discovery of the lovers in the garden, we catch only the end of a description of Iseut lying in Tristan's arms: "Entre ses bras Iseut la reine" (v. i). But we can see Iseut doubling that moment in the final scene, as she arranges their bodies in the quintessential position of lovers, body to body, mouth to mouth: Embrace le, si s'estent, Baise la buche e la face E molt estreit a li 1'enbrace, Cors a cors, buche a buche estent, Sun espirit a itant rent, E murt de juste lui issi Pur la dolur de sun ami. (Sn2, w. 809—15) (She takes him in her arms and then, lying at full length, she kisses his face and lips and clasps him tightly to her. Then straining body to body, mouth to mouth, she at once renders up her spirit and of sorrow for her lover dies thus at his side. p. 353) Within the story it is Iseut who has the last vision of their love, one of complete fidelity and unity. But the narrator's summary quickly reminds us of the duality in their experience of death: Tristrans murut pur sue amur, Ysolt, qu'a tens n'i pout venir. Tristrans murut pur sue amur, E la bele Ysolt par tendrur. (Sn2, w. 816-19) (Tristan died for his love, Iseut because she could not come in time. Tristan died for his love, fair Ysolt because of tender pity. p. 353, slightly modified)

44

Chapter 2

These four verses catch the essence of doubling in Thomas's text, not only as stylistic pattern but as organizing principle in the lovers' lives and story: the exact repetition of the verse describing Tristan's death reminds us of the lovers' desire to fuse two into one, to eliminate all distinction, all difference within the couple; the variations of the two verses describing Iseut's death, alternating, interwoven with Tristan's own, both say the same thing, but with different words, suggest a distinction, a difference that is as fundamental to the couple as their desire for unity. The lovers die together, but consecutively; they die for love, but not exactly for the same reasons. Their unity in death is marked visually by the final description of the two bodies, tightly embraced, but necessarily two, male and female. The difference of gender—the essential difference the lovers obviously have to maintain in their love, however strong their desire to experience and be exactly what the other is and experiences—remains an important aspect in the duality that characterizes their representations of death. Can we attribute Iseut's more positive view of that death—and therefore of the couple's life—at least in part to her identity as woman, life-giver, healer?12 Franchise Barteau suggests that if the couple never renounces their essential unity, even in Thomas, it is Iseut even more than Tristan who retains such faith: "sans doute parce que femme, fille de magicienne et magicienne elle-meme, elle est plus pres des secrets de la vie."13 Iseut as life-giver is completely focused on Tristan: their love never bears fruit in the production of children, but rather exercises its power to give life by restoring Tristan's, that is, ultimately, by keeping the couple alive. Iseut's vision then is more simple, that is to say undivided, than Tristan's double vision, which repeatedly fails to unite them because of the interference of his own uncertainties, jealousies, and desires, while those very desires still compel him to seek the goal of unity across and in spite of the obstacles of diversity and multiplicity. Doubling functions both as attachment and detachment, fusion and separation within the couple, each member supplying more or less of the positive and negative charges that keep the current alive between them. Both lovers die for the sake of the other, but without correct understanding of the other's death. Their desire to share everything, to be doubles of each other, falls necessarily short of identity, though they may come as close to doubling each other as any two human beings, male and female, can. In the ordering of the last scene we are presented first with Tristan's view of the couple and death, then with Iseut's: a mixture of positive and negative views, followed by a fully positive one. With the lovers' own final visions, we have an example for prac-

Thomas and the Legend of Tristan and Iseut

45

tically any judgment of their success or failure—any judgment that is, except a totally negative one, since the opposition presented is not a clear distinction between negative and positive views, but rather an asymmetrical opposition in three overlapping parts: negative + positive vs. positive.

The Narrator's Double View What about the narrator's role here? We must now move into the second phase of analysis to explore the narrator's relationship to the story he narrates. He has shown himself on the whole to be particularly in tune with the kind of errance that characterizes Tristan's feelings and less "conscious"—at least in his commentary—of the singleness of mind more characteristic of Iseut. On the other hand, it is interesting to see a verbal link in the closing commentary between Iseut's own description of herself as "true friend" and the narrator's insistence on the truth of his version, as promised at the beginning: "[E dit ai] tute la verur, / [Si cum] jo pramis al primur" (Sn2, w. 828-29). What is the truth of Thomas's version!1 Is it to be located in the mixed pessimism of Tristan's view or in the optimism of Iseut's? In order to answer such questions, even partially, we need to take a closer look at Thomas's narrator, his own self-representation, as well as his relation to the story told. The kind of sympathetic interjection or exclamation that Beroul's narrator typically uses to mark a new episode or emphasize a dramatic moment in his narrative can also be found in Thomas's romance. But, if we can judge by the extant fragments and thus generalize about his narratorial habits, such exclamations appear infrequently—in fact, only twice in the three thousand or so verses that remain (C, v. 6; D, w. 1583-84). Much more characteristic of Thomas's narrator are the long commentaries in which he analyzes and elaborates his characters' situations. He frequently seems to enjoy the status of omniscient narrator, able to report the actions, words, and inner feelings of his characters— a narrator with special interest and penetration in the area of human psychology, as commentators of Thomas have long remarked. The vocabulary of his commentaries reveals, moreover, a continuity between his own discourse and that of his characters: a shared linguistic experience especially connects his analytic discussions to Tristan's own interior monologues, as they both explore the same sets of oppositions—desir/poeir, desirlvoleir, amurlhaur, colvertise/franchise (desire/power, desire/will, love/hate, base

46

Chapter 2

action/nobility of heart).14 They play the same verbal game of numbers in which two opposed terms present the challenge of finding a third that will resolve or replace the opposition: two's flirting with one's—or is it three's?—but invariably remaining two.15 This pattern of debate that shows Tristan to be a homo duplex16 is mirrored by the narrator's own pleasurable entrapment in stylistic doubling. The narrator's knowledge suggests then a kind of participation in his character's lives that recalls the fusion sought by Tristan and Iseut within the unit of the couple. We may notice, however, as the narrator picks up and repeats his characters' vocabulary, in order to explore and then magnify their particular experience into generalities about human nature, that the process of generalization occasionally leads his commentary off the mark. For example, when the narrator generalizes from a specific analysis of Tristan's motivations in wanting to marry Iseut aux Blanches Mains (Sn1, w. 233-91), his description evokes a mentality that we can easily recognize: "the grass is always greener on the other side." Novelerie fait gurpir Buen poeir pur malveis desir, E le bien, que aveir puet, laissier Pur sei el mal delitier; . . . (Sn1, w. 255-58)

(Thirst for novelty makes [some people] sacrifice the good in their power for wicked craving, and to give up the good which they can have in order to seek pleasure in evil; . . . p. 304) But does this really apply to Tristan's case? Is he really giving up "what he can have" to get something else, which only his fancy presents to him as something better? While this may be true on the spiritual level of Tristan's love for the Queen, it is not true on the physical level: in exile from England, serving first in Spain, now in Brittany, he does not presently "have" the Queen—and physical possession is precisely what most bothers him at this point in the narrative. Furthermore, again the narrator singles out women as especially prone to this sort of behavior (Sn1, v. 287-90), while throughout the story it is Tristan alone, and never Iseut (not even Iseut aux Blanches Mains), who exhibits a desire for novelerie. In fact, the narrator himself seems to recognize the irrelevance of his commentary, as he interrupts it by remarking: "Ne sai, certes, que je en die" (Sn1, v. 291).17 The last set of generalities in this long series is supported by the nar-

Thomas and the Legend of Tristan and Iseut

47

rator's own direct observations, since he has seen such things happen to many: "A molz ai veii avenir" (Sn1, v. 345). His guarantee of veracity comes from the realm of experience, not as personally lived, but as personally observed. This is a distinction that signals another important aspect of the narrator's self-representation vis-a-vis the story he narrates: he has not experienced for himself what his characters suffer. Hici ne sai que dire puisse, Quel de aus quatre a greignor angoisse, Ne la raison dire ne sai, For ce que esprove ne Pai. (T1, w. 144-47) (I do not know what to say here as to which of the four was in greater torment, nor how to tell the truth about it, because I have not experienced it. p. 317) Thomas's stance here is remarkable in its particular differences. First, it calls attention to the narratorial stance of romancers (like Chretien) who are typically disengaged from their story material, as far as any personal experience is concerned. As Thomas's disengagement becomes explicit, it seems to anticipate or acknowledge that some author/narrators are not so uninvolved. Thomas's narrator thus recalls and at the same time differs from romancers, like the anonymous author of Partonopeu de Elois or Renaut de Beaujeu in Le Bel Inconnu, who are personally engaged in the stories narrated, even linking their heroes' fortunes in love to the author/ narrators' own success with a lady. Thomas's German translator, Gottfried von Strassburg, does not follow his example here, but rather specifically identifies himself as a lover who addresses a select public of lovers and shares in Tristan's and Iseut's experience. When he comments on their life in the "Cave of Lovers," for example, Gottfried adds: "There was a time when I, too, led such a life, and I thought it quite sufficient" (Hatto, p. 264). There is no such empathy announced between Thomas and his matter. With his claim of ignorance, on the contrary, Thomas's narrator excuses himself from judging which of his main characters suffers most. Lack of experience cuts him off from both sets of triangles that link all four together in a web of love and suffering. As the narrator nevertheless continues his discussion of their respective states, he stresses in particular his ignorance in relation to Iseut aux Blanches Mains. She sleeps in a bed with her husband, but the narrator knows neither their joy nor their troubles (T1, v. 187).

48

Chapter 2 Ne sai se rien de delit set Ou issi vivre aimme ou het; Bien put dire, si Ten pesast, Ja en son tens ne le celast, Com ele Pa, a ses amis. (T1, w. 190-94) (I cannot say what she knows of pleasure, or whether she loves or hates such a life. But I can aver that, had it irked her, she would never have hidden this from her friends as she has done. p. 319)

The narrator twice repeats his ignorance and then volunteers an interpretation of her feelings that seem strangely at odds with her particular situation, since her silence on the matter of her unusual marital state is directly motivated elsewhere by Tristan's request for secrecy regarding his supposed wound (Sn1, w, 626—28). Similarly, in a later commentary on "woman's anger" (D, v. 1323), motivated by Iseut aux Blanches Mains' plans for vengeance, the narrator first warms to his typically misogynistic theme and then backs off abruptly, claiming that he does not dare to speak his mind on a topic that has nothing to do with him: "Mais jo ne os mun ben dire / Car il n'affirt nient a mei" (D, w. 1334—35).18 Iseut aux Blanches Mains appears, on the one hand, to be the character most alienated from the narrator's experience and thus the one least illuminated by his analytic skills (while Tristan, as we have seen, seems the most central to his narrator's psychological concerns). On the other hand, these verses suggest, however mysteriously, a possible correspondence he prefers to suppress: the narrator does not dare say what is bothering him, since that would be inappropriate to his role. His silence capping any further speculation on his own experience in love, he thus ends the discourse on "ire de femme" and resumes an uncommented description of the actions and feelings of his character. Ruthmarie Mitsch contrasts Thomas's narratorial habits with Gottfried von Strassburg's, whose extended metaphors and ornate descriptions allow the narrator to express his knowledge, superior to that of the characters, and thus move the reader to compassion, while Thomas allows the lovers to speak for themselves in monologue; his narrator makes no judgment on them, only interprets by way of general comment.19 It is remarkable how little the narrator does intervene in the course of telling his tale—which consists chiefly of his characters' monologues and dialogues, with the narration of actual events down to a minimum. This may seem

Thomas and the Legend of Tristan and Iseut

49

paradoxical in a version particularly known for its narrator's commentaries: when he intervenes, he tends to do so at length. On the other hand, once he begins narrating, he does so with very few of those fillers—"ce m'est avis," "a mien escient," and so on—with which a Chretien de Troyes, for example, keeps reminding us that our reception of the tale is constantly filtered by the narrator's point of view. Our first impression of omniscience obviously must be corrected: Thomas as narrator is quite willing to step aside from what his characters know and experience. Indeed, since he seems to accept the same premise they do regarding the link between shared experience and knowledge, we are led by his own explicit commentary to question the limits of his reliability. How well he knows, understands, and represents his characters varies with the character involved. His "omniscience" dips considerably, as he moves from Tristan to Iseut to Iseut aux Blanches Mains, as his level of "sympathetic" participation rises and falls. But even when he seems most fused with a character, as he is with Tristan, we have been warned not to confuse the narrator's experience and that of the lovers. Alternately, and sometimes simultaneously, attached to and detached from his story and characters, Thomas is a narrator above all bent on delineating his role as that of the teller of a tale, while the best judges of the tale told are the lovers who read or hear his version. After his claim of ignorance about which of his characters suffers the most, he adds: La parole mettrai avant, Le jugement facent amant, Al quel estoit mieuz de 1'amor Ou sanz lui ait greignor dolur. (T1, w. 148—51) (I will put the case before you—let lovers pass their judgement as to who was best placed in love, or who, lacking it, had most sorrow. P- 3i7) He repeats the same idea, as he concludes a review of the four sufferers, inviting one who knows ("qui set" T1, v. 152) to decide now who has the best of love, who the worst suffering. While dissociating himself from the characters, the narrator simultaneously associates his public with them: lovers know and can therefore judge what the narrator does not know and can only present. Once again, such statements imply that only direct, lived experience gives us knowledge of others—at least in the domain of love.20

50

Chapter 2

The Epilogue Such instruction returns us once again to the end of Thomas's romance to follow through on our evaluation of his narrative stance in the concluding verses. As we begin to read the epilogue, we are immediately struck by an abrupt shift from the end of the lovers' story—narrated with dramatic concision, its high points marked by the direct discourse of the characters themselves—to the narrator's final commentary, where we are removed from the world of fiction and placed in its surrounding context, that of the storyteller and his literary public. After the four verses quoted above linking and differentiating Tristan and Iseut in death, Thomas begins his epilogue by naming himself in the gesture of closing his work and sending a greeting to his public of lovers, to all those who will hear his verses (Sn2, w. 810-11, 825). We should notice what is left out by this narrative shift. After the lovers die, there is nothing, no survivors, no miraculous plants growing out of the lovers' tombs, as we see in Eilhart's version or in the later French prose romance.21 In Thomas, our final vision of the lovers remains fixed on two bodies, side by side, closely embraced; no transcendance moves them beyond that final, static vision. Tristan's earlier message to Iseut, as elaborated to Kaherdin, prepares us for the final blocking of the death scene. The first part of Tristan's message lyrically embroiders his salut d'amw, playing on the different meanings of salut, as his salutation gradually leads into the question of health and salvation (D, w. 1195—1208). This is not a religious salut, but one consistently tied to the saving powers of Iseut and love. The two alternatives envisaged by Tristan are death without Iseut, life and health ("Salu de vie ne sante," D, v. 1203) with her. As Pierre Le Gentil has pointed out, the lovers call upon God only to make sure that they remain together in death.22 Tristan as martyr of love may play Saint Alexis under the staircase, but he does not project a view of joyous union in Paradise, nor does Thomas's narrative suggest any such extrapolation beyond death. In contrast to the enormous drive for doubling that characterizes all aspects of Tristan and Iseut's story until this final moment, their death is represented as an absolute end to all further doubling. While it doubles back to the earlier scene in the garden, it refuses to include the double beyond death offered by the legend of the miraculous plants, new perennial doubles for the lovers. From this point of absolute stasis—which is,

Thomas and the Legend of Tristan and Iseut

51

in fact, the infinite mirroring of the doubles already constructed reflecting each other back and forth within the confines of Thomas's version—we in the public are ejected by the closing comments into another order of discourse, where the narrator's commentary opens the doors so dramatically closed by the plot.23 Two lyric enumerations mark the epilogue's opening and close. The first evokes Thomas's public: Tumas fine ci sun escrit: A tuz amanz saluz i dit, As pensis e as amerus, As emvius, as desirus, As enveisiez e as purvers, [A tuz eels] ki orunt ces vers. (Sn2, w. 820-25) (Here Thomas ends his book. Now he takes leave of all lovers, the sad and the amorous, the jealous and the desirous, the gay and the distraught, and all who will hear these lines, p. 353) The second describes the example he has offered to lovers and enumerates their sorrows: Aveir em poissent grant confort, Encuntre change, encontre tort, Encuntre paine, encuntre dolur, Encuntre tuiz engins d'amur! (Sn2, w. 836-39) (May [lovers] derive comfort from it, in the face of fickleness and injury, in the face of hardship and grief, in the face of all the wiles of love. p. 353) Like Tristan to Iseut, the narrator sends his scdut to all lovers, lovers of all types.24 In between these two enumerations, the storyteller describes his own role in relation to the story told and the possible reactions of his public. His vocabulary here recalls not only his earlier commentaries, but also key words in Tristan's own discourse: the saluz, the opposition between voleir an&poeir, the important verb recorder, and of course the words that express love's pain (paine., dolur, engins). What is striking in this clos-

52

Chapter 2

ing commentary is the combination of voices, one presented by the named author/narrator designated in the third person, the other by the pronoun jo. One is the lyric voice of the poet/lover we recognize from troubadour and trouvere poetry: that lyric CT' claims personal experience at the base of his exploration of love's conventions; it is an "I" fused with the poetic expression of love. Thomas's narrator is fully capable of speaking with such a voice, as he has abundantly shown through the language of Tristan's monologues. The narrator can thus dub/double the voice of the lover, but it is not his own; he has a distinct voice as storyteller, the detached voice of the romancer who mediates between story and public, who identifies himself in the epilogue by name, just as Chretien does in some of his romances. In the closing commentary we notice again the narrator's capacity for fusion and separation in regard to his matter—just as the final image of the lovers captures the essence of their union and disunion. As narrator, Thomas makes no comment on the lovers' story per se, but rather on his treatment of it and its usefulness for the public. The story is exemplary for the public of lovers. A model for emulation or warning? We are not told, though we might speculate that the same story would function differently for the different types of lovers enumerated. We are told the story has been embellished to please lovers and to offer them the comforts of remembering their own love suffering as mirrored in that of Tristan and Iseut. The doubling now ended within the story is projected out beyond the world of fiction to link characters and audience. It is a lyric voice that has the last word here, expressive of an identification with the bitter experience of love. The last series of nouns—arranged in a pattern of anaphora and doubling synonyms culminating in the single, all-inclusive "tuiz engins d'amur" (v. 839)—highlights the negative effects and powers of love, as does Thomas's version in general. Does this suggest a negative comment on the matter that Thomas has not made explicit? But whose personal experience and knowledge is furnishing the sense of identification with the lovers expressed here? The narrator has explicitly denied such involvement and refers us to our own experience as lovers. The audience's sentimental energies are surely engaged by the story's example and by this last encantatory enumeration, but our powers of judgment may be as varied as our capacity for different types of love.25 Indeed, we might attribute the ambiguity of Thomas's narrative stance, neither definitely opponent nor apologist, to his desire to attract such a varied public.

Thomas and the Legend of Tristan and Iseut 53 Thomas and the Tristan Legend Thomas is aware, nevertheless, that not all will like his version, though he is ready to guarantee its truthfulness: [S]i dit n'ai a tuz lor voleir, [Le] milz ai dit a mon poeir, [E dit ai] tute la verur. (Sn2, w. 826-28) (If I have not pleased all with my tale, I have told it to the best of my power and have narrated the whole truth, p. 353) These verses allude to a problem raised earlier in the romance, the famous passage where Thomas discusses different versions of the significant events leading to the end of the story (D, w. 835—84). Thomas names himself in both places, here situating himself in relation to a community of lovers (his public), there in relation to a community of storytellers (his predecessors, colleagues, and rivals). A discussion of that key passage inevitably focuses on the narrator/author's relationship with the Tristan legend as a whole. The issues raised in Thomas's commentary concern the diversity of the Tristan matter and the unity of his own version. As such, it raises further significant perspectives on the issues of doubling, fusion, and detachment. Seignurs, cest cunte est mult divers, E pur c,o Funi par mes vers E di en tant cum est mester E le surplus voil relesser. Ne vol pas trop en uni dire: Ici diverse la matyre. (D, w. 835—40) (My lords, this tale is quite diverse and for this reason I have brought it together through my verses. I tell of it as much as is needed and I want to leave out the remainder. I do not want to include too much in one version. The matter diverges at this point.) Thomas thus interrupts the flow of his narrative and speaks directly to his audience about the nature of his material. He places his intervention just

54

Chapter 2

after the account of Tristan's two returns to Mark's court, first as leper, then in an unspecified disguise in order to see about Iseut's hairshirt and participate in knightly sports. The theme of vengeance, which seems to take over the second episode with the death of Cariadoc and the hasty departure of Tristan and Kaherdin recalls Beroul's treatment of the same materials. These are magnificently orchestrated in the drama at Mai Pas and Blanche Lande, a theatrical spectacle that gives full play to Tristan's antics as leper and the high drama of Iseut's ambiguous oath. Thomas may not have known Beroul's particular rendition of the legend, but he was certainly familiar with what Jean Frappier has called the "common version," an example of which we can recognize in Beroul's somewhat archaizing romance—that is, the sort of version that predates what we are accustomed to call Thomas's "courtly" adaptation. While the dating of Thomas and Beroul is too inconclusive to determine any definite precedence, I think it significant that such comparisons emerge just as Thomas's passage on different versions of the Tristan legend comes into view. I am further struck by Thomas's insistence on the unity and economy of his own version, just when I have noticed the contrast between Beroul's use of the materials to create a single, dramatic spectacle (even if played out in various acts) and Thomas's more choppy treatment, which takes bits and pieces of the same materials and reweaves them into the episode of Brengain's quarrel with Iseut, along with the scene under the staircase not found in Beroul's version. Thomas appears to have "cut and pasted" a series of scenes elsewhere "unified," but here "doubled:" Tristan's two returns, Tristan as Saint Alexis, and so on. Given this narrative context, we can now focus on some key terms in Thomas's commentary, the verb unir and the expression en uni dire. Douglas Kelly has given us a masterful explanation of these terms, as they relate to the contemporary discussions in rhetorical treatises on the art of composition.26 Thomas has brought together in his poem diverse Tristan matter: "1'uni par mes vers." But the kind of unity he seeks is not based on inclusiveness: "E le surplus voil relesser / Ne vol pas trop en uni dire" (D, w. 838-39). The governing principle here is based on continuity and coherence: some part, the surplus, will be excluded when the diverse materials of the Tristan legend are put together in a logical, unified narrative. How is this coherence to be judged? Thomas spends a good deal of time discussing just that problem, both in general terms and in specific examples within the story. Reason not surprisingly plays a significant role here.27 It is called upon twice within this commentary to frame, with in-

Thomas and the Legend of Tristan and Iseut 55 troduction and conclusion, Thomas's discussion of the "wrong" version he rejects on logical grounds. Thomas accepts neither Kaherdin as the lover of the Tristan le Nain's wife nor Governal as Tristan's messenger to England: Thomas icx> granter ne volt, E si volt par raisoin mustrer Qu'igo ne put pas esteer. (D, w. 862-64) (Thomas declines to accept this and is ready to prove that it could not have been the case. p. 338) The argument made is based on the question of recognition: as earlier episodes in the story indicate, Governal is too well known by Mark and his court not to be recognized as Tristan and Iseut's messenger.28 Those who tell the story in this way have therefore strayed from the story and the truth, but the authority of reason will eventually guarantee Thomas's authority as storyteller (D, w. 879-84). He does not, however, rely on reason alone to establish his authority and that of his text. Even before calling upon the powers of logical demonstration, Thomas has been careful to associate himself with recognized authorities among storytellers, both in written and oral accounts: Entre ecus que solent cunter E del cunte Tristran parler, II en cuntent diversement: O'i en ai de plusur gent. Assez sai que chescun en dit E go que il unt mis en escrit, Mes sulun go que fai oi\ Nel dient pas sulun Breri Ky solt les gestes e les cuntes De tuz les reis, de tuz les cuntes Ki orent este en Bretaingne. Ensurquetut de cest'ovraigne Plusurs de noz granter ne volent (/) que del naim dire ci solent. (D, w. 841-54; emphasis added) (Those who are accustomed to narrate and tell the tale of Tristan tell it differently—I have heard various people do so. I know well enough

56

Chapter 2 what each says and what they have put into writing; but to judge by what I have heard, they do not follow Breri, who knew all the deeds and stories of all the kings and all the counts that had lived in Britain. Moreover, many of us are unwilling to assent to what narrators are accustomed to say about the dwarf, p. 338; slightly modified)

Thomas here stresses his personal experience as listener and reader of other versions in order to establish his authority. It is what he has heard and read that gives him knowledge of the true version, and that knowledge of truth is clearly distinguished from the habit of storytelling that is reflected in the many erroneous versions of Tristan's tale (N.B. the two repetitions of the verb soler). The logical argument given in w. 865-76 comes as confirmation of the truth that stems first from Thomas's direct knowledge of the version given by Breri, whose extensive knowledge in the matter of Britain is emphasized.29 The preterite solt associated with Breri, in contrast to the present tense sai of the narrator, establishes the order and transmission of authority. I have stressed another important verb in this passage, the verb voter, used frequently throughout this comentary to distinguish what Thomas wants or rather does not want to tell from what other storytellers want or do not want (see also w. 838, 839, 862, 863, 881, 882). Desire is the omnipresent feature of Tristan and Iseut's story: the verb voler plays a key role in representing that desire, both in the characters' own discourse and in the narrator's descriptions. The desire associated with the storyteller is, of course, different from that of his characters. We have seen earlier how he carefully distinguishes his knowledge as teller from any direct experience of love—which is rather the bond between story and audience. The personal experience and desire that Thomas does associate with himself are all concentrated in his activities as transmitter of a diverse matter, his identity as textual space across which the tale moves and from which other versions stray (N.R.forsveie., v. 879). The truth appears along one path, the path of unity and coherence; the author/text that demonstrates that oneness in itself becomes ultimately its own authority. Despite the essential difference in the types of experience described, Thomas thus represents himself, in regard to the multiplication and unification of the Tristan matter, to be involved in the same problematic as the lovers. Interwoven on both levels are the problems of truth and recognition, diversity and unity, desire and reason. The verb diverser, here used to describe the diversity of versions, has appeared earlier to describe

Thomas and the Legend of Tristan and Iseut 57 Tristan, not only the variety of his thoughts ("E pense molt diversement," Sn1, v. 2), but his perception of the difference between his life and Iseut's ("'Molt diverse nostre vie,5" Sn1, v. 6). Tristan likewise echoes the narrator's "par raisun mustrer" (D, v. 863), when he accepts Tristan le Nain's identification of Tristan the Lover ("Par grant reisun mustre 1'avez," D, v. 1012).30 Jean Frappier has clearly demonstrated the way Thomas's version associates reason and love, reason as the guide of love's proper conduct.31 Reason also guides the storyteller's desire to guarantee one version of the diverse Tristan matter. Reason will be the judge of two desires, jo's for his story, other storytellers' for theirs: Ne voil vers eus estriver; Tengent le lur e jo le men: La raisun s'i pruvera ben! (D, w. 882-84) (I don't want to quarrel with them. Let them keep to their version and I to mine. Reason will demonstrate who's right!) At the beginning of his intervention, Thomas tells us that he wants to leave aside le surplus (v. 838), telling us only what is necessary. The word surplus is not uncommon among twelfth-century writers. We may remember it in other contexts: the surplus of lovers most romancers decline to narrate, the surplus of meaning that later writers add to or find in earlier works.32 Le surplus thus offers further connections between the world of the author/text and that of the characters. If for the moment I willfully misread Thomas's surplus in its own context and seek out its erotic connotations, I am tempted to contrast the author's and the lovers' conduct in desire. Thomas describes for us his discretion, his ability to give up something—one kind of unity based on totality—in order to achieve a different kind of unity through coherence.33 The lovers, on the contrary, show themselves unable to give up the surplus^ the sexual connection between them that generates doubling, multiplies the objects of desire. Thomas thus differentiates himself or detaches himself from the lovers' conduct. But this initial contrast yields quickly to a more profound correspondence between author and characters. Having sacrificed a unity based on non-selective inclusiveness, Thomas then fuses the different materials of his text through his own (re)invention of doubles: paradoxically, simplification proceeds through doubling. His romance expands enormously, and reminds us, on another level, of Thomas's earlier remarks about no-

58

Chapter 2

velerie: that thirst for novelty he finds in Tristan is a characteristic we might just as readily apply to Thomas himself, as he exhibits the storyteller's desire to extend his tale, put off the ending, with new characters, new episodes, new adventures. Thomas's thousands and thousands of verses (most of them, unfortunately, known to us only through translations) are a testimony to the general human principle of novelerie. On the other hand, the penchant toward unlimited expansion is still controlled: the romance remains coherent through its own pleasurable self-reflection, the mirror images that multiply apparent differences, but remain unified through repetition and doubling. This is the same pattern that underlies the couple's own life: the repeated cycles of separation and re (union), separation and (re)union. The logic of doubles that operates here in the organization of the story—with its double tendency both to multiply and simplify—is the same logic that proves to be so fruitful for the interpreter of Thomas's version. Two opposing yet complementary patterns characterize that enterprise. First, "the more you look, the more you find": it seems impossible to exhaust the list of doubles, since the more you think about the different aspects of the story together—at whatever level—the more doubles you begin to perceive, especially since the same element, character, scene, and so on, can have more than one double, as suggested above. Secondly, "the more you find, the more you find it's all the same": what at first appears different becomes the same; the interpreting eye puts together the double, sees the unity of meaning behind the varied appearances of the text. We readers, already invited by the narrator to share our experiences as lovers with that of the lovers in the story, find ourselves equally involved in the storyteller's desire for doubling. Do we see one or two, or something that constantly shifts back and forth between unity and difference!1 Such questions lead us to another well-known context for le surplus in the twelfth century: the Prologue to the Lais (w. 10-16) where Marie de France discusses the surplus de sen, the addition of meaning—gloss, commentary, interpretation—that subsequent writers bring to or discover in the works of their predecessor. It is both true and untrue that Thomas leaves out such a surplus of meaning. On the one hand, his version itself represents a surplus as it reshapes the traditional material of the Tristan legend. On the other hand, here as in Marie's Lais the interpretation that any reshaping implies remains quite elusive—if we are to judge by the diversity of interpretations that mark our attempts to characterize the exact nature of Thomas's "truth." Le surplus is somehow present and absent

Thomas and the Legend of Tristan and Iseut 59 at the same time. The narrator speaks at length in the mode of commentary and yet, paradoxically, he remains extraordinarily discrete, allows the actions and characters of his story to speak for themselves. We are thus forced to take seriously the narrator's own claims to ignorance. Whether or not such ignorance is a fiction, it is certainly a ploy, endorsed by the author's blocking of the story, to shift the burden of concluding onto us.34 The narrator's lack of experience in matters of love removes his capacity for judgment. He can neither endorse nor condemn, but only—to the best of his limited ability—tell. If we return after this long excursus through narratorial experience and authorial practice to the judgments offered by the lovers themselves, we have an example of no single judgment, but rather a continually oscillating one, a judgment always renewed out of the asymmetrical tension that opposes Tristan's double vision of wandering and fidelity to Iseut's single vision of union in love. Thomas has not given us a key to eliminate any of the possible points of view represented within his romance. He has instead left us spellbound, caught between the two voices of lyric identification and critical distance that echo in our ears, as our eyes remain fixed on the final image, two bodies intertwined in death.

3. A Case for mise en abyme: Chretien's Chevalier de la Charrete

"Or est venuz qui 1'aunera!" ("The one comes who will take their measure!") J The herald's repeated cry to announce Lancelot's unequaled performance at the tournament of Noauz applies with equal accuracy to Chretien's accomplishments as romancer. He has certainly been "the one who has given the measure" for twelfth century and subsequent romance writing, as attested by the judgments and practice of medieval writers themselves and by the enormous critical and scholarly bibliography that continues to recognize and explore the exemplary status of his works for medieval romance.2 But just as the herald's cry sounds like a riddle that provokes questions from the spectators who first hear him, we modern readers of Chretien may be moved to ask what and how his romances are measuring. Questions of measurement—judgment, evaluation, interpretation—seem to be the central issues posed by Le Chevalier de la Charrete both internally for the Arthurian world and externally for the romance public. Around the unique figure of Lancelot, the problems of interpretation and judgment crystallize throughout the romance, but they do so most explicitly in the tournament episode itself. By focusing internally on that episode, we can begin to explore how Chretien plays with the resources of fiction as the object and instrument of measurement. Intratextual analysis will naturally lead to the extratextual world of author and patron, where the same issues arise. That exploration will lead in turn to the intertextual play first within Chretien's corpus, then between his romances and the Tristan story. The complete circuit, in its perambulations along the complex and beguiling byways of Chretien's reinventions, should finally give us a better understanding of how medieval textuality and fictionality interact, in order to represent the contradictions and complexities that arise in translating an Arthurian ideal into the context of twelfth-century feudal society.

Chretien's Chevalier de la Charrete

61

The Tournament at Noauz: mm en abyme Beginning in the middle is an approach that Chretien clearly favors, since all his romances (except Glides] begin in medias res, locate us in a series of events already underway. But in what sense can we say that the tournament episode itself is "in the middle?" It is not, as far as verse numbers are concerned: the episode's approximately seven hundred verses (5359-6056) occur rather toward the end of a romance just over seven thousand verses long. Centrality appears rather on the metaphorical level: the tournament episode operates as a mise en abyme of the entire romance, both internally with respect to its narrative and thematic organization and externally in terms of authorial production and reader response. Through its repetitions and variations, the tournament recapitulates what has already happened and anticipates what is to follow. Caught in the network of echoes and analogues, which typically link the juxtaposed episodes of romance, the tournament is at the same time isolated from the rest of the narrative chain in a way that enhances its ability to collect together, intensify, and thereby crystallize the narrative threads as well as the conflicts raised by the Charrete. As mise en abyme the tournament is located at the heart of those issues most crucial to the matiere et san (v. 26) of the Charrete. The episode is literally a center insofar as the tournament at Noauz is sandwiched between Lancelot's two places of imprisonment which mark, as we learn later from Godefroi de Leigni's epilogue, the pivotal moment when the first author turns over his charge to the second. The tournament appears then as the last major episode separating and tying together the work of both authors and thus sheds light on their complementary relationship vis-a-vis story, patroness, and public. Displacement, isolation, gratuitousness, fictionality—these are the key terms that will recur, somewhat paradoxically, throughout my analysis of the tournament episode as central to the problems posed by the Charrete. The tournament's isolation within the series of episodes is motivated and guaranteed by its geographical position—significantly not at Arthur's court—and by the secrecy required by Lancelot's desire for anonymity. Not referred to elsewhere in the romance, the tournament enjoys on one level a certain gratuitousness within the plot structure, as suggested by the doubling of Lancelot's imprisonment before and after the episode.3 Since Lancelot has already been imprisoned by Meleagant, strictly speaking we could omit the tournament from the list of events and still arrive at the ending in pretty much the same way. On the other hand, what appears

62

Chapter 3

gratuitous, doubled, at the level of plot, is often significant for that very reason when we move to other levels of analysis. Just as the heraldic mise en abyme reproduces in its center the shield's general design, in such a way that none of the repeated pattern touches any of the other figures,4 just so the tournament's isolation sets it apart while at the same time focusing its ability to reflect in miniature the surrounding narrative. The special location of the tournament episode is further highlighted, if we compare the narrative structure of the Charrete with that of two other romances by Chretien. Peter Haidu has demonstrated how Erec and Tvain reflect a typical model for twelfth-century romance structure that, with variations, serves quite well to demarcate the narrative structure of numerous other twelfth-century romances.5 When we try to locate the Charrete narrative with respect to this general model, however, we discover that it does not really fit into the model; it cannot be adequately represented by the divisions. 6 This difference constitutes one of the major anomalies that sets Lancelot apart within Chretien's corpus. It is then all the more curious that, in spite of significant changes in the Charrete's overall narrative structure, the tournament episode occurs in approximately the same position (w. 5359-6056) as the Joie de la Con episode in Erec (w. 5319-6358) and the Pesme-Aventure episode in Tvain (w. 5101—803), identified by Haidu as the extraordinary, "superfluous" adventure that crowns the hero's second series of successful exploits, well before the actual end of the romance. Although the marvelous element that characterizes these episodes "hors serie" has been displaced to earlier adventures in the first part of the Charrete—especially the cemetery episode—the structural similarity (which mirrors their similarity in content, since all three episodes entail formalized conflict) lends the tournament scene a kind of authority as summation of the hero's exploits. As such, the tournament at Noauz places Lancelot on a level beyond normal chivalric performance, as he demonstrates with particular clarity (at least for the cognoscenti) his own, anomalous heroism. Chretien's choice of a tournament for this kind of recapitulation is not without significance for the romance. A tournament is, by its nature, already a displacement—not "serious" combat, such as we saw earlier between Lancelot and Meleagant; not war, but combat as game, display, pageant. We know, of course, from contemporary accounts that tournaments of the twelfth century were by no means harmless and resembled more the melee of battle than the highly stylized jousts of later medieval tourneys. But if we consider how the tournament is represented in the

Chretien's Chevalier de let Charrete

63

Charrete's description, what we see is precisely tournament as show, a spectacle to elicit commentary, favorable and unfavorable. Ahatine, the word most frequently used throughout the episode to name the event taking place is, significantly, a synonym for tornoi that means not only tournament but discussion, provocation, quarrel.7 In fact, we hardly see any fighting at all. Only forty-two verses describe actual combat, all focused on Lancelot's performance: the first day's prowess and cowardice (w. 5657—73), the second day's prowess alone (w. 5934—51,5966—70,5980—81). All these descriptions given by the narrator are generalized accounts, except for Lancelot's first joust the second day against the King of Ireland's son. Throughout the tournament, the narrator has for the most part turned his job over to the spectators, through whose eyes and comments is filtered knightly and unknightly conduct.8 Rather than offer us direct description of the fighting, the narrator multiplies the spectacle through the varying and contradictory perspectives of herald, damsels, Queen, and knights, combatants and non-combatants (including Gauvain), who witness and evaluate what they see. The enumeration and description of the best knights' shields, which initiates the second day of the tournament, exemplifies this tendency to see the tournament as an object of contemplation. The catalogue (w. 5773—822) is offered by knights who are not themselves fighting, whether because they are prisoners or have taken the cross (w. 5769—70). Furthermore, what they describe is not what the most admired knights are doing, but rather what designs are painted on their shields or where the shields came from. What stands out here is the poetry of naming knights and places, whose orderly rhythms are marked by the repetitions of "Veez vos . . . Et veez vos . . . c'est . . . cil . . . Et cil autres . . . " ("Do you see . . . and do you see ... it's . . . the one . . . and that other one . . ."). The visual pleasure of the spectators is doubled by our verbal pleasure, clearly the author's as well, though he allows his characters to speak for him. The heraldic images fix each knight's identity as representation and recognition of prowess, rather than prowess itself. The displacement from action to commentary is, in a sense, the raison d}etre of the tournament as planned by the female organizers, who want to marry shortly: s'anpristrent a eel parlemant une ahatine et un tornoi. Vers celi de Pomelegoi Panprist la dame de Noauz.

64

Chapter 3 De eels qui le feront noauz ne tandront parole de rien, mes de ces qui le feront bien di'ent que les voldront amer. (w. 5366-73) (They decided in the course of their discussions / To organize a splendid tourney, / In which the lady of Pomelegloi / Was challenged by the lady of Noauz. / The women will refuse to speak / To those who fare poorly, / But to those who do well / They promise to grant their love.)

In this description the narrator cleverly sets up the pun on noauz (whose two meanings are matched at the rhyme) and anticipates the Queen's subversion of the damsels' project: despite the organizer's intentions, no one at the tournament will remain silent about the knight who fights au noauz, except the Queen herself, who has arranged the spectacle of Lancelot's contradictory performance. This conflict between diverging intentions or contradictory arrangements emphasizes again the character of the tournament as displacement, which we might identify here as a displacement toward fictionality.9 This is not to deny that tournaments are a recognizable part of social reality in the twelfth century, which had for contemporaries important social and economic functions. As war games viewed both positively and negatively, tournaments provided a training ground for knights and an arena in which to develop and display reputations.10 The career of Guillaume le Marechal gives ample testimony to the way tournaments functioned as a kind of job fair and potentially lucrative source of equipment and money, in the exchange of prisoners and ransoms. Chretien's representation of the tournament at Noauz, however, focuses not on economic realities, but on the tournament as replica, a game or model consciously used and manipulated, set up to be witnessed and evaluated as such. Arranged by the Queen's commands, Lancelot's performance in particular calls attention to the tournament as "pretend": a pretend war whose real object is matrimony; a series of real combats, in which Lancelot pretends to be a coward. Of course, the same sort of displacement also characterizes romance as a genre whose self-consciousness as fiction has often been remarked.11 Displacement within displacement, the tournament seems to designate the Charrete as a sign referring to what it is or is not, a fiction that plays with

Chretien's Chevalier de la Charrete

65

history and reality as it crosses over their borders, opens them to communicate with the world of imagination and play. Lucien Dallenbach points out that it is precisely those texts that are most inclined to call attention to themselves as recits that voluntarily duplicate themselves through mise en abyme (pp. 78-79). The declared fictionality of Chretien's romance thus plays an important role as the starting point for any attempt to understand how and what Lancelot measures.

Intratextual Recalls Given this overview of the tournament episode, let me turn now to its specifics in order to substantiate how it reflects and reorients the entire romance, from the levels of plot and characters to those of themes and meaning. Lancelot's conduct at the tournament has often been recognized as a reprise of the cart scene, a correction of his former two-step hesitation to obey Love's commands, an obedience promised during his reconciliation with the Queen, which itself sets up a link between her cold reception and their eventual night of love.12 Clearly this is one important chain of events, linking Lancelot and the Queen, which the tournament episode recalls and elaborates. But these are not the first events thus recalled; we need to situate them within the opening gestures of the episode to appreciate how Chretien signals to his readers that we should be prepared to read both forward and backward the linear progression of the scene. Having just described the Queen's return to Arthur's court—a return that reveals Lancelot's absence and the deception of the letter that earlier announced his return—the narrator opens a loop in the narrative by going backward to the beginning of the tournament plans made while the Queen was out of the country (w. 5359—60).13 In the Queen's absence, the damsels are worried about their marriage arrangements (cf. Guenevere's role in Cliges as matchmaker for Alexandre and Soredamor). When she returns, they are anxious to have her attend the tournament already organized. In order to secure the Queen's presence, the damsels approach the king and ask for a gift, which he grants before knowing what it is they want to request (w. 5386-96). This is an obvious, if benign, replay of the opening scene in which Keu's Rash Boon leads to the Queen's departure from court. The description of Arthur, in particular, insists on his habit of granting unspecified requests ("cil qui rien veher ne sialt," v. 5395) and

66

Chapter 3

shows that he has learned nothing from his previous experience about the possible dangers of granting requests before they are specified. Even his last-minute deference to the Queen's wishes ("s'ele le vialt," v. 5396) disappears behind the fact that he has already granted the request, whatever it may be, according to the damsels' desire (ccil feroit quan qu'eles voloient," v. 5391), before making any condition to this promise. The King's position vis-a-vis the Queen has not changed, nor has his essential lack of concern for those far from his court and imprisoned in Gorre. His resignation in the face of Meleagant's announcement about the prisoners (w. 61-62) corresponds to the ease with which he forgets his grief for Lancelot in the joy of the Queen's return (w. 5352-58). The King's shortsightedness, even and especially in relation to the "object" he loves best ("la rien . . . que il plus vialt," v. 5357), plays off most disadvantageously against Lancelot's own obsessive devotion and passionate action on behalf of the Queen. Of course, the serious and dangerous consequences of the Rash Boon in the opening scene are now transposed into a comic mode: no danger lurks in wait for the Queen as she leaves Arthur's court to attend the tournament at Noauz. The repetition calls attention to the pattern itself and thus constitutes a displacement toward fictionality, as it emphasizes the character of play and the games, both serious and comic, the characters play on each other. Though Guenevere has once again been the object of other people's manipulations, she will soon be the one pulling the strings, taking control of other people's actions at the tournament, laughing secretly to herself after others have laughed foolishly in public. When he reveals the location of Lancelot's imprisonment, the narrator vigorously curses the disloyal traitor, Meleagant—"may he burn in hellfire" ("que max feus arde!" v. 5427). At this juncture, Meleagant's name signals one of the major displacements of the tournament episode, which sidesteps the ongoing quarrel between Lancelot and Meleagant. Their suspended combat is yet to be completed, will never be completed, if Meleagant can keep Lancelot from Arthur's court. The tournament as combat, however, refers indirectly to the series of combats between Lancelot and Meleagant, and it does so most particularly through the figure of the King of Ireland's son, who performs well at the tournament, although the unrecognized Lancelot pleases the crowd four times as much (w. 5631-33). Seen immediately in comparison with Lancelot, the same knight is described at the end of the first day as mistakenly believing himself the best:

Chretien's Chevalier de la Charrete

67

li filz le roi d'Irlande pansse sanz contredit et sanz desfansse qu'il ait tot le los et le pris; mes laidemant i a mespris qu'assez i ot de ses parauz. Nei's li chevaliers vermauz. . . . (w. 5709—14) (The son of the king of Ireland felt / That beyond any doubt and without contradiction / He himself deserved all the esteem and reknown; / But he was terribly mistaken, / For there were many equal to him. / Even the Red Knight. . . .) Once again the narrator links the King of Ireland's son to Lancelot, as he prepares their encounter on the second day: the knight who overestimates his own performance is easily unhorsed, once Lancelot unleashes his true prowess. As Bademagu repeatedly points out (e.g., w. 6304—26), the knight who praises himself is not the one who earns real praise and esteem. The King of Ireland's son acts as a displaced, minor version of Meleagant (cf. the Hospitable Damsel's suitor), who recalls and anticipates Lancelot's suspended and final victories over the self-deluding traitor. This preliminary inventory of the tournament as mise en abyme concludes with one last set of characters and events from the first part of the romance, introduced in this episode by the seneschal's wife and her agreement (covant). As any number of readers have noted, Lancelot constantly attracts the attention and help of ladies, married and unmarried, whom he encounters along his route.14 It seems as if Lancelot's intense passion for the Queen generates around him a magnetic field of erotic potential. I have elsewhere shown how the Hospitable Damsel episode sets up a chain of analogues that chart the contradictory directions of desire.15 The seneschal's wife specifically echoes that earlier episode by her offer of a covant (v. 5477). She will grant Lancelot permission to go to the tournament in exchange for his promise to return and give her his love. "—Dame, tote celi [amor] que j'ai vos doing je voir au revenir. —Or m'an puis a neant tenir, fet la dame tot an riant; autrui, par le mien esciant,

68

Chapter 3 avez bailliee et comandee Pamor que vos ai demandee. Et neporcant sanz nul desdaing, tant con g'en puis avoir, s'an praing." (w. 5482-90) ("My lady, I will certainly give you / All that I have upon my return." / "Now I see that I am nothing to you," / The lady answered with a laugh: / "It seems to me that you / Have assigned and given to another / This love I have asked of you. / Nevertheless, without disdain / I shall take all I can have.")

The wife's light-hearted attitude suggests once again that this is a brief, comic replay of that earlier, more serious encounter—serious, that is, if we consider the dangers of the rape scene and Lancelot's anguish in fulfilling the condition that he share his hostess's bed.16 Here Lancelot demonstrates in his reply not only a prudence that contrasts well with King Arthur's unhesitating accord of Rash Boons, but also a subtlety of language that recalls the ambiguous oath scene with Meleagant and anticipates the Queen's own play on verbal ambiguity in the message au noauz. "All the love that I have" rings with generosity, a generosity that characterizes Lancelot's relationship with the fair sex in all his encounters. The damsels at crossroad and ford, Meleagant's sister (in the Orgueilleux episode and later when she rescues Lancelot from the tower), all witness Lancelot's largece^ whether in word or deed. But the seneschal's wife is not fooled by this apparent gift of love: like the Hospitable Damsel and the Queen's pucele, she knows that Lancelot's love has already been given, is already under someone else's command. Their night of love may remain a secret, but Lancelot's devotion to the Queen seems to be common knowledge, particularly among women, whether in Logres or in Gorre. On the other hand, the seneschal's wife does not disdain Lancelot's offer, nor does she seem to consider the words empty of value. Lancelot's great heart and his love for the Queen generate an excess that overflows to the benefit of the Arthurian world, collectively and individually. Lancelot as the nexus of desire is precisely what we will see in the tournament episode itself, in the secret reactions of the Queen and the public comments of the damsels. They have organized the tournament for this very purpose: prowess demonstrated in combat is to be rewarded by love and marriage. But the connections between prowess and love, love and marriage are disrupted by Lancelot's presence incognito and by the

Chretien's Chevalier de la Charrete

69

Queen's commands. The game of disruption begins when Lancelot dons the red armor and rides off on the excellent horse offered by the seneschal's wife: they belong to her husband. She has asked for Lancelot's love; he has promised what he has to offer. The bargain struck, she has logically put Lancelot in her husband's place. Again the tone here is comic, but this is not the first time Lancelot has occupied the place of a husband. The displacement of that role is one of the major problems raised in the margins of silence that operate so ambiguously in and around the complex texture of the Charrete. In some sense, we can situate Lancelot's current position with respect to a general principle that seems to underlie his characterization: where Lancelot is, there marriage is not (literally and figuratively). Where Lancelot is, marriage is suspended, put aside, displaced. The declared goals of the tournament organizers are to single out those knights who perform the best, neglect the worst, and marry the former. Instead, they find their attention riveted by one knight who alternately performs as the best and the worst. Lancelot's singularity, both as paradox and hyperbole without equal, eliminates the competition. This seems to be true even from the male point of view, since among the noncombatants, nothing pleases Gauvain more than to contemplate the unknown's exploits (w. 5958-60). All the damsels desire only Lancelot, though none believes she deserves such a knight (w. 5985-6001). All vow to marry only the Red Knight and decide to remain unmarried for the year, when he disappears at the end of the second day (w. 6002-6, 6051—54). Lancelot's feats and his disappearance have opened up an interval of non-action on the marriage scene, which echoes a pattern of suspended action characteristic of the romance as a whole. The Red Knight's performance as a kind of disruptive or disappearing center points out the distance that separates Lancelot from other Arthurian knights and isolates him from the court. The location of the tournament at Noauz is significant in this respect, as is the catalogue of the best knights fighting: while a number of names are familiar from other romances as members of the Round Table, they are really nothing more than names, without even the status of secondary characters. We might usefully compare this with the variation in Cliges. Though Cliges's presence at Arthur's court is brief, he is privileged to encounter the best, when he attends a three-day tournament at Oxford. Incognito, he triumphs over Sagremor, Lancelot, and Perceval in the first jousts of each day. The tournament closes when Arthur ends the contest between Gauvain and Cliges before a winner can be recognized (w. 4595—919). The knights Lancelot encoun-

70

Chapter 3

ters at Noauz do not include even the main secondary figures like Keu or Sagremor. Gauvain, already outdistanced by Lancelot's performance during the quest for the Queen, remains among the spectators, content to reflect Lancelot's glory in his admiring view. Although Cliges himself is an episodic character with respect to the Arthurian kingdom, his temporary centrality brings out by contrast the more enduring character of Lancelot as permanently and crucially located on the periphery of Arthur's court. Lancelot is caught in the play between center and margin, or perhaps between competing centers, in a drama to determine who will occupy the court's heart, literally and figuratively. The leitmotif of Lancelot's absence from court is, in some sense, the counterpart of his uniqueness.17 Lancelot is not exemplary in the positive mode: his is not an example that can be imitated. A hint of Lancelot's otherness is suggested in the mysterious reference to his childhood with a fairy (w. 2345-50). Does this difference motivate his preference for anonymity, amply demonstrated during his quest for the Queen and reaffirmed at the tournament by both the narrator (v. 6043) and himself (to the herald, w. 5550-55)? This is part of his modesty surely, but also part of his mobility as a character who crosses boundaries, sheds reputations and identities, even his own insofar as they would limit action. By choosing to appear as the unrecognized Red Knight, Lancelot remains officially absent from the tournament and isolates his presence there in the same way that characterizes the night of love. If bloody sheets indicate that someone besides Arthur has been in the Queen's bed, the person they point to is not Lancelot. Where the issue of marriage (and adultery) appears, Lancelot disappears.

The Process of Recognition: The Herald If we would get to the heart of the issues raised by the tournament episode, we must focus now on the recognitions that do occur at Noauz: Lancelot is not invisible to all. The distribution of knowledge before, during, and after the tournament is a critical ingredient here. First and foremost, we in the public know who the Red Knight is, because this time we have followed his actions "from the beginning." This is a major change from the first part of the romance, where we were almost as ignorant as the participants about the identity of the knight coming to the Queen's rescue. Carefully engineered by the narrator's use of circumlocutions that

Chretien's Chevalier de la Charrete

71

repeat the known (e.g., "the one who rode in the cart") without revealing the desired unknown, our ignorance of Lancelot's name in the first part enmeshed us in the same process of discovery repeatedly undertaken by the romance characters themselves, even though we shared some of the narrator's more comprehensive point of view. By contrast, at the tournament we do not recognize Lancelot per se. Since we think we know the answer to the riddles posed by the herald's cry and the Queen's message, the process of recognition and non-recognition set off by those enigmatic words leads us to a different kind of recognition, not limited to a character's name and identity, but linked rather to Lancelot's role as touchstone for the entire Arthurian society and its conflicting values. But our more general process of investigation necessarily begins by observing that of the characters. The herald is the first to recognize Lancelot: how and why does he do so? In theory, his recognition seems logical enough. That is what we expect heralds to do at tournaments: identify knights by their shields and announce who they are to the public. As mentioned earlier, however, this function is actually performed on the second day of the tournament by the knights who are watching the fighting. Just out of the tavern where he has lost coat and shoes (w. 5535—41), the herald fails to identify the one shield he comes across, the one Lancelot has left outside his miserable lodging. We know it belongs to Meleagant's seneschal: we do not know what it looks like. The shield remains a blank on which those watching the Red Knight's performance will inscribe an identity, cowardice or courage. The herald, however, sees Lancelot before he goes into action: he peeks inside, sees Lancelot unarmed, and recognizes him immediately (w. 5548-49). The measure of his surprise is given by the gesture of crossing himself, while the instant recognition tells us that Lancelot is a figure already well-known and easy enough to recognize "in the flesh." At this stage, there seems to be a simple division between outside (cover) and inside (truth), but that apparent simplicity will soon be obscured by a more complex interplay of appearance and reality. The herald's cry itself helps spread a sense of confusion, insofar as he has been forced by Lancelot's most ungentle threats to curb his professional desire to call out the knight's name. Wanting to reveal, but required to conceal, the herald uses the same sort of circumlocution favored by the narrator in the first part of the Charrete, but this time phrased in the future (cf. the inscription on the tombstone), to describe what Lancelot will do instead of what he has already done: "Or est venuz qui 1'aunera!" It is at this point that the narrator

72

Chapter 3

identifies the herald as our master ("nostre mestre," v. 5572): as the first one to use this expression, he is the one who has taught us to say it.18 Whatever its historical origin, the fictional etiology presented here focuses our attention on the cry itself before its dramatic role is fully developed. It is surely noteworthy that this is the only time in the narration of the tournament episode when the narrator intervenes with more than a passing remark to fill out a verse. His comment associates the herald directly with himself and the romance public. We may see ourselves a bit in the herald's shoes (though his have been left at the tavern), as we peek behind the scenes and imagine we know what is going on, since we know something other people are dying to know. Like the narrator, the herald is promising a good story. He has set up expectations among his hearers, but for what? The cry itself seems enigmatic. The verb auner may mean either to measure with an aune (a kind of square stick) or to strike with it.19 And what about the pronoun ('Taunera") that appears and disappears, depending on the manuscript?20 Its reference is definite and yet unspecified. Of course we know, once we have read the entire scene, that this call will be another focus for the crowd's mockery, when Lancelot's cowardice seems to contradict the herald's expectation. Once the gap between announcement and action has been created, however, it is just as gleefully and unexpectedly suppressed by Lancelot's superlative prowess. From the point of view of the crowd, a disjunction between words and performance has appeared: the cry seems sometimes true, sometimes false; now it fits, now it doesn't. This is a pattern that we recognize as typical of the Charrete narrative, which seems to be motivated by a series of puzzles: Meleagant's challenge and Keu's desire to leave in the opening scene, the Cart Knight's identity, the Queen's cold reception at Bath, the bloody sheets, Gauvain's and then Lancelot's whereabouts. As soon as one puzzle is solved, another appears to keep open the forward momentum of the narrative.21 More specifically, the herald's cry echoes those rumors that circulate, telling stories about Lancelot during the quest, or about Lancelot and the Queen during his first (mistaken) capture by the people of Gorre. The two sets of rumors that follow Lancelot's journey to Gorre are both true: one describes past action, his ride in the cart; one describes his identity as liberator of the prisoners and predicts future action. When both rumors come together, their apparent incompatibility causes upset. What crime was Lancelot accused of that he deserved a cart ride? Surely no one ever saw such a courageaous knight, if he were free of such a reproach (w. 2606-22). Seen

Chretien's Chevalier de la Charrete

73

through their reactions, the problem of rumor as an expression of truth begins to emerge here: it is true that Lancelot rode in a cart, but the logic of expectation based on the customs surrounding cart usage in those days, such as the narrator is at great pains to explain to his contemporary public (w. 321-44), has been disrupted by Lancelot's voluntary entrance into the cart for quite other reasons than crime and punishment. The rumor is true, then again not true, an anomaly confirmed by its eventual disappearance from public view, despite all the predictions that a ride in the cart constitutes an ineradicable stain. This intertwining of truth and its opposite will concern us again shortly.

The Process of Recognition: The Queen If the herald is able to recognize Lancelot in a sudden flash, the Queen's recognition is played out, by contrast, through the main action of the tournament itself. She sees Lancelot only when everyone else does, when he appears in full armor, after the first jousts have already taken place. His entrance dramatized by the herald's cry, Lancelot's identity is the subject of questions even before he begins to show off his prowess, which, in turn, intensifies the crowd's desire to know who he is: Trestuit de demander s'angoissent: "Qui est cil qui si bien le fet?" Et la reine a consoil tret une pucele cointe et sage et dit: "Dameisele, un message vos estuet feire, . . . a ce chevalier m'an alez qui porte eel escu vermoil; et si li dites a consoil que cau noauz' que je li mant." (w. 5634-39, 5642-45) (They all troubled over the question: / "Who is this knight who fights so well?" / The queen whispered to a clever, pretty girl, / And said: "Miss, you must / Take a message . . . / . . . to that knight / Carrying the red shield, / And tell him in secret / That I bid him to do 'his worst.'")

74

Chapter 3

The Queen's action is immediately set off by contrast and juxtaposition with the other spectator's questions, which furnish a context for understanding something about her conduct. The romance public can see Guenevere's message as indicating a recognition already made—or at least, as a test of her hypothesis. Certainly the private conversation between Queen and pucele recalls an earlier one concerning Lancelot's identity that took place when an unknown knight first fought Meleagant. While there the initiative came from the pucele and led to a public announcement of Lancelot's name—a revelation for the reader/listener as much as for the characters—here it is the Queen who acts first, in order to reveal Lancelot's identity by concealing it from everyone else. This ambiguity, protected by the verbal ambiguity of the message itself as it puns on noauz, plays on the same pattern as the herald's cry. It is, in fact, the conjunction of these two messages that engineers the contradiction and confusion attached to Lancelot's subsequent performance—an echo of that earlier moment when the combination of Lancelot's two reputations as cartrider and liberator sowed anguish and confusion among his hosts. To the disjunction between words and meaning signaled by competing interpretations of noauz corresponds the disjunction between words and deeds generated not only by Lancelot's paradoxical conduct, but also by the crowd's mockery. Mes tex dit sovant mal d'autrui qui est molt pires de celui que il blasme et que il despit. (w. 5759-61) (But often the one who speaks ill of another / Is far worse that the one / He slanders and despises.) The narrator's warning about the misdirection of words could stand as an epigraph to the entire romance. Where both types of disjunction continually complicate the characters' and the public's efforts to figure out what is going on, we need to clarify as much as possible what motivates the Queen's recognition, apparently immediate and yet requiring a process of proof stretched over two days and three messages. The capacity to recognize Lancelot is not, I think, to be explained in terms of wdisemblance. Gauvain's failure to recognize Lancelot, both here and earlier during their first night of shared hospitality, may appear invraisemblable to a public whose expectations are based on a systematic use of realism. Realism has its own episodic impor-

Chretien's Chevalier de la Charrete

75

tance as an element of representation in the Charrete^ but much more significant for any overall interpretation are the patterns of repetition and variation that structure and implicitly comment on character and narrative development. I suggested earlier that ladies are especially attuned to Lancelot and thus more likely to figure out the puzzles surrounding him. In this respect, the Queen is the lady par excellence, the beloved whose secret connection with her lover gives her an advantage no one else can share within the world of romance. We who are outside that world are given access to her privileged position through the narrator. What are we invited to recognize through Guenevere's message, as it signals and plays upon Lancelot's hidden signature? Consider again Lancelot's entrance into the tournament fighting. The elaborate attention given to his appearance incognito should remind us first that Lancelot consistently and voluntarily keeps his identity a secret throughout his quest for the Queen. Neither extreme scorn nor praise leads him to reveal his name or show any concern about the rumors that double his progress toward the Queen. He allows each one to invent a different story about his identity, creating multiples that grow up outside the reputation he has clearly established before the romance begins. Lancelot's own use of incognito is the sine qua non that underlies the play inscribed in both the herald's cry and the Queen's messages. In some sense, they are repeating what he has already done, though their variations on the pattern are nonetheless significant. There is another detail, less dramatic perhaps, but just as important for recognizing what is typical of Lancelot and uniquely his own. Lancelot was not present at the tournament's first jousts: "Mes n'i ot point de Lancelot / a cele premiere asanblee" (w. 5612—13). Lancelot's initial absence at the tournament recalls his crucial absence from the opening scene of the romance and it signals his essential difference. Romance heroes, like Erec, like Cliges, typically demonstrate their superiority and enthusiasm from the very first joust to open a tournament—not so Lancelot. Over and over again, we have seen Lancelot hesitate before achieving his ultimate triumphs.22 Should he go to the Hospitable Damsel's rescue? Should he cut off the Orgueilleux's head or grant him mercy:1 The title event of the romance constitutes the primary instance (first and foremost) of Lancelot's hesitation: his two-step inner debate before stepping into the cart. It is a commonplace of Charrete criticism to recognize Lancelot's immediate acquiescence to the Queen's commands, au noauz and au mialz, as the mirror image that voluntarily corrects his earlier hesitation to follow the

76

Chapter 3

dictates of love. Lancelot's delayed entrance into the tournament suggests that the issues raised by hesitation are more complicated than any simple reversal of the common pattern.23 They involve the proper and improper conduct not only of Lancelot but of the Queen, not only of the lovers in the private realm but of knight and Queen in the public. If, as I am suggesting, the Queen recognizes Lancelot through his slight delay before a brilliant performance—the pattern typical of Lancelot and of Lancelot alone throughout the romance—her immediate recognition is nevertheless subject to caution. The last time she received news of Lancelot it turned out to be a deception, despite the convincing appearance of the letter whose signs convinced all that they should believe it: "entresaignes tes / qu'il durent croire, et bien le crurent" (w. 5270-71). The narrator does not specify what these believable signs were, but his phrase emphasizes their believability. He suggests, on the one hand, how important is the role of belief in any interpretation of signs and, on the other, how unreliable credibility per se may be. We can find multiple examples in the romance that corroborate this problematic situation, from the messages on the tombstones in the cemetery of the future to the blood found simultaneously on the Queen's and Keu's sheets. Understandably suspicious of (possibly misleading) signs, even of highly credible ones, the Queen devises a test that will confirm or deny her belief that the knight carrying the red shield is indeed Lancelot, despite her lover's disappearance into a kind of black hole from which not even news of his whereabouts manages to escape. The form of the test itself indicates how well the Queen has read Lancelot's character (and the previous events of the romance). First, she frames her message as a pun whose double meaning will be apparent only to her obedient lover; secondly, she invents a scenario for the knight/lover that accurately reflects in its own paradoxical demands the paradoxical character Lancelot himself demonstrated earlier. Under the Queen's command, Lancelot acts out his own hesitating style, as he alternately races toward and withdraws from chivalric action, pretends to be afraid and seeks out shame (w. 5669-73). In the exaggerated fashion that calls attention to the fictionality of Lancelot's play-acting (at least from our point of view, if not that of all the spectators), we may recognize a voluntary, comical replay of Lancelot's unintentionally comic eclipses as the knight so lost in meditation on the Queen that he falls unceremoniously into a ford, unhorsed by its knightly guard; the knight so occupied by contemplation of his beloved that he defends himself only weakly from behind against a Meleagant intent on

Chretien's Chevalier de la Charrete 77 taking advantage of his opponent's foolishness. We are reminded further of Lancelot's immediate obedience, when he hears the Queen agree to Bademagu's request that she stop the combat between himself and Meleagant. With the same semblance of non-action, Lancelot both repeats the pattern of hesitation that confirms his love and reverses any hesitation that disputes it. Lancelot has learned (not) to hesitate—under the appropriate circumstances. On the second day, when Lancelot once again fails to appear in the opening engagements, the Queen sends the same message via the same pucele^ who, in expectation of just such a repeat, had duly noted the place of Lancelot's withdrawal the previous evening. Why this repetition, so insistently signaled by the details about the messenger? On the one hand, we can see it as a necessary confirmation that the first day's believable signs were the product neither of chance nor deception; they truly indicated the Red Knight's real identity. Repetition in the world of romance is never gratuitous or empty of meaning; it constitutes the very path by which we can attain meaning through interpretation, measurement of repeated elements in the play of variation. Here the variation itself shifts the nature of the Queen's test: the second day her message au noauz is repeated, but what matters is not the performance it requires, but the response it generates (w. 5836—44). Lancelot's unqualified intention to carry out the Queen's command is all Guenevere needs to be sure not only that the Red Knight is Lancelot but that Lancelot is as entirely hers as she is his ("ce est cil cui ele est tote / et il toz suens sanz nule faille," w. 5874—75).24 Once this truth is confirmed without any possible doubt by the very repetition of proofs, the Queen hastens to send the new message au mialz—with the spectacular results we have already seen. While Lancelot's temporary cowardice (an exaggerated form of hesitation) thus serves as a kind of foil to enhance by contrast his real display of prowess, we should not forget that even as the worst knight Lancelot held the attention of the crowd more than any other knight at the tournament. Lancelot's unique heroism is tied to this paradoxical combination of extremes. Under the impetus of his love for the Queen, Lancelot acts out his secret and paradoxical identity under the watchful eyes of all, in the public as in the private domain. If only those who have penetrated the secrets of his private world can truly perceive and understand the complexities of who he is, all (regardless of their knowledge or ignorance) are attracted to the singularity of Lancelot's performance.

78

Chapter 3

Delay and Characterization of the Lovers While the Queen thus deepens her understanding of Lancelot, we may do the same in regard to the Queen. As Douglas Kelly has pointed out, Guenevere's character is revealed only in stages.25 In that process, the tournament constitutes a pivotal moment that casts light both on her previous conduct and the role she will play in the closing scene of the Charrete. The treatment of Guenevere's character corroborates, like so many other elements, the narrator's desire not to clarify too quickly the mysteries and misinterpretations that abound within and around his story matter. The expression used to describe Lancelot's feigned cowardice ("fet sanblant" v. 5672) appears earlier in the romance, when the narrator recounts Guenevere's enigmatic behavior at Bath. After the first combat between Lancelot and Meleagant, Bademagu brings Lancelot to the Queen, who appears angry: "fetsanblant de correciee" (v. 3940). The verb faire sanblant is ambiguous in this passage: when we first read it, we probably interpret it as an accurate description of how she appears and give no thought to any possible gap between her semblance and the reality of her inner feelings. This is fully in accord with Old French usage, which differs from the more limited modern use of the expression. In a first reading, what occupies us, as it does Bademagu, is the gap between what we expected the Queen to do in greeting Lancelot and her refusal to show him the kind of gratitude his service so clearly deserves. Only later during the Queen's monologue, when she thinks Lancelot dead, do we learn that her appearance of anger was feigned, a joke that misfired, since Lancelot took it seriously (w. 4205-7). Only when Lancelot returns to Bath, with sufficient courage to ask the question he dared not ask earlier, do we learn what the Queen was joking about. Without explaining to Lancelot that it was a joke, she relates her anger to Lancelot's hesitation before the cart, which she interprets as an expression of fear or doubt about the competing demands of love and honor (w. 4484-85). This piecemeal process of revelation about the Queen's underlying motivation—never totally available to any character within the story, not even Lancelot—engenders a good deal of puzzlement and suspense, while at the same time giving ample room for the misinterpretations of all those who try to understand the Queen's actions, from Bademagu to Lancelot himself. The perfect lover is sure the Queen is justified in her response (w. 4339-40) and correctly guesses that her "sanblant" must have been related

Chretien's Chevalier de la Charrete

79

to his ride in the cart, but he attributes the misunderstanding of love and shame to the Queen (w. 4336-96): "Ce deust ele amor conter; et c'est la provance veraie: Amors ensi les suens essaie, ensi conuist ele les suens. Mes ma dame ne fu pas buens cist servises; bien le provai au sanblant que an li trovai." (w. 4372-78; emphasis added) ("She should have counted this as love, and this is the true proof, [for] thus does love test her own and thus know her own. But this service did not please my lady, as I proved from the welcome [appearance] she showed me.55) The path from sanblant to provance veraie, from sign to meaning, true or false, is treacherously misleading, even for those with the most privileged, inside information. The fact is, the Queen herself recognizes that she made a serious mistake with a bad joke. She judges herself to be foolish—worse, treacherous and cruel (w. 4201—4). We need to compare the Queen's vision of her mistake with that of Bademagu, who chastised her earlier for her lack of welcome (w. 394751). Bademagu is described there as noble and courtly, "frans et cortois" (v. 3948), probably to furnish a rhyme for his identification as "rois." But this description is not just an empty filler. It is precisely his possession of appropriately courtly attributes that enables him to arbitrate what is proper, courtly behavior—by which standard the Queen fails to do what she should. She is in the wrong (v. 3997), Bademagu later repeats to Lancelot, since his risk of death in her service demands a warm welcome from her, while the non-welcome given presupposes some misdeed (v. 3991) that neither he nor Lancelot can fathom (cf. the erroneous expectations generated by Lancelot's cart ride). The mistake Guenevere makes—as we can understand it from outside the romance, by combining both Bademagu's and the Queen's perspectives—is to act out in public what should remain a private matter between herself and the man who has acted under the sole impetus of love for her (however much such action also rebounded to the public good).

8o

Chapter 3

Once the Queen sees the effects of her mistake and the possible dangers imposed for Lancelot, she is never again represented in the romance as trespassing the bounds set up by her public role as Queen. Her ability to learn and change contrasts markedly with that of the King, who, as we saw earlier, is able neither to control his emotions nor to adjust his behavior to correct past mistakes. Whatever the intensity of grief or joy inwardly felt for her lover, Guenevere contains these feelings without interrupting the smooth flow of her public function. While a certain amount of feeling, positive or negative as circumstance requires, is appropriately shown for the knight who has done so much on her behalf, these feelings are released as part of her public role as Queen, disconnected from any private passion. She thus adopts the position of Bademagu and repeats it back to him, when the rumor of Lancelot's death has been corrected by a true report: "Trop me fust ma joie estrangiee, s'uns chevaliers an mon servise eiist mort receue et prise." (w. 4420-22) ("My joy would leave me altogether / If Death were to claim and take / A knight in my service.") When Gauvain returns without Lancelot, she knows she must show him a semblance (sanblant, v. 5201) of great joy, despite her unhappiness. When Lancelot returns in the last scene, the narrator elaborates the difference between her public presence and private absence, the split between body and heart that echoes the description of Lancelot's eyes and heart during their first meeting at Bath (w. 3970-80).26 The narrator alludes explicitly to her earlier cold reception and feigned anger (v. 6834) when explaining how reason encloses and ties up her foolish heart and thoughts, fol cuer andfolpanse (v. 6847). Reason makes her accept the postponement of the joyous welcome her heart desires to share with her lover, but cannot in the presence of the court, where prying eyes might discover their love. Inevitably we are drawn to compare this with Lancelot's first inner debate between Reason and Love: the comparison at once brings the lovers together and reveals how different is the lesson each of them has had to learn. The Queen has learned the virtue of delay, the very motif that so uniquely expresses Lancelot's character as knight and lover. Whereas for Lancelot the term refers to his ability to cross boundaries set up by love and prowess, public and private, for the Queen it operates to maintain

Chretien's Chevalier de la Charrete

81

these divisions in her own behavior. At the simplest level, there is an exchange of roles, a chasse-croise that places them on opposite sides: one learns to hesitate, the other not to hesitate. Understood with greater complexity, the motif of delay sets the lovers apart from their society, as they choose to hesitate or not under the dictates of love. The interplay of hesitation and its opposite thus not only traces the trajectory of Lancelot's chivalric quest and his subsequent return to court but also characterizes the overall representation of his love relationship with the Queen. If we consider the night of love at Bath as marking a center, the moment of maximum union both physical and spiritual, we can see the first half of the romance as a series of movements toward and away from and again toward that culmination, just as the second half appears to reverse the process in the gradual stages of Lancelot's returns to court, first secretly at Noauz, then openly at Arthur's court in the final scene. If, during the first part, we and the lovers are uncertain about the nature and mutuality their love—an uncertainty figured by the lovers' hesitations— during the second part the physical separation of the lovers no longer puts into doubt their spiritual engagement. The tournament episode plays a pivotal role in this respect, since in Chretien's romance it is the first time the lovers are brought together within Arthur's kingdom. Though their contact is indirect, channeled through the mediation of publicly performed messages and role-playing, this displaced repetition of their night of love confirms, through its variation on private union within public distance, the total commitment each has made to the other. The redefinition of the Queen's role on the side of hesitation furnishes a correction not only for the cold reception scene, but also for the opening scene at Arthur's court. There the King's lack of hesitation, unchecked by the Queen, leads first to Guenevere's falling in a most unqueenly fashion at Keu's feet, then to her being led off by a defender whose weakness is apparent to all the court. While only Keu himself is blamed for this in the Charrete (w. 185—87), the characters in Yvain, who twice allude to this scene, are less reserved in their criticism (w. 3702, 3920). Did the Queen momentarily forget the circumspect guidance she demonstrated in Erec et Enide when she made sure the King's observance of the White Stag Custom did not cause a total disruption of the court's order and peace!1 Whatever the reason for her passivity in the opening scene of the Charrete, she herself falls as first victim to the combined manipulations of the King and Keu, who acts as if in anger (v. 86). Keu's fiction becomes apparent only when he obtains the Rash Boon—through Arthur's own unwilling con-

82

Chapter 3

nivance inasmuch as he sends the Queen to fall at Keu's feet until he promises to stay. Guenevere is thus made to beg her own misfortune, as the role of her defender passes by default from the King to Keu to Gauvain and ultimately must fall to Lancelot for her successful return to court. If the lesson drawn from this opening scene should lead the Queen not to hesitate before Arthur's precipitation (cf. w. 6174-80) and the lesson of the cold reception leads her to hesitate before Lancelot's hesitation, we may see the results of both lessons acted out in the tournament episode. Where the Queen herself assumed a sanblftnt, in her unwelcome of Lancelot, she now gives the role of actor to Lancelot, who has already given himself the role of anonymous Red Knight. While the displacement of roles among her defenders earlier placed her in jeopardy, now she will do the manipulating and displacing for very different effects.27 As we saw earlier, her arrangement of Lancelot's actions in the contrasting patterns of worst and best leads to a kind of joy and knowledge between the lovers that echoes their union in the single night of love represented in the romance—or rather alluded to by the narrator's discretion (w. 4676-86). Just as that union was brought about in part through the lying rumors of death that preceded Lancelot's return to Bath, just so Guenevere has used her message au noauz to create a lying picture of Lancelot's prowess that fools the crowd, while at the same time, paradoxically, it exhibits (for a selected audience) the truth of Lancelot's identity as knight and lover. The Queen's enjoyment of her own joke—at the expense of the damsels who each vow to marry only the Red Knight—shows that her comic impulse is still active, but has been rechanneled with happier results, at least for herself and Lancelot (w. 6007-14). By using her command over Lancelot, the Queen has created a fiction—the paradox of Lancelot's performance an noauz and au mialz— that secretly corresponds to the truth of Lancelot's identity and thus reveals the mutuality of their love. Their equality in love, as perceived by the Queen, is based on that total gift of self: each one belongs entirely to the other (w. 5874-75, quoted above). But elsewhere the narrator has judged Lancelot superior to the Queen, based on the relative quantity of love that each has to give: if she loves him greatly, Lancelot loves her a hundred thousand times more (w. 4662-68). The great size of Lancelot's heart and its enlarged capacity for love will be beautifully elaborated in the Prose Lancelot, but already in the Charrete it signals in yet another way Lancelot's uniqueness. His total, unhesitating subordination to Love and

Chretien's Chevalier de Id Charrete

83

the Queen results in his heightened capacity to achieve what no other character in the Arthurian world can do. Lancelot's mobility as a character who moves across and beyond normal limits contrasts markedly with the limitations we have seen placed on the Queen's role in the distribution of private desire and public function. Whereas Lancelot's heart opens up new realms for him in the mixing of private and public roles, the Queen must carefully separate the two and give priority to the latter.28 The relative size of the lovers' hearts introduces again, after a long detour, the question of measurement raised in the opening paragraphs of this chapter. Measurement and evaluation seem to operate as much and as problematically in the domain of love as they do in the domain of prowess. The analysis of the tournament episode as mise en abyme, carried out thus far within the dimensions of the romance world, has suggested to what extent the question of measurement is tied to the manipulations the characters perform on each other, for good or for bad, intentionally and unintentionally. Meleagant's imprisonment of Lancelot appears in this light as both an admission of inferiority and a desire to demonstrate his superiority over Lancelot. The increased security of Lancelot's second imprisonment in the tower (the direct effect of the tournament episode in terms of plot) is a measure of Meleagant's real, but denied fear. Meleagant's selfdeception—represented at the tournament by the King of Ireland's son— never fools anyone, not even his own father, who repeatedly tries to save his son from fatal misjudgment. As a character, Meleagant is easily judged and classified. The measurement of Lancelot's value remains, on the contrary, a matter of dispute not only for the inhabitants of romance but for a long line of critics and scholars who have argued for and against Lancelot as hero, emasculated or inspired by love.29 The question of outside judgment recalls my earlier suggestion about our perspective on the tournament: since we already know what the participants in the story are trying to discover, the enigmatic character of the herald's cry implies for us a different process of discovery, informed by that of the characters but operating on a more general level. In order to act upon that implicit invitation, we need to understand more fully how the author/narrator has himself manipulated the story and our perceptions of it. We may operate at a higher level of recognition and judgment than most of the characters in the Charrete (including the narrator), but we have been shown over and over again that we are engaged in the same process of interpretation and evaluation as the characters themselves. We, no less than they, must

84

Chapter 3

sift through the evidence, formulate and verify hypotheses, as we try to figure out what is really going on in this romance and what it may ultimately mean.

Authors, Story, Public, Patroness The connection of author to story and public immediately introduces the question of Chretien's relationship to Marie de Champagne and Godefroi de Leigni, the patroness who gave him "matiere et san" (v. 26: story matter and meaning) and the "clerk" to whom he passed on his story for completion. Both of these relationships have posed problems for scholars trying to understand what exactly Marie gave Chretien and why he did not complete the romance himself. In the absence of any real historical evidence to furnish an explanation, there has been a good deal of storymaking about Chretien and Marie that more often than not casts Marie as Queen and Chretien as Lancelot in the tournament episode: capricious lady manipulates obedient servant. This scenario is used to suggest that Chretien, an apologist for love in marriage a la Erec and TCvain^ reluctantly takes on the mission to write a romance of adulterous love, the san extrapolated from Andreas Capellanus's own literary representation of Marie de Champagne in De Amore. Godefroi's intervention is then seen as confirmation of Chretien's distaste or as a sign of his failure to imagine an ending for the romance as commissioned.30 To a certain extent Chretien himself invites these fictionalizations of his role vis-a-vis Marie and Godefroi. As noted by any number of readers, he sets up explicit parallels between the relationship of author and patroness as represented in the Prologue and Lancelot's obedient service to the Queen's commands as demonstrated at the tournament. Not only does the narrator have Lancelot repeat to the Queen his own description of complete subordination to Marie ("come cil qui est suens antiers," w. 4 and 5656: as one who is completely hers), in each case the lady supplies a fiction, story, and meaning that will be acted or written out by the recipient.31 If, as I indicated earlier, Lancelot is a character particularly appreciated by and attractive to women, it is entirely fitting for a lady to be identified as Chretien's literary patron for the Charrete, however we may understand her role as the source of his "matiere et san" (cf. Philippe de Flandres as patron for the Conte du gmcd]. Just as Chretien uses echoes and analogues within the romance to

Chretien's Chevalier de la Charrete

85

generate meaning through the interplay of repetition and variation, so he extends the pattern to operate in the world outside romance, as projected in the doubling of relationships between himself and Marie, himself and Godefroi. The diegetic and extradiegetic worlds mirror each other. Chretien as author/narrator "repeats" himself through duplication and the displacement of his contract with Marie. With the introduction of Godefroi, Chretien sets up a second set of relationships between author, public, and story, that echoes his own triangle with patroness and story. Whatever personal motivation may lie behind that transfer, the effect for us as readers of the Charrete is to make us wonder about and interpret the role of authors and patroness, based on the "story" partially and enigmatically recorded in Prologue and Epilogue, just as we are fascinated and puzzled by what Lancelot and the Queen are up to within the story.32 If the meaning of Lancelot's fiction is unclear for the crowd of spectators confused by contradictions, the san of Chretien's romance is no less ambiguous for his public and will not be captured or determined by any simple reference to the historical figures named. Their historicity in the romance exhibits the same problematic status as Arthur's in the context of twelfth-century writing. As Wace comments in his Brut, the stories about Arthur and his knights are a mixture of truth and lies neither wholly believable, nor wholly unreliable: En cele grant pes que je di Ne sai se vos 1'avez oi', Furent les mervoilles provees Et les avantures trovees Qui d'Artur sont tant racontees Que a fable sont atornees: Ne tot man^onge, ne tot voir, Ne tot folor ne tot savoir. Tant ont li conteor conte Et li fableor tant fable Pour lor contes anbeleter, Que tot ont fet fables sanbler.33 (During that great period of peace I'm speaking of, I don't know if you've heard about it, were the marvels proved and the adventures found which are told so much about Arthur that they have been turned into fables, neither complete lies nor completely true, neither

86

Chapter 3 complete foolishness nor complete wisdom. Storytellers have told so much and fablers fabled so much to make their stories more attractive that they have made them all seem made up.)

Medieval historiography already identifies Arthur as an historical figure about whom legends have accumulated to such a degree that even today we find it difficult to disentangle fact and fiction. In the context of romance Arthur has become as much a marker of fiction as of history. Indeed, in Chretien's work we might argue that his fictional status outweighs the historical dimension.34 Fiction tends to pull history into its own domain. This is precisely what Chretien does to himself and to Marie and Godefroi. Each name is an oblique historical marker, a reference to real people who lived at a specific moment in history. But outside the romance I can learn nothing about them upon which to anchor any hypothesis regarding the exact nature of their transactions. Among the three, only Marie de Champagne can be verified with respect to non-literary documents of the period (which tell me nothing about her involvement with Chretien or the story of Lancelot). Marie herself has been drawn into the game of fiction, not only in Chretien's text, but in that of Andreas Capellanus, whose DeAmore enjoys as problematic a relationship with Chretien and other literary sources as it does with the verifiable and documented history of Marie's court. Whether I am convinced or not of Chretien's reliability in giving me information about his relationship to Marie and Godefroi, I can only understand that triangle in terms of the fictional context Chretien has created in the Charrete. Given the unavoidable loop that ties together in a common story pattern both the characters of the romance and its producers, we can only deal with Chretien, Godefroi, and Marie as figures in the game of fiction, the implied author/narrator(s) and their "narrataire."35 If we allow ourselves to be guided by Lancelot's example as analyzed in the tournament episode, we may be able to see that, whatever Marie de Champagne actually gave Chretien, once the gift has been given, he is no more the hapless victim of a woman's manipulation than Lancelot himself, whose own pattern of fiction-making supplies a model for the Queen's scenario. As indicated in the Prologue itself, Chretien's rhetorical play on the contradictions and harmonizations of truth and hyperbole furnishes ample proof of his own manipulative skills. In general, his capacity and taste for irony—say one thing and mean another, say and not say at the same time—should make us wary of (mis)interpreting his words. Moreover, the role of Godefroi de Leigni makes an ironic comment, in retro-

Chretien's Chevalier de In Charrete

87

spect, on Chretien's submission to Marie's command. As represented in the Epilogue, Godefroi's relationship to Chretien and the story duplicates that of first author and patroness. Now it is Chretien himself (in the role of Marie) who commissions the clerk/author and supplies the story: Godefroi de Leigni, li clers, a parfinee LA CHARRETE; mes nus horn blasme ne Fan mete se sor Cresti'en a ovre, car c/a il fet par le boen gre Cresti'en, qui le coman^a. (w. 7102-7) (The clerk Godefroi de Leigni / Has put the final touches on the Knight of the Cart; / Let no one blame him for completing the work of Chretien, / Since he did it with the approval of Chretien, who began it.) This pattern of turning over his work to others appears in anticipation and with particular clarity in the tournament episode, in the way the narrator allows so much of the storytelling to pass through the comments and perceptions of the characters. But while the narrator may seem to efface himself, Chretien as author remains fully in control, as he blocks out the action and our view of it—not unlike a screen director who controls our vision through the movement of camera, the play of actors and locations, the editing and montage. The analysis of the tournament episode as mise en abyme confirms to what extent Chretien controls his story before passing it on. By mapping out the pattern of relationships between Lancelot and the Queen, Lancelot and Meleagant, Lancelot and the court, the romancer offers a model for closure, insofar as the tournament summarizes all that precedes and anticipates all that follows. The kind of closure implied in the tournament episode is, however, a program for open-endedness at the levels of story and meaning—and even at the level of authorial control, if we play out further the implications of the tournament as mise en abyme and the role Godefroi de Leigni attributes to himself. Like Marie de Champagne, named but silent in the romance Prologue, Chretien as author and narrator remains silent about the transfer of commission and story to another. In the Epilogue he has become (again like Marie), a name cited to authorize and guarantee a fiction. We have to take Godefroi's own words as proof for the relay of authors. Chretien's use

88

Chapter 3

of characters to tell his story may echo in advance the role of Godefroi, but a fundamental difference remains to be explored between authorial discretion and the duplication of authors revealed in the Epilogue. Can we be sure that Chretien even wrote the tournament episode himself:1 Where did Godefroi's part begin? There is no obvious signal given in the linear development of the romance. Statistical analysis of the narrator's interventions shows a marked increase in Godefroi de Leigni's part, but we are unlikely to have noticed that change in a first reading, certainly not in an oral reception of the romance read out loud in installments.36 Even most scholarly rereadings of the romance are more likely to stress the continuities of style, rather that any discontinuities. We are forced to rely for the most part on Godefroi's own retrospective location of his beginning—but that location is itself ambiguous: tant en a fet des lors an ^a ou Lanceloz fu anmurez, tant con li contes est durez. Tant en a fet, n'i vialt plus metre ne moins, por le conte mal metre, (w. 7108—12) (He worked on the story from the point / At which Lancelot was walled in / And finished it. / This much only has he done. He wishes to add nothing further, / Nor omit anything, for this would harm the story, v. 7109 slightly modified) Godefroi seems to be concerned with exactness ("ne . . . plus . . . ne moins"), yet the identification of the episode where Lancelot is "anmurez" is uncertain. It seems to designate the second imprisonment when Lancelot is actually walled into the tower, but at least one scholar has argued for reading "anmurez" as a general term that would refer to Lancelot's first imprisonment.37 While most of us continue to locate the beginning of Godefroi's part after the tournament episode, further uncertainties about the tower and its location (geographical and diegetical) complicate the puzzle of Godefroi's role. Is the tower on an island or simply next to the sea:1 The first description (w. 6117—26) does not seem to correspond to the second, when Meleagant's sister approaches the tower (w. 643941). Is it by accident that Train's third allusion to the Charrete^ this time by the narrator himself, refers to a tor that apparently designates Lancelot's second imprisonment in its use of the specific term "tower," yet the

Chretien's Chevalier de la Ckarrete

89

temporal indication locates it during Lancelot's first imprisonment with reference to the Queen's return to court three days earlier (Tvain, w. 4734-69)? This apparent inconsistency (cf. the false rumors circulating in the Charrete) has led a number of scholars to postulate the simultaneous composition of Tvain and the Charrete: the inaccuracy of the Tvain allusion could reflect the unfinished state of the Lancelot; the unnoticed movement of the tower from island to seaside might occur in the transition from one author to the next.38 Given our irreducible ignorance of the particular circumstances surrounding Chretien's production of his romances, we are again forced by these puzzles to speculate about how the ill-fitting pieces go together. Even the explanation I just alluded to does not completely satisfy. Why, for example, should either Chretien or Godefroi, depending on who was writing when, change the location of the tower? Is it simply an unimportant detail, unlikely to be noticed when we read the romance out loud (cf. the shift between glove and stick in the Chanson de Roland or Beroul's dead felon who later comes back to life to be killed a second time) ? Or is the problematic location of the tower itself a figure for the problematic location of authorial production and meaning? The direction our speculations will take depends in large measure on the character we give the author. Once we credit Chretien with certain attributes—irony, cleverness, even craftiness, a taste for the enigmatic, a desire to exploit ambiguity—our explanation of the "facts," the inconsistencies that crystallize around the tower within the Charrete, as well as between Tvain and the Charrete—will tend to assume that their apparent incoherence, like the "loose ends" that are frequently left hanging in the plot (whispers, dwarfs, damsels, etc.), may be intentional and programmatic, part of the san Chretien is elaborating in his romance. Such a position strikes me as more convincing and more critically rewarding than one that sees Chretien as an incompetent storyteller; it places the "burden of proof" on us as readers. What then do I make of the uncertainties that surround the transition from Chretien's part to Godefroi's? The first result is an intensification of the problem concerning endings—a problem that becomes more and more complicated in Chretien's last three romances: if the achievement of closure is already problematized within the apparently closed dimensions of Tvain and the Charrete, we may consider the unfinished state of the Conte du Graal as the ultimate, suspended expression of that problem, as (re)invented in Chretien's corpus.39 Godefroi de Leigni's epilogue, while clearly marking the end of the

90

Chapter 3

Charrete^ opens the question of closure by revealing that the romance ends twice, at least in terms of authorial production. Yet we cannot exactly find the first ending where Chretien stopped writing and we may not be satisfied with the second by Godefroi, however much it agrees with Chretien's own plan, since we remain in the dark about what will happen to the lovers. A contrast appears between Chretien's and Godefroi's desire for the appearance of closure (located on the levels of rhetoric and a particular segment of narrative events) and a contradictory desire for openness (operating more obliquely in terms of narrative, romance production, and^w).40 The uncertainty that weighs over the future of the lovers (and the future of the Arthurian ideal) parallels the ambiguity we have seen operating with no less indeterminability in the relationship between the romance's authors and patroness. If we focus on the doubling of authors, we suddenly see the Charrete as a romance that contains a continuation, despite its relatively compact size.41 Although, on one level, the Charrete continues to operate as a single, independent romance, whose plot is unified by the Lancelot-Meleagant quarrel, on another, we are invited to read the conclusion, from the end of the tournament to the brief description of the entire court's joy at the defeat of Meleagant, as Godefroi's continuation of a story begun by Chretien.42 Once we learn that a second author has completed Chretien's own task, the unanswered and unanswerable questions that emerge from such an unexpected tmnslatio carry the interplay between closure and open-endedness from romance plot into romance production. The worlds of fiction and "reality" mirror each other in a pattern of infinite regression: if we would understand the possible meanings they bring into play, there is no place to go but further into fiction and the play of textuality.

Tvain and the Charrete: Patterns of Multiplying Stories As we focus on the Charretis single and double stories of authorship or on the doubling of story and continuation, we may be reminded of a different kind of doubling Chretien introduces, when he particularly aligns the Lancelot with another of his romances through Yvairfs explicit allusions to the events of the Queen's abduction. As we saw above, these allusions not only connect certain plot elements in both romances but also involve the authorial question as it is tied to the problematical tower/im-

Chretien's Chevalier de la Charrete

91

prisonment. Both romances introduce a second narrator who takes over Chretien's role: Calogrenant within the fiction, who initiates Yvain's adventure by a tale of his own failure, Godefroi de Leigni outside the fictional world (and yet known to us only within it), who ends Lancelot's adventure with his triumph over Meleagant. Although the direct allusions of Train seem to give the Charrete a certain chronological precedence in the order of composition, other implicit allusions within the Charrete suggest just the opposite.43 We can only appreciate certain ramifications of Keu's and Arthur's behavior in the opening scene, for example, if we already know how they acted in TCvain. Whatever the conclusions to be drawn about the actual process of composition, the result of these textual echoes is to create a pattern of interlace between the events of both romances that simultaneously establishes and undermines any linear, chronological ordering of their episodes. Although the main events of Lancelot's and Gauvain's quests are clearly embedded in the middle of Yvain's plot, the opening events of both romances set up an intertextual dialogue that is non-mimetic, nonlinear. The tower with its uncertain location in time and space not only maintains a link between the romances but also introduces gaps in the way they fit together. The two romances clearly stand in a special relationship to one another, yet they cannot be neatly matched and ordered.44 This explicit interplay between Tvain and the Charrete appears exceptional within Chretien's corpus, and yet we can nevertheless see it alluding to a typical pattern in which the romancer repeatedly experiments with combinations of stories. The title of his first romance is significantly based on the coordination of two names, as he indicates in the Prologue to Cliges: "d'Erec et d'Enide" (v. i). The stories of husband and wife are conjoined and both must set off on adventure, as we wonder whether Erec or Enide is to blame for a fault of recreantise or untimely speech. The title of his second romance gives pride of place to Cliges's story, yet the romance itself neatly divides into two generations, with the story of Alexandre and Soredamors followed by that of Cliges and Fenice. Chretien is clearly following here the biographical model offered by the Tristan story (such as we can still see it in Gottfried, for example), where the love story of the parents furnishes the context for that of the son. Throughout the middle three romances of Chretien's corpus the story of Tristan and Iseut, like an eminence grise, peers over the shoulders of his lovers, combining or contrasting with their stories more or less explicitly, depending on the romance. While Cliges plays the most obviously and insistently on the Tristan story, the Charrete more profoundly accepts and

92

Chapter 3

rewrites the love triangle through a process of displacement that in Tvain continues to move away from the Tristan story and yet still makes reference to it through the hero's madness, his life in the forest, and so on.45 The special ties already described between the Charrete and Tvain are particularly interesting in relation to their mutual play on the Tristan model. In some sense, we can interpret their intertextual links as part of Chretien's strategy to disengage Lancelot's story from the tragic fascination of Tristan's. Buttressed by Train and the kind of space it maintains around the Charrete, the latter is securely integrated into an Arthurian "cycle" of stories. The most explicit link between Tvain and the Charrete is furnished by Gauvain, whose many adventures on the way to the Water Bridge in quest of the Queen are alluded to by both romances—in Tvain as the reason for his inability to help Lunete and his own family, in the Charrete as the reason for his rusty armor (w. 5117—21). The embryonic interlacing suggested in the Charrete by having two heroes on their way to rescue the Queen remains unrealized; it is replaced rather by the intertextual entwining of Yvain's and Lancelot's adventures. Since neither romance actually tells us what happens to Gauvain on his long and perilous route, his adventures constitute an open and uncharted space through which the Charrete and Tvain are interwoven. Le Conte du Graal will move into that potential space, by placing Gauvain's adventures in the main narrative, interlacing them with Perceval's—the last combination of stories we have from Chretien, a puzzling, but powerful one for the production of further narratives, as continuators seek to complete Chretien's unfinished story. As we reflect on this pattern of variously combining stories, we may distinguish a dual tendency in Chretien to control or contain stories and at the same time allow them to continue and change in combination with others. Placed together, the epilogues of the Charrete and Tvain show concern about the problem of authorized and unauthorized continuations.46 Seignor, se j'avant an disoie ce seroit oltre la matire, por ce au definer m'atire: ci faut li romanz an travers. (w. 7098—101) (My lords, if I were to tell more / 1 would be going beyond my matter. / Therefore I draw to a close: / The romance ends here.)

Chretien's Chevalier de la Charrete

93

Godefroi thus opens his epilogue with an insistence (repeated in w. 7111-12) on his exhaustion of the story matter. His comments are oriented back toward what we have just read, in order to justify his continuation and ending within the dimensions staked out by Chretien. Chretien's own epilogue in Train is shorter, but he too insists on the end of his matter as a limit: Del Chevalier au Lyeon fine Crestiens son romans ensi, n'onques plus conter n'en oi ne ja plus n'en orroiz conter s'an n'i vialt mangonge ajoster. (w. 6804-8) (Chretien thus ends his romance; he never heard anyone tell more about it and you will never hear any more told, if one doesn't want to add any lies to it.) While Chretien's name and authority guarantee the limits of Godefroi's work, Chretien himself claims to use the story heard to set up a boundary for Tvain: anything beyond the limit thus established constitutes an addition and, worse, a lie. (Of course, we can only measure this boundary with respect to Chretien's own written text.) Both epilogues want to convey a strong sense of inside closure in relation to an outside world of other stories; both claim or imply an inner consistency and an outer difference, which Train's closing remark identifies as a contrast between truth and lies.47 Chretien's own reference to the oral context of storytelling (cf. Eretfs Prologue) suggests how well aware he was (and we should be) that many people tell (and write) stories, even the same stories, in different ways. All medieval storytelling and writing participates in tradition and change— mouvance, as Paul Zumthor has dubbed it.48 Like many other medieval writers, Chretien shows his awareness of mouvance by simultaneously refusing and acknowledging it as the process in which he himself operates. Across time and space stories exist independently of authors; they grow and change, continue in conjunction with other stories, other storytellers—or even the same storyteller returning again and again to the same one, as Chretien does with the underlying pattern of Tristan and Iseut. Like Marie de France and Thomas d'Angleterre, other proud "Moderns" of the twelfth century, Chretien is anxious to set his own stamp on the traditions he inherits and passes on, as the pun on his own name in the

94

Chapter 3

Prologue to Erec testifies (w. 23-26). The contrasting humility he shows in the Charrete Prologue is more to be understood as part of Chretien's strategy to praise Marie de Champagne than as a real expression of authorial self-effacement. He does not want us to forget the author whose "painne et. . . antancion" ("effort and . . . diligence," v. 29) has translated "matiere et san" into the romance we still read eight centuries later. But Chretien necessarily situates himself in the context of tradition and mouvance. The dual authorship of the Charrete reflects (on) Chretien's play with the mouvance of tradition by problematizing in the romance both the way medieval stories are shared or passed along from one author to the next and the way these stories are combined and reinvented to make new ones. And if there is one story that is shared by numerous authors and storytellers in the twelfth century, it is the story of Tristan and Iseut. In some sense, we might say that the dual authors of the Charrete are really Chretien and the Tristan legend, in whatever form or forms Chretien knew it. Although their names are never mentioned explicitly, we cannot read the Charrete without being aware of and exploring the ways in which Chretien has displaced, taken apart, and reinvented the story of Tristan and Iseut—and King Mark. As deeply and intensely as the Queen lies embedded in the heart of Lancelot, the problems and issues raised by the Tristan story lie at the heart of Chretien's rewriting of the lovers' passion and their relation to king and courtly society.

Tristan, Lancelot, and the Measure of an Arthurian Ideal Given the degree to which Chretien seems to (re)invent Lancelot in order to conjure the subversive myth of Tristan and the fatal disorders sown by his love for Iseut, it is not surprising that we need to measure the play of displacement and repetition between the two stories to understand some of the implied meaning of the Charrete*9 The Charrete^ own model suggests that the byways of fictionality are essential to any quest for meaning. Although Chretien makes no explicit reference to Tristan and Iseut in telling Lancelot's story—unlike the author/narrator and Fenice in Cliges, for example—the unmistakable allusions to the Tristan material are surely part of Chretien's strategy to highlight the fictionality of his romance.50 The Charrete is a story that plays constantly against other stories: as another version among many competing versions of the Tristan legend, it is doubly fictional, the story of a story.

Chretien's Chevalier de la Charrete

95

In characterizing Chretien's general attitude toward the Tristan story, I would suggest that his position in the Charrete is hopeful within the limits of human potential. Chretien seeks to avert the tragic potential of the Tristan story; his bent is comic, in both senses of the word. But he is cognizant—and makes sure that we are, as well—of the complexities, the paradoxes, and the contradictions of the human condition that complicate the translation of an Arthurian secular ideal from system to praxis, within the fictional realm itself, and even more so when we pass from fiction back to what we are pleased to call reality. Chretien's Lancelot does not simply borrow the Tristan model of the love triangle and replay it with different characters; his romance is a powerful recasting and reimagining (in the medieval and modern senses of that word) of the relationships that tie together the principal characters, and the social and psychological forces brought into conflict through their entanglements. In order to measure the changes wrought by Chretien, let me summarize briefly the most salient features of the Tristan legend brought into play (regardless of any particular version at this point). First and foremost, the love between Tristan and Iseut is a factor of disorder vis-a-vis feudal society; it remains throughout their story irreconciliable with and distinct from the other values reflected in the romance world. The doubling of family and feudal relationships in the triangle—linking husband, wife, and nephew, King, Queen, and vassal—both establishes and reinforces this antagonism between private and public, love and society. Tristan's triumphs as Mark's best knight are unconnected to his love for Iseut. The most spectacular example of his prowess, his defeat of the Morholt, predates it, and Tristan's service to Mark, his service on behalf of courtly society in general, is interrupted once he falls in love with Iseut.51 In the Charrete, there is total synchrony (at least until the night of love) between Lancelot's service on behalf of the Queen and his role as liberator for the citizens of Logres held captive in Gorre. As suggested earlier, his love for the Queen generates an excess that benefits the entire Arthurian society. Love and prowess are not simply coodinated in terms of individual motivation for combat; love is shown to be the force that leads Lancelot to achieve his greatest triumphs, without which Arthur would have lost his Queen and left his subjects undefended. This shift introduces a profound difference in the Tristan model, which cannot be overlooked in any general evaluation of Chretien's rewriting of the love triangle. Lancelot's love and his heroism are totally integrated, even though at times they may appear (to the characters or the romance public)

96

Chapter 3

to be opposed. Lancelot's two-step hesitation before the cart, his two reputations that clash when they meet, these (and many other elements in the romance) still recall the Tristan-type love that is "opposed" to chivalric service and courtly society. They remind us further that Lancelot's love for Guenevere is not unproblematic even if it serves noble ends, especially once that love is shown to be mutual and consummated by the lovers (a Tristan-type love is inconceivable without physical passion, however spiritualized their union may be). We have already seen how the tournament dramatizes the link between love and prowess, now (apparently) disrupting it, now reinforcing it, connecting it further in equally problematic ways with the issue of marriage. Here is the crux of the problem, which Chretien's romance both displaces and relocates through displacement: the role of adultery as treason to the King. Chretien has eliminated one particularly nasty aspect of the Tristan story, the incestuous links of the "family romance."52 The nephew who carries off his uncle's (or father's?) wife still reverberates in the background of Arthurian history, in the later actions of Mordred and the destruction of Arthur's kingdom. This may indeed be the link which allowed Chretien to conflate two of the greatest tales circulating in the twelfth century. But it also offers the model for an ending that Chretien's rewriting desires to change, or at the very least, to put off for as long as possible. In this respect, Lancelot's connection with Arthur is simplified, although it is not at all simple. His link with the King passes not through family ties already in place before a woman enters the scene, but exclusively through his love and service for the Queen. In the Charrete we know nothing of Lancelot's role at Arthur's court before his arrival there in the very last scene of the romance (cf. the marginalization of Lancelot that I spoke of earlier). Lancelot does, however, identify himself to the monk at the cemetery as a knight born in Logres (w. 1929-30). He thus qualifies to participate in the custom of Logres, which sets up rules for correctly or incorrectly appropriating damsels circulating in the realm. I have elsewhere analyzed in detail how this custom, operating in Arthur's own kingdom, furnishes implicit (if limited) justification for Lancelot's night of love with the Queen, inasmuch as she has been won fairly and squarely through combat with her current defender, Meleagant.53 Without reiterating all the stages of that argument, I would emphasize here the complexity of the romance's multiple views of Lancelot, which allow us to see him as Arthur's best knight, not in spite of his role as the Queen's lover, but rather precisely

Chretien's Chevalier de la Charrete

97

because of his passion for Guenevere. While neither role cancels out the other—and thus reflects all the better the messiness of real life experience, despite the fictional quality of Lancelot's resolution of paradox—the paradoxical and unsettling combination leads to our perception of Lancelot's further role as touchstone. He is the one who reveals a whole range of conflicts and problems that characterize the Arthurian world: competing courtly and chivalric values, feudal and religious institutions, established customs, family ties, the proper authority of the King and the proper interpretation of signs. If, in some sense, Chretien has shown that the Tristanian model of love has the same potential for coordination with prowess and courtly service as the kind of love explored and finally integrated into marriage in Erec or Tvain, the introduction of an adulterous triangle has nevertheless forced out into the open other issues that should also command the public's attention, if we are to face without blinking the contradictions of an Arthurian ideal cast in terms of twelfth-century feudal society. In order to follow this development more closely, we need to consider now in some detail how Chretien reinvents the Tristan model in the specifics of the Charrete. We cannot determine with any certainty what version or versions Chretien knew. He may allude to Beroul's Tristan (or one like it) in the tournament episode, by recalling Tristan's appearance incognito as the Black Knight who takes vengeance on his enemies during the tournament at Blanche Lande.54 Chretien's use of two other Tristan elements—the bloody sheets and the ambiguous oath—aligns the Charrete here with a Beroul-type romance, there with a version like Thomas's, at least if we can rely on Gottfried's translation as an accurate representation of the way Thomas situated the blood drops and tied that scene to the ambiguous oath. In Beroul's version, there is no ambiguity about the lovers' guilt, as indicated by the location of the blood drops both on Iseut's sheets and the floor between the lover's beds (w. 747—70). When the King sees Tristan's bleeding leg, he declares, "Trop par a ci veraie enseigne: I Provez estes!" (w. 778—79, emphasis added: "Here is surely a true sign. You are proven guilty!"). While Meleagant seems to borrow King Mark's words (or is it the other way around?), the true signs ("ansaignes bien veraies," v. 4774) in the Charrete are at least as ambiguous and rather more misleading than the blood drops in Thomas/Gottfried, which appear only in the lovers' beds and not on the floor between them.55 Indeed, it is this ambiguity that causes Mark to waver still about the lovers' guilt and demand that Iseut publicly demonstrate her innocence by the ordeal of the red-hot iron. Al-

98

Chapter 3

though Chretien maintains this ambiguity and the demand for public innocence, he has not completely abandoned some of BerouPs most spectacular effects: the order of events reflects that of Thomas/Gottfried, but leaves out the ordeal. For obvious reasons, the ambiguous oath suits Chretien's purposes much better, as he plays on the uncertainties that connect truth and lies. His transformation of bloody sheets and ambiguous oath makes significant use of displacements, in location and personnel, from moral problem to epistemological puzzle. The initial change involves a shift from the husband's home and court to another's: while Mark was at once husband, king, and accuser in the Tristan story, these roles will be splintered and duplicated, multiplied and transformed in the Charrete, in such a way that the inadequacies of accuser, husband, father, and king may ultimately appear more problematic than the lovers' guilt.56 At the very least, the problem of adultery and the issue of loyalty will be shown as part of a much larger set of conflicts which cannot all be attributed to the fault of Lancelot and Guenevere. As a king and husband, Arthur is left far behind during most of the Charrete. This is itself a significant problem, as indicated in the earlier discussion of role substitutions, and is still an important factor here in the replay of the bloody sheets and ambiguous oath episode. Bademagu provides the place, but is unrelated to the Queen—though he is the father of the accuser, a relationship that will play a crucial role in the development of the accusation and the judicial combat arranged to verify it. Just as Tristan did, Lancelot bleeds on the Queen's sheets, but the two beds in the room with bloody sheets are hers and Keu's. Keu, too, has a wound that bleeds—thanks to Meleagant, who keeps Bademagu's good medicine from curing the wound he himself inflicted. Keu becomes the obvious target of Meleagant's accusations: Arthur entrusted the Queen to Keu because he had such faith in him (w. 4854-57), but now every night he treacherously visits the Queen's bed (w. 4810-17). Meleagant misrepresents the opening scene, as much as he does the present one; his exaggerated fable spares the lovers the necessity of inventing a scene that would cover their private truth with an acceptably public one (as we see in the Mai Pas episode of Beroul, for example). In fact, the Queen has gone to sleep thinking her sheets "molt blanc / et molt bel et molt avenant" ("pure white / And fair and proper," w. 4742-43).57 Later she truly believes ("Et ele cuide dire voir," v. 4784) that her nose has bled and stained her sheets, since she is unaware of Lancelot's wounded finger. The narrator seems to emphasize the Queen's own sense of innocence

Chretien's Chevalier de la Charrete

99

in this scene, while at the same time showing her alertness (cf. BerouPs Iseut), when she secretly sends for Lancelot and announces to King Bademagu that she will have a champion to defend her, if Meleagant dares to accuse her (w. 4901-6). The truth concerning Lancelot and the Queen's love affair is not so much left out of this scene, as distanced and problematized, displaced by Meleagant's false accusation against the Queen and Keu. Meleagant may remind us here of Beroul's felons: as both counterpoint and contrast, they present a model of accusation that is at the same time truthful and lying—truthful concerning certain facts, but not according to different value systems, lying and treacherous in terms of their own motivation in envy and feudal power plays. We may be reminded at this point of yet another displacement in the Charrete'. Lancelot is not under accusation here, as Tristan was in similar circumstances. But his entire quest for the Queen has been marked by a series of accusations against the Cartrider—true insofar as Lancelot did indeed ride in the cart, false insofar as Lancelot is not guilty of any of the crimes the cart ride presupposes, whether those connected with capital or non-capital offenses, both types paradoxically combined in Chretien's description of the cart custom.58 Both lovers are accused in the Charrete^ but they are accused separately, and each of the accusations appears to be true, as well as false. Such displacements seem to be the Charrete's key maneuver for uncovering the nexus of issues at stake. The oath that Lancelot makes is not ambiguous in the sense that Iseut's is in Beroul. Her oath is true, but apparently refers to a scene just witnessed by the crowd of spectators at Mai Pas, while in reality—from the lovers' and the romance public's point of view—it refers as well to the love affair between her and Tristan. Its truth is double, but appears unequivocal. Lancelot's oath in relation to the Queen and Meleagant's accusation is entirely true (though we know what has been left out in Meleagant's reading of "enseignes bien veraies"). What will prove to be ambiguous is the second half, where Lancelot swears to show no mercy, if he manages to defeat Meleagant (w. 4974-84). This oath and the dismay it causes Bademagu (w. 4985-86) immediately precede the judicial combat between Lancelot and Meleagant, presided over by the King. This is the escondit (defense, exculpation) that Beroul's Tristan constantly claims, but never obtains, since all are convinced that his prowess alone guarantees him victory (e.g., Beroul, w. 155-56, 799-818). If Lancelot is thus granted the opportunity to let God show that Meleagant has lied, the true demonstration ("voire demostrance," v. 4976) he swore is interrupted by

TOO

Chapter 3

Bademagu's fear that Lancelot will successfully execute his oath of no mercy on Meleagant. In a replay of the interrupted conclusion to Meleagant and Lancelot's first combat, Bademagu asks the Queen to stop the battle, and Lancelot does so, as soon as he hears the Queen's agreement. But while Lancelot was clearly the winner in that earlier contest, as indicated in the narrator's description and everyone's reaction except Meleagant's, the interruption comes so quickly here that there is no clear advantage on one side or the other (w. 5004-10). Of course, we expect Lancelot to triumph, as much as Bademagu does, but the description of the combat up to this point emphasizes the equal balance between the two combatants. It is Bademagu's fatherly love that makes him intervene so quickly that we and Lancelot are deprived of a "voire demostrance" of Meleagant's lie; it is Lancelot's love for the Queen that makes him agree to suspend execution of his second oath. Does his oath then become a lie? In any case, the proof of who is lying is thus deferred and added to the stakes of Lancelot and Meleagant's last combat—by which time we have learned a lot more about the depths to which Meleagant's treachery leads. When Lancelot renews his promise to show no mercy (w. 7008-9), we know how richly Meleagant deserves his fate, truly executed by Lancelot. Do we at the same time forget (at least momentarily in the joy of the moment) Lancelot's own breach of loyalty to the King? If our attention has been temporarily deflected from the guilt of the lovers, where and why has it been redirected? Chretien has displaced the accusation of adultery first by placing it in the mouth of a treacherous liar. He has made it another example of the kind of misinterpretation that repeatedly arises around the actions of Lancelot and the Queen. Where Tristan and Iseut are obliged to invent a scene to cover the private truth of their love (and thus prove its innocence through fiction?), Meleagant assumes the creative role in this scene. His unintentional lie moves away from the moral problem raised (implicitly rather than explicitly in Chretien's romance) by the lovers' consummation of their love and substitutes an epistemological one: how can true signs not indicate the truth? They did at one point in Beroul, even if Mark's love for nephew and Queen later gives them the opportunity to produce yet more signs that multiply and cover that truth. In Chretien's transformation, the moral problem of the lovers is simultaneously displaced and combined with the problem of signs and their contradictory meanings. From the reader's point of view, we can see how even false and misleading interpretations can nevertheless lead to true problems posed by and in the text.

Chretien's Chevalier de la Charrete 101 A second displacement triggered in the accusation operates through the choice of language. Adultery itself is not mentioned, since Meleagant speaks rather of Keu's treason against his lord, King Arthur: "Le roi Artus a Kex trai' / son segnor" (w. 4854-55). The vocabulary here activates specifically feudal values, rather than religious concerns or views (cf. the way Chretien has eliminated the stigma of incest in Lancelot and Guenevere's relationship). Instead of focusing exclusively on the couple formed by the lovers, Chretien's version highlights the interaction between Meleagant and his father, which ultimately stands as an analogue for the proper behavior ofsejjnor or king with his vassal/subjects.59 Bademagu's fatherly love is repeatedly shown to be a dangerous weakness that allows his son's treachery to go on unchecked. If, as Keu tells Lancelot, Bademagu has kept Meleagant away from the Queen, except where the presence of others guarantees her safety (w. 4048-57), what was Meleagant doing so early in her bedchamber:1 Why has the King not prevented Meleagant from poisoning Keu's wounds? Meleagant's lie is here unintentional; later with Lancelot's imprisonment, his lies are fully prepared with malice aforethought, yet Bademagu is powerless in searching for Lancelot. There is a kind of interlace between Meleagant's perfidy and Bademagu's goodness that allows the romance to continue for thousands of verses after the first combat at Bath, where we may have initially expected the romance to end. Bademagu's fatherly love has its good side and its bad—an ambiguity that we usually attach to Lancelot's love for the Queen. Neither Bademagu's nor Arthur's kingship is unproblematic.60 In the Charrete the disorders in their kingdoms are not shown to be the result of an adulterous love. Meleagant's treachery acts as a catalyst, but his evil is aided and abetted by the good intentions of Arthur and Bademagu—even of Lancelot, too trusting to believe ill of a deceiving dwarf. The automatism of Rash Boons, the selfishness of Keu, the irrationality of long-standing customs, all these conspire to make trouble in the Arthurian world. But trouble is not simply the result of bad elements, it may arise as well from the competing demands of different value systems, each valid in its own domain. It is this paradox that is chiefly represented by the figure of Lancelot, who must satisfy the contradictory requirements of love and chivalric honor, fidelity to the Queen and fidelity to his word, his customary generosity for damsels and his equally customary mercy for defeated adversaries. Lancelot is shown again and again as the only one able to achieve the impossible by truly satisfying these opposed demands; he alone turns the impossibility topos upside down and repeatedly makes oppositions and contradictions meet. Lancelot is indeed the "one who will measure," but he will

102

Chapter 3

not do so in the standard way of Gauvain, the second-best-after-the-hero who conventionally measures other knights' achievements in Arthurian society. Lancelot figures as a touchstone whose contact reveals the positive and negative alloys of the society itself. Chretien's romance thus operates like the Case, one of Andre Jolles's "simple forms." In the Case, a situation arises in which we cannot simply measure the conduct of the individual; we must also measure the norm by which we usually judge his conduct. As an incident unfolds, a given law may be shown to produce good and bad results: the weight itself is not giving true weight. One must then judge between norms, evaluating each in turn by a superior one. But the Case does not conclude; it poses a question: where are the weight and norm necessary for that evaluation? While it imposes on us the obligation to decide, the Case does not contain the decision itself.61 The temptation to make a decision is very strong in the Charrete, where examples of interpretation and evaluation proliferate in the narrative, where we in the public are implicitly asked to interpret and evaluate Lancelot's behavior and its impact on Arthurian society. Sometimes Lancelot himself appears as the superior norm, the only one who succeeds to an extraordinary degree on the different scales of values, public and private. But if on these same scales he may be shown to be a traitor and an adulterer, there is possibly something faulty in the system of measuring the society's values. Chretien does not cut the Gordian knots posed within his romance. He neither rejects the Arthurian world and its chivalric values nor remains uncritical of the gaps that appear between an idealized system of values and the realization of those ideals in the praxis of romance adventures. Both within and outside the Charrete, the process of interpretation continues, since the text continues to supply more elements than can be worked into any of the systems it contains within itself.62 As undecided Case then, the Charrete postpones the tragic ending of Tristan and Iseut and substitutes the death of Meleagant, about which all (in the absence of Bademagu) can express unmitigated joy. Chretien achieves this "happy ending"—which is, in fact, more a potential for such an ending than a fully realized one—through his power to invent a fiction, to reimagine the Tristan story through the kinds of displacements and repetitions just analyzed, and through his ability to cut away the biographical and chronological development of the Tristan story—and of Arthur's history—in order to focus on the essential kernel of the lovers' story, the basic unit represented as well in the Folies Tristan and Marie de France's

Chretien's Chevalier de Id Charrete

103

Chievrefoil: separation, recognition, and (re)union.63 Such an organization may not be immediately apparent in the multiplicity of episodes that form the basis of the Charrete's narrative structure. Indeed, in some respects, their enchamement recalls the kind of organization we can see in Thomas/ Gottfried, where the story of Tristan's life up to the moment when he and Iseut drink the love potion can be represented as a linear development (like Lancelot's quest for the Queen), while the events following that crucial moment are more circular in organization (cf. the second half of the Charrete\ as the lovers are periodically joined and separated until the tragic ending in death.64 But if we stand back from the nitty gritty of the Charrete's specific episodes, we can perceive the overall organization in terms of the Tristan kernel, though it is not quite in the "simple" form we saw in the Folie Tristan d'Oxford. The initial separation of the lovers is clear, though we are not sure they are lovers: the Queen's whispered words seem to invite such speculation (in every manuscript but Guiot's) and suggest in retrospect that, while the spiritual character of their passion has already begun, this particular (re)union adds for the first time a physical dimension that is the sine qua non of a love modeled on Tristan and Iseut. After the cold reception and Lancelot's capture have added the necessary complication to help the lovers (especially the Queen) recognize the intensity of their passion, the night of love comes as the fulfillment of Lancelot's linear quest and drive for (re)union. But, of course, the romance does not end where the basic unit or lai ends: it continues by adding a series of separations and displaced (re)unions that gradually lead the lovers back to Arthur's court, from which their "original" union was removed. In this light, the tournament episode's mise en abyme recapitulates and redirects the lovers' recognition and their secret contact in the Queen's bedchamber, allowing it to operate at a distance within Arthur's realm. In some respects, the night of love is a unique episode, unrepeated in the Charrete and, therefore, problematic, since we cannot evaluate without repetition. Will the lovers continue the physical nature of their love? Their expectations for further rendezvous imply as much, but the romance itself gives us no indication about their future history.65 On the other hand, the tournament episode does, in some sense, repeat their secret union through its own variations, which also prepare the interplay in the final scene on the Queen's simultaneous absence and presence in greeting Lancelot at Arthur's court. The continuity of their love is projected beyond the limits of the romance, but the exact nature of its continuation remains open.

iO4

Chapter 3

The Charrete contains then not just a single instance of the basic unit, but one and a half: separation, recognition, (re)union; separation, recognition. In this aspect, as in so many others, Chretien's romance emphasizes the open-ended nature of its organization and meaning. But the addition of another "half" not only keeps open the lovers5 story; it also serves to integrate their story into the larger context of Arthurian history. While in the Folie Tristan d'Oxford, for example, the issues of evaluation and judgment are totally focused on the lovers and the character of their relationship, we have seen how the tournament episode and its recognitions engage not only the Queen but the entire crowd of spectators, and finally the romance public as well, in the process of evaluating the one who will give the measure, "celui qui 1'aunera." Moreover, Lancelot will give the measure not only for prowess, not only for love, but as the touchstone who reveals an entire range of conflicting values competing in the secular ideal of Arthurian society. The cropping Chretien has performed on the Tristan story to carve out Lancelot's in the Charrete also appears as the key maneuver of fiction vis-a-vis Arthurian history. We may know from Geoffrey of Monmouth or Wace about the end of Arthur's reign through the treachery of Mordred—a nephew trusted too much by a loving uncle (Bademagu and Meleagant?). We may know about Guenevere's cooperation, her vilenie (shamefully regretted afterward) in submitting to Mordred's secret love and marrying him against the law. But Chretien's romance "ignores" that ending, projects the possibility for a different kind of ending by rewriting the secret love that engages the Queen as an ennobling, even redemptive passion. Of course, Chretien does not write such an ending himself. The Charrete closes the story of Meleagant and Lancelot—in Godefroi de Leigni's "continuation"—and keeps open the stories of Lancelot, Guenevere, and the Arthurian court, for those unauthorized continuations, medieval and modern, in narrative and criticism, that continue to pursue the "matiere et san" given by Marie de Champagne.

The Fictionality of Romance We may legitmately wonder, along with medieval writers and thinkers who have expressed opinions on the subject, about how seriously we should take any romance and its claims to have something to do with truth and meaning. How seriously we take Chretien's Charrete is likely to de-

Chretien's Chevalier de la Charrete

105

pend on what we think happens to truth when it is mixed with lies—or fictionalized—"a fable . . . atornees," as Wace phrases it in his discussion of the marvels and adventures told about Arthur (Brut^ v. 1252). From the religious and didactic point of view of Chretien's contemporaries, romance and truth (or rather, Truth) are clearly and necessarily separated by the gulf of lying fictionality: romance is only serious in terms of the damage to one's soul for anyone seduced by its charms.66 Such antagonism may be clarified, if we consider the different roles played by repetition in the Middle Ages, especially the contrast in attitudes toward repetition that characterize theological discourse and what we would call literature or fiction.67 Neoplatonic thought, the dominant mode of medieval Christian theology, requires focus on the underlying sameness, the repetition of the Idea, the permanent, good, and true, unaffected by the meaningless play of variations, which belong rather to the realm of accidents, the concrete and changing. From such a point of view, fictional works like romances are either tolerated or condemned as empty space, a vacation of meaning; fiction is a lie precisely because it is emptied of redemptive discourse, a language turned away from its extralinguistic reference.68 But with the hindsight of centuries or from a point of view that assumes a position outside the mode of Neoplatonic thought, we can see that the play of repetition is neither meaningless nor empty, inasmuch as repetition and its pattern of variations are ultimately capable of introducing real differences. This is precisely the kind of play we have been following in the Charrete, through the repetitions and variations of the tournament episode as mise en abyme^ through the various displacements, apparently gratuitous, but in fact potentially capable of introducing significant changes in the models replayed. Whether or not romancers like Chretien and their aristocratic public were aware of such possibilities for serious play, romances as well as other medieval documents and institutions give ample testimony to crosscurrents of difference and disagreement throughout the Middle Ages. As Douglas Kelly points out, following the example of recent historical scholarship, the aristocracy, as well as the Church, employed the clergie to express its own system, ideals, and values.69 We may well imagine that Chretien's audience expected to find substantial fare mixed into his agreeable fictions.70 Chretien implies as much when he describes Marie de Champagne as supplying him not only with story matter but also san\ the story told within the romance will have some meaning for those outside it, as it refracts (however indirectly) the society that commissions and en-

io6

Chapter 3

joys it. The Charrete itself seems designed to suggest that fictionality is not merely an accessory to truth and san, but part of the process through which we must seek them.71 While truth and meaning are by no means synonymous, Chretien connects them in his Prologue through the figure of Marie as patroness. It is worth noticing that Chretien makes no claim there about the truth of his story, as Marie de France does repeatedly for her lais.72 Nor does Chretien claim to have found his story in a book, as he does in the Conte du Graal and Cliges. In the latter, the book itself guarantees the truthfulness of his story and makes it more believable: "Qui tesmoigne Festoire a voire: / For ce fet ele mialz a croire" (Cliges, w. 23-24).73 The issue of credibility in romance is repeatedly and intimately tied to the problem of truth, even (especiallyI1) where a woman and not a book stands as the source and guarantor of the story and its meaning. In exchange for her gifts of "matiere et san," Chretien offers his praise of Marie de Champagne, praise that he claims to be true, despite its possibly deceptive appearance in the form of a typically exaggerated and possibly lying hyperbole, borrowed from praise of other ladies (w. 7-20). I have analyzed this passage and others to show that one of the impediments to truth-telling in the Charrete is rhetorical: if both truth and lies share the same appearance, Chretien must speak and not speak both at the same time.74 The disjunctions introduced by the mediation of language (and we have learned that the same ambiguity applies to non-verbal as well as verbal signs) are as active in the category of truth, as in that of meaning. Chretien does not need to make any claims for the veracity of his story in terms of its source in history or book; the contradictory and ambiguous nature of signs bridges the worlds of fiction and lived human experience (evoked by the figure of Marie de Champagne).75 This is where romance joins the world we live in, where something can be true and yet lack meaning, where something can be true or not and nevertheless have meaning, where meaning itself is produced by interpretations that may be true or false. These may be meaningless variations in the context of a Neoplatonic Ideal for which Truth is always one, however numerous the repetitions in the realm of the accidental. They are not meaningless in the realm of romance. In the diegetic and extradiegetic worlds that mirror each other in the Charrete^ we will not find Truth, but we may encounter truths, even if they resist totalization and problematize their representability. Chretien's romance suggests that san as direction is not lost: if we follow Lancelot's and the narrator's droit chemin (straight or right road) through all its displacements, its twists and

Chretien's Chevalier de la Charrete

107

turns, the detours of fiction may yield complex truths and varied meanings to those who reflect on them.76 The analysis of the tournament episode has already demonstrated not only the problematic character of signs, but that of the truths they represent: both ends of the equation may be paradoxical; both signified and signifier are inherently ambiguous. If signs are interpretable in different ways from different points of view, subject to the vagaries of belief and the limits of knowledge, the unpredictability of their truthfulness and their meaning reflects the discontinuous and paradoxical nature of an Arthurian ideal, expressed as a courtly and feudal society seeking to combine and coordinate the contradictory demands of competing values. In romance we inhabit a place of multiple truths seen through multiple points of view. There are times when something is true for one character and not for another. Point of view is not simply placement in space, but choice of value system from which to judge a given action. There are other times when we need to see that one character is right, sees, speaks, or interprets truly, and another is wrong, whether because he lies or simply fails (for lack of knowledge) to see, speak, or interpret correctly. Truth is a category we cannot do without, even in the world of fiction. Indeed, the world of fiction may be necessary to show how difficult and necessary the category itself is, even and especially when it cannot be easily circumscribed. As we have seen repeatedly in the Charrete, truth and lies, appearance and reality, inside and outside, may be opposed at times, but more significantly, they are often intertwined in ways that resist polarization: the rumors about Lancelot as cartrider are both true and false, Bademagu's love for his son is both good and bad, Lancelot is the worst and the best at the tournament, love makes Lancelot appear as an adulterer who betrays his lord and king and as the best of Arthur's knights because he is the most faithful lover of his Queen.77 Chretien's fiction offers a privileged space for experimentation and discovery, where displacement and repetition are key modes that allow us to evaluate and arrive at judgment. I have emphasized throughout the analysis of the tournament episode not only how it displaces and repeats the story related elsewhere in the romance, but how it draws attention to the process of displacement and repetition as the core of fictionality. Paradoxically, this same process may also constitute our only pathway to truth(s), since what is not repeated through the echoes of variation and difference may escape our capacity to understand, measure, and interpret. We need repeated observation, repeated verification to arrive at true

io8

Chapter 3

evaluation. The patterns of delay and open-endedness in the Charrete, the rhetoric of saying and not saying, suggest we read Chretien's romance not as a statement containing truth, but as a movement toward truth(s), as an exploration of the problems that movement entails. Given the ambiguities of both signs and the meanings they generate, we cannot express a given truth without the detours of fiction that are part of the way we perceive it. The Charrete makes us focus on the process of perception and proof, the continual displacements that permit and require measurement. Not excepted from the drive to find truth and meaning, we are cautioned about the pitfalls, the complexities, the open-ended nature of the quest.

4. The Interplay of Gender and Genres in Partonopeu de Blois The Protean character of romance as the genre that constantly reinvents itself through fission and fusion with other traditions and its own everrenewable self is nowhere better exemplified than in the anonymous Partonopeu de Blois, probably written toward the end of the twelfth century (ca. 1182-85). Little known today, even to medievalists, Partonopeu was a best-seller in the Middle Ages: it circulated in numerous manuscripts, was imitated, adapted, and translated into German, Dutch, English, Italian, Norwegian, Icelandic, Danish, Spanish, and Catalan.1 Described by Anthime Fourrier as a kind of summa of twelfth-century writing (pp. 440-41), the multiplicity of threads ingeniously entwined in Partonopeu's romance tapestry are elaborated through more than 10,000 verses in the romance proper and then prolonged for an additional 4,000 verses in a Continuation that may or may not be by the same author.2 Partonopetfs experiments in fusion, first apparent on the level of form, operate with equal significance in the semantic field, whether in terms of the romance's major themes or in the deeper political and philosophical implications of its crossovers, shifts, reversals, and recombinations. These will involve not only the interaction of history and fiction, truth and fable, but also the interplay of gender and genres, as the romancer explores the links between knowledge, power, and, choice. As Partonopeu moves from the unidirectional linearity of its opening genealogy into the amiable and open-ended wanderings of romance and continuation, the narrator interjects his own love story between us and the romance plot, connects us to that world through his mediating experience, and thus involves his public, at least by implication, in the same kind of judgments and choices required of his characters. We discover, as we observe and meditate on the actions of the characters and those of the narrator who reports them, that the categories among which we must choose are neither so easily separated nor so stable and immovable as we might have first surmised. A quick summary of the romance will help situate the analysis of

no

Chapter 4

Partonopeu^ which I might facetiously subtitle "Beauty and the Beast." Its story belongs to the same type as that well-known French fairy tale.3 Readers unfamiliar with Partonopeu^ however, would probably guess that Beauty designates Melior and take Partonopeu for the Beast. So he is, at one point in the story, at least metaphorically, and certainly Melior's beauty plays an important role in the romance, but the title roles—more appropriately viewed from the variation of the Cupid and Psyche story— are reversed in terms of gender.4 Like the fairy mistresses of Celtic to, Melior, the Empress of Byzantium, lures the thirteen-year old nephew of the French king to Chef d'Oire, where she plans to keep him secretly for two and a half years, until he can be knighted and presented to her barons as a suitable husband. While he remains invisible to all the inhabitants of Chef d'Oire, Partonopeu enjoys the pleasures of love with Melior each night, but only on condition that she remain invisible to him. After a year and a half (during which he twice returns to France), Partonopeu betrays the taboo imposed, in spite of Melior's warnings. Once seen by her lover, Melior's magic powers are destroyed and Partonopeu is banished. Seeking death in the Ardenne forest, Partonopeu is discovered by Melior's sister Urraque, who persuades him with false news of Melior's pardon to accompany her to Salence. There Partonopeu recovers his health and prepares for the three-day tournament arranged by Melior's barons to choose her husband. Winner of the tournament for his prowess and for his beauty, Partonopeu is finally married to a forgiving Melior (who throughout the second half of the romance has exchanged the role of powerful fairy mistress for that of haughty, but hesitating lyric domna). In an epilogue, the romancer promises to tell us more, if the lady for whom he is writing winks appropriately in his direction—which she does, according to the Continuation.

The Narrator's Double Stance Presented as motivation for the entire romance project, the narrator's desire is repeatedly explored in personal interventions, borrowed from lyric poetry. This lyric persona of the narrator is one of the major innovations of the romance: it appears already in the springtime opening, elaborated in the Prologue as an incitement to write: Si me somont joie et jovens Que je ne soie uisous ne lens.

Gender and Genres in Partonopeu de Blois

in

For c,o wel par envoiseure En escrit metre une aventure, Et bele et bone et mervellouse."5 (Love and youth summon me to be neither idle nor slow. For this reason I want to amuse myself by putting into writing an adventure, beautiful and good and marvelous.) Situation and language evoke the courtly love lyric and the persona of the poet/lover, but the call to composition is translated into the narrative mode of romance and adventure. Before and after his lyric enumeration of bird songs and the link between springtime and love, the narrator presents another image of himself as the typical romance narrator, trained in the arts of rhetoric, a cler who knows about Greek and Latin books and who can bring to a vernacular public an "estoire d'antif tens" (v. 78: "a story of ancient times"). When the Prologue begins with his thanks to God for all the gifts he enjoys from Him (w. 1—4), first of all his knowledge in words and deeds (v. 2), we can recognize in this a variation on the exordial topic that connects knowledge and talents with the obligation to share them (cf. the opening verses of the Roman de Thebes}. These topics, as well as his acknowledgment of an unnamed patron (w. 10, 74), his concern for the criticism of other derc (w. 77-80) and the qualifications of his readers, show the romancer to be as accustomed to the didactic mode as the lyric. This combination of narratorial personae has a significant impact on the triangular relationship set up between author/narrator, the story narrated, and the vernacular public. Partonopeu's story, as he tells it, seems to attract and intermingle two different kinds of responses, typically associated with two different genres: the subjective engagement of lyric and the objective stance of romance. Insofar as the narrator speaks in the authorial and didactic modes, he stands apart from his story and allows us to judge it with the objectivity of distance that mirrors his own. Yet intertwined with that stance is the narrator's lyric persona, which, on the contrary, is repeatedly involved in the characters' stories as negative and positive models for his own love story.6 He is engaged in the romance events and sentiments, as he seeks to engage in them one particular member of his public, Passe-Rose (the name he gives his lady in the epilogue to the Continuation). Not only does he hope that she will draw out the appropriate lessons to be learned from the conduct of his characters (taking the good and

ii2

Chapter 4

avoiding the bad, as his Prologue invites); he also makes her responsible for the story's existence as written document, both as causa scribendi of the main part and arbiter of the continuation. The narrator's own experience is the bridge that connects our world with the world of fiction, marking it simultaneously as other and our own. This connection is all the more striking to the extent that the romancer has not named himself in prologues or epilogues, but only given a kind ofsenhal (secret name) to his lady—as if to short circuit any attempts to singularize his story to that of a particular historical person, while at the same time supporting the human character of the universal model by the claims for lived, personalized experience. Robert Manning analyzes the differentiation in romance between an author designated in the third person and the narrator in the first as an emblematic warning to the reader about the shifting authority and credibility of the narrator.7 In Partonopeu de Blois, where the romancer offers no third person designation for himself, the same sort of play seems to shift to the mix of narratorial personae. The anonymity of the romancer, motivated by whatever irretrievable personal reasons, remains caught up in the fictional network of associations to be analyzed in the link between Melior's invisibility and Partonopeu's incognito. His identity as author appears only indirectly in his masquerades as clerkly educator and lyric lover: the concealment empowers his writing, as he shares with us what he knows about stories of ancient times—and hopes thus to catch in a snare of love his reluctant lady. We might compare these two contracts between narrator and public, one detached, the other sexually engaged (at least by desire), to the contrasting spectacles arranged for different publics by Melior: the knightly jousts and animal combats conjured up without risk for father and tutors contrast with the more exciting and dangerous ones shared with Partonopeu in the darkness of her bed. Their erotic engagement brings into risk the exercise of Melior's powers, which must be protected by the taboo of invisibility. Like the romancer, Melior is a creator of fiction, as she arranges Partonopeu's story to bring him to Chef d'Oire and entertain him during his sojourn. Like the narrator, she tells him stories of ancient times (v. 1868) and even contemporary French history (w. 1911-18). Both the narrator and Melior are lovers whose love generates romance fictions; both seek to use those fictions to invite and control the response of a beloved. Manning views Melior and her sister as artist figures who offer two different models of romance for the creative reader Partonopeu.8 He interprets Partonopeu's betrayal as a challenge to Melior's absolute control of

Gender and Genres in Partonopeu de Elois

113

their story and sees Urraque's manipulations as a more realistic romance model attuned to the demands of feudal society. If Melior's first part does indeed appear more marvelous than the later events, we can see, nevertheless, how her arrangements anticipate the ending more or less as we have it, in the general lines, if not in the specifics. Rather than oppose Melior's fabulous world to Urraque's more realistic one, I would suggest the romance situates them both as stages within the larger experience, each necessary and formative as Partonopeu's and Melior's identities interact and develop through a series of male and female roles, each chosen from a whole gamut of literary models synthesized in the unique blend of Partonopeu.

History and Romance Manning's distinction between romances that are more or less marvelous, more or less realistic, may reflect medieval writers' own comments on one of the more suspect aspects of romance writing, which the author of Partonopeu seems to allude to in the Prologue, when he specifies that moral lessons may be extracted even from pagan tales, "fables as Sarrasins" (v. 104).9 Fables is a key word for understanding the problematic intersection of history and fiction in the romance tradition. Wace's well-known comments in the Roman de Brut describe how Arthur's adventures have become so embroidered in the retelling that they are a mixture of truth and lies in which truth has been transformed into fiction (fable).10 While some romancers may explicitly claim truth and defend their works from charges of falsehood, the author of Partonopeu first deals indirectly with the issue by phrasing it as a linguistic one, occasioned by his shift from Latin to the vernacular, Latin being the usual, authorized language for a "story of ancient times." n In the face of possible criticism from fellow clerc who might accuse him of wasting his time, since he is not writing in Latin, the narrator offers a defense taken from Saint Paul: everything written in books may be put to good use by those who know how to discern good and evil, following good examples and avoiding bad ones (w. 95-100). In this passage the romancer strongly resists the tendency of medieval clerics (and even some modern scholars of the Middle Ages) to stake out an exclusive claim for literacy defined as Latinity; he crosses the boundary between clerical and lay and declares the vernacular, even pagan fictions, as fitting a location as Latin for serious, if also enjoyable, learning.12

H4

Chapter 4

Thanks to this writer, edification and pleasure will go hand in hand (w. 91-94)—not the wasteful pleasures of courtly games like backgammon, checkers, or chess, but the profitable game of learning from examples of all sorts: Mai et bien i doit on trover Par conoistre et por deviser. Nus horn n'eschive mal por sens S'il nel connoist por garder s'en, [Ne ne fait bien dont il ait gre S'eins nel connoist en son pense.] (w. in—16) (One should find good and evil there to know and distinguish. No one avoids evil through good sense unless he can recognize it to keep away from it. Nor does he do good for which he can receive credit unless he recognizes it in his mind.) Knowledge of good and evil, gleaned from someone else's experience thanks to the preservation of writing, gives us the power (at least in theory) to recognize and choose the, good, as well as avoid the bad, even before we make our own mistakes. Although the narrator does not mention here the story of Adam and Eve in the Garden of Eden, as he will much later, we might almost read these verses as an apology for Eve's felix culpa, inasmuch as her fall is the source of our present capacity for such knowlege. But we will better understand that development once we have analyzed Partonopeu's own entrance into and exile from the garden of Melior's love. In the meantime, we can observe with the narrator's specific commentary that any moral problem that may arise in reading stories, even "fables as Sarrasins," is not to be located in the role of the romancer or the language of the tale, but rather in the character of the readers (who either see for themselves or listen to someone's oral reading). Foolish ones can find only the simplest meaning you can touch (v. 118); wise ones know how to achieve the subtlety of bees, who draw honey out of the bitterness of nettles (w. 121-23). The important distinction established in these verses is located between wisdom and folly: these categories will divide the characters into those who will do good and those who will do evil, the readers into those who will fail and those who will succeed in recognizing the difference. As the narrator assures us, there will be "maint bien et maint

Gender and Genres in Partonopeu de Blois

115

mal dit" (v. 130: "many good and bad words") in his story, which guarantees that the discerning reader/listener will have much to learn about imitating the good and leaving behind the bad. The problem as stated here then is not the difference between truth and fiction, but rather our ability and desire to learn, acquire knowledge, and exercise choice.13 The overall strategy of Partonopeu, whether in the Prologue or the romance as a whole, is designed to put into question any simple opposition between what Wace distinguishes as fable and truth. If we focus on the relationship here between truth and fiction, history and romance, what strikes us most forcefully is Partonopeu's seamless fusion of elements, its erasure of clear boundaries. Certainly shifts in the threshold of credibility between the twelfth and the twentieth centuries should make us cautious about deciding what is realistic or marvelous from a twelfth-century point of view. The bestiary lore, for example, that strikes us as pure fancy may have appeared to a medieval audience as scientific fact. But when I speak of fusion, I do not mean to imply a total disappearance of the constituent elements. The medieval public undoubtedly viewed Partonopeu as a blend of fiction and fact. Fourrier has described how different medieval allusions to the romance, and even certain rewritings of it, highlight its fictionality or its historicity.14 Any individual reader may lean to one aspect or the other, but the romance itself seems to revel in the combination, the crossovers between categories of all sorts. It thus invites us to explore the methods and effects of fusion. Consider how the long genealogy (w. 135-498), narrated at the end of the Prologue, provides a kind of transition from history to romance. Genealogies are a popular genre in the twelfth century. As form, they furnish a kind of presumption of historical veracity without the narrator making any specific claim to that effect. The narrator here introduces Partonopeu's genealogy by referring first to his source, not a particular book or author, but rather a group of unspecified Latin and Greek books. This topos authorizes the 364-verse genealogy he is about to elaborate, in which he chooses one of two rival historical traditions, both accepted in the twelfth century, concerning the descendants of Troy. Geoffrey of Monmouth and others may have claimed Trojan ancestry for the British monarchy, but the narrator of Partonopeu reports on the Trojan descent of the French kings through Priam's son Marcomiris. What differentiates Partonopeu's genealogy from the standard of the day, however, is the place where it stops, not (as is usual) with the contemporary figure who illustrates the glories of his lineage, but rather with King Clovis and his

n6

Chapter 4

nephew Partonopeu, still situated so far in the past that the romancer can easily use Clovis's son Lohier to allude both to his actual son, Clotaire I, who lived in the sixth century, and the tenth-century Lothaire (954-86), during whose reign lived the founder of the Blois lineage. Just as the historical tradition followed here contains a gap of eight centuries between Priam and Marcomiris, a gap for which the author of Partonopeu invents his own solution by bringing them together as father and son, so the romancer opens a gap at the time of King Clovis, into which he inserts the story of Partonopeu of Blois, now linked to the Merovingian line through Lucrece, a sister of Clovis unattested by historical documents.15 The transition from genealogy to romance is smoothly effected through the description of the Ardenne forest, which falls within the limits of Clovis's kingdom. According to the narrator, it was filled in those days with exotic animals and other great marvels. There the King goes hunting, accompanied by his nephew, who is lured away from the rest of the king's party and launched on his own marvelous adventure by the hidden machinations of Melior. Thus begins the romance that spins out of the genealogical introduction, leaving the succession of generations behind as it spreads out horizontally within the king's clan and remains fixed in the generation of Lohier and Partonopeu, in order to elaborate the hero's adventures with Melior.16 The historical and the fabulous are not necessarily opposed in twelfth-century historiography—or romance—although they may at times be distinguished. As Paul Zumthor points out, romance itself grows along with and out of medieval historiography; the nature of the "facts" narrated in histories and romances does not per se differentiate them.17 Certainly the marvelous is also the subject of much history writing in the twelfth century, although the separation between history and romance widened as the marvelous became the focal point for romance and the truth it signified superseded claims for historical veracity.18 Most important in this context, the combinations of realm, probabilia and mimbilia in Partonopeu do not so much invite us to cut apart and analyze them into separate boxes, as to explore where the combinations take us, once we admit that the world of romance has many lessons to offer along with its considerable pleasures. Just as the characters move back and forth between West and East, complementing and embellishing the genealogy's single movement from East to West, so are we moved with them through worlds of fiction and history, reality and romance. Chef d'Oire, Melior's capital city built especially for Partonopeu (and modeled by the narrator on

Gender and Genres in Partonopeu de Blois

117

Constantinople), emblematizes the blend of literary genres and the intermingling of reality and fiction that characterizes Partonopeu.19 Fourrier has pointed out how much of the apparently fabulous character of Chef d'Oire and the exercise of Melior's power there corresponds, in fact, to contemporary history: a marriage between Louis VIFs sister and the heir to Byzantium took place in 1180; on a number of occasions in the eleventh century women exercised imperial power in Constantinople; Byzantine emperors were known for their interest in the occult sciences; and twelfthcentury travel reports describe spectacles not unlike those put on by Melior for her father.20 In Partonopeu, as in Chretien's Cliges, we find combinations of both the marvelous and the real to be explored in France, as in Byzantium. Reality is enhanced and extended by the play of fiction; edification and enjoyment are both increased by the combination. Partonopeu demonstrates not only in its representation of Chef d'Oire, but in its whole fabric as story, that the logical tendency to set up truth in a series of oppositions—truth/fiction, truth/deception, truth/lies, truth/appearance—is undermined by the romance model and its supplementary logic, its own tendency to fuse disparate elements into a beautiful semblance that signifies its own truth.21 Douglas Kelly associates this kind of literary truth with what we nowadays call ideology and identifies topical invention as the means to find truth in matters considered by authors and audience to be "credible, if also debatable."22 Partonopeu^ ideology (understood in this sense) might be summarized in borrowed words as "Beauty is truth and truth beauty." But we have to understand all that beauty stands for in this romance, as it constitutes a nexus of interrelated values for the courtly world of Partonopeu.

Beauty and Birth As the narrator tells us in one of his personal interventions, in the time of Partonopeu many courtly virtues went into the winning of love, but first and foremost on the list we find beauty (w. 8010—14). Beauty here is not simply outward charm but the expression of noble blood and lineage, whose goodness is reflected and verified in one's physical identity. We can begin to see this idea developed first in the context of the genealogy, by looking briefly at the three figures who play key roles in forming new generations out of the dying Troy. Through the interrelated actions of Anchises, his wife, and Priam's son Marcomiris, the genealogy anticipates

n8

Chapter 4

some of the major themes explored in the romance proper, as it puts into place the positive and negative poles of lineage and the beneficial deception of women. In the highlighting of engin (wit, cleverness, deception), it even hints slightly at the role reversals that will have such a significant effect on Melior and Partonopeu's story and on the notion that beauty guaranteed by blood alone can be sufficient to found the social good.23 Good and bad blood appear in the story of Marcomiris, who, we are told, was smuggled out of Troy as a baby, protected by his nurse from the traitorous Anchises, presented here (as in the Roman de Troie and its source) as the one who betrayed Troy to the Greeks. Marcomiris grows up the image of his noble brothers: by the age of fifteen, he so resembles Hector and Paris that his true identity is revealed. To save his life, he is taken off to France and thus becomes the founder of the French royal house. Through Marcomiris, the heroic valor of Hector continues to flow in the blood of the French kings. In presenting Anchises as traitor to King Priam, the narrator begins to elaborate a theme that will reappear prominently in subsequent episodes of the romance proper and the Continuation: the "fils a vilain," a serf raised to a position of authority by the king, alienates him from his nobles and leads him inevitably to dishonor. From Anchises's actions in the genealogy we may surmise that noble lineage is no less important for the king's highest counselors than for the king himself. Both the "fils a vilain" theme and the glorification of the royal lineage are based on the common notion that blood determines good—or bad— fruit.24 The theme of beauty, which physically and spiritually embodies that concept, is thus the positive counterpart of the "fils a vilain" theme. But if lineage determines the quality of the offspring, then the narrator must invent a new father for Eneas to escape the ill effects of Anchises's lineage. Adultery offers a possible solution. Eneas's mother thus appears as a kind of mat mariee whose supposed infidelity saves her son from Anchises's bad blood. When the narrator excuses what might otherwise appear to be a moral infraction as evidence of good family planning, he anticipates Melior's later corollary about Partonopeu's good blood inherited from Hector (w. 1501—14): Maus fruis ist de male rai's, Por ce di qu'il [Eneas] n'ert pas ses fis [Anchises]. Franche dame soit honoree

Gender and Genres in Partonopeu de Elois

119

Qui a frarin est mariee, Qui si bel maine son engin Que ses fix ne soit de put lin. Miex vaut bons fix en pechie nes Que mavais d'espouse engenres. (w. 307-14; emphasis added)25 (Bad fruit comes out of a bad root. For this reason I say that Eneas wasn't [Anchises's] son. May a noble lady be honored who, married to a base man, manages her deception so well that her son escapes from an ignoble lineage. A good son born in sin is worth more than a bad one born in wedlock.) Both the narrator's comment and the franche dame's conduct authorize Melior's considerable use of ingenuity and craft to protect herself and her future progeny. When her barons insist that she marry, Melior conducts a world-wide search for a suitable husband. His suitability, as Melior explains later to Partonopeu himself, is to be defined in terms of beauty and character ("por biaute et por mors," v. 1351), since she is sufficiently rich not to have to sell her love. Earlier the narrator had introduced Partonopeu with an elaborate description of his beauty (w. 541-78), recognizable with respect to the ideal of twelfth-century portraits.26 An explicit comment highlighted somewhat mysteriously how important a role that beauty would play: Avant el livre iert bien mostre Por coi je lo tant sa biaute; Car o ce qu'il ot bones mors, Li fist sa biautes grant secors. (w. 579-82) (Later in the book it will be shown why I have praised his beauty so much. For along with his good character, his beauty brought him great help.) Of course, one expects romance heroes to be handsome; what surprises here is Partonopeu's age—a tender thirteen—and the use of his beauty as the primary cause for love, a motif more readily associated in courtly tradition with the lady as object, not the first subject of love. Although he is only thirteen and has not as yet shown his prowess except in the hunt, Melior already knows that Partonopeu will be one of the elect, a "chevalier eslis" (v. 1501). Beauty abounds in this cousin of Hector: blood will not lie

120

Chapter 4

and nature will not suffer him to do anything but that which is noble (w. 1507-14). Melior's explanations to Partonopeu give voice to the underlying rationale that motivates—at least in part—the narrator's inclusion of a genealogy as transition from Prologue to romance. It assures us that the values of birth and beauty passed on from the Trojan royal family to the line of French kings will operate no less in Partonopeu and the more fabulous world of the romance proper. It seems all the more significant then that a romance that began with the carefully enumerated generations of the French royal lineage does not end with any announcement about Partonopeu and Melior's descendants (though we may assume they are to be considered the forebears who illustrate the house of Blois for the romance's contemporary audience). In the course of the romance, the importance of beauty and lineage will be affirmed, but also shown to be inadequate if used as the single principle of harmonious social integration. Melior's choice of Partonopeu at age thirteen makes his beauty the dominant cause of her love: his prowess is expected as the corollary of his beauty, but will only be demonstrated post hoc. The crisis of Partonopeu's betrayal suggests the instability of such an arrangement. In the second half of the romance the use of a tournament will give an appropriate role to the demonstration of male prowess as a sine qua non of male beauty. Or rather Partonopeu's beauty will come to exemplify not only his lineage but also his extraordinary accomplishments as knight. In this respect, we might see the hugely amplified tournament episode, used to lead into the romance's culminating events, as a kind of counter-balance to the genealogy as introduction to the romance proper: where birth and noble line dominate the genealogy leading from the Trojans to the Merovingian royal house, individual performance stars during the tournament fighting and furnishes the basis for judging the finalists. As we shall see, the realization of the tournament itself as a kind of beauty contest, in which prowess decides the finalists and beauty the elected winner, serves to undermine any disjunctive opposition, as it intertwines both sets of values in a more comprehensive system.27

Crisscrossing of Gender and Genres In order to see more precisely how the romancer sets into place, only to complicate and interlace the values of beauty and blood, on the one hand, and the accomplishments of the individual, on the other, we need to follow more closely the romance's developments from Melior's initial

Gender and Genres in Partonopeu de Blois

121

selection of Partonopeu to his final choice as winner of the tournament. In so doing, we must focus particularly on the crisscrossing patterns of gender and genre, which characterize the romance's progress. Through reversals, shifts, and recombinations, this experimentation is crucial for developing the interplay between lineage and personal worth as exemplified in the main characters' beauty and will lead ultimately to further questions about the stable or movable nature of the categories in play.28 At thirteen Partonopeu may already show the good effects of being Hector's cousin, but he cannot be presented to Melior's barons as a candidate for marriage. She has secured a two and a half year delay, so that Partonopeu can be knighted and then chosen by all as her husband (w. 1482—1506). Marriage is necessarily delayed, but Melior is unwilling to delay everything. Like Partonopeu, we learn retrospectively that Melior arranged for the hunt in the Ardenne forest that opens the romance. She arranged further for a lost Partonopeu to find the ship whose beauty calmed his fears (w. 761-62) and led him to Chef d'Oire. This is the city built by Melior on the most beautiful site in her realm as soon as she hoped to bring Partonopeu there and provide for him a secret, but pleasant sojourn (w. 1721-24,1736-40). Chef d'Oire is emblematic of Partonopeu and Melior's relationship in a number of significant ways. Its beauty like theirs is described at length by the narrator in a rhetorical set piece, but we see it as if through the eyes of Partonopeu.29 Our view is thus two-staged: we know only as much as Partonopeu, who is overwhelmed by its beauty and suspects some enchantment ("faerie," v. 809)—all the more so since he sees not a single human being there (as likewise on the ship that brought him). The mystery of invisible people raises the specter of magic, deception, and deviltry, but the beauty itself is reassuring (cf. the connection of beauty and good blood).30 Partonopeu's reactions are seconded by those of the narrator, who fears his efforts to describe the hyperbolic beauty of Chef d'Oire will not be believed and may even generate the same reaction in us as in Partonopeu: Ne vous wel plus loer la rue, Que nel tegnies a fafelue; Mais nus ne set tant de favele Qui par deist com ele est bele; (w. 859-62) (I don't want to praise the street anymore lest you take it for craziness. But no one is so good a talker that he can say how beautiful it is;)

122

Chapter 4

We may or may not worry about faerie, but the narrator's choice offavele may remind us of the Prologue comment on "fables as Sarrasins," especially since Partonopeu himself has been wondering if all he sees in Chef d'Oire is fable (w. 983-84). Although a standard device of romance narrators, hyperbole may sometimes call attention to itself as problematic; as in Chretien's Prologue to the Charrete, it may invite us to ask if we are in the realm of fiction or truth. Of course, the narrator promises an explanation that will convince us that all these marvels are true (w. 867-72). Here, as elsewhere, we are asked to locate the fabulous in the realm of the real, despite the appearance of unreality. Before that explanation is given, however, Partonopeu must encounter the Empress Melior, as he did her city and her palace. And while their beauty is immediately revealed to him, Melior herself remains invisible. Much later in the romance several descriptions will demonstrate the extent to which Melior's beauty is faithfully mirrored by the fabulous beauty of her substitute, Chef d'Oire (w. 7471-510, 9828-40). For now, however, Partonopeu once again equates invisibility with the diabolic (w. 1131—32), until her exclamation in the name of the Virgin tells him that she is no devil (w. 1160—62) and must be quite beautiful, since she speaks so beautifully (w. 1164-65). Just as Partonopeu boldly chose to seek hospitality in the main palace, his fears reassured by beauty and his own nobility inspiring him, so now he overcomes Melior's (apparent) resistance and makes love to her. In their subsequent conversation Melior explains who she is and what she has arranged for Partonopeu. Chef d'Oire will be for him a place of uninterrupted pleasures, on condition that he accept Melior's prohibition against seeing her for the next two and a half years. During the day, Partonopeu can enjoy all the visible beauties and delights of Chef d'Oire, in the darkness of each night, the unseen Melior and the pleasures of love. Partonopeu accepts Melior's condition and, predictably, fails to keep it, once he falls under the influence of his mother and the Bishop of Paris during his visits back to France. What is fascinating in this setup and elaboration of the plot is how the romancer plays with male and female roles and the consequences of their reversals. This ploy situates Partonopeu de Blois at the crossroads of a number of literary models, each of which furnishes elements for the romancer's unique reinventions. Scholars have long recognized the Cupid and Psyche story in Partonopeu's variation and puzzled over what form of it the romancer might have known.31 By putting Melior in the role of Cupid, the author has given her the usually active, male role and endowed

Gender and Genres in Partonopeu de Elois

123

her with supernatural powers that may not raise her to the status of a god, but do furnish her with sufficient magic to maneuver Partonopeu through the first part of the romance. In this maneuvering we are explicitly invited to notice another layer functioning in Melior's identity, especially when the narrator describes Melior's approach toward a fearful Partonopeu with the words, "Atant vient une fee al lit" (v. 1125: "Now a fairy comes to the bed"). Attentive readers have by this time identified Melior with the fairy mistresses of Celtic lais.32 But we soon learn, along with Partonopeu, that Melior is not a fairy, although her powers and conduct in the first part of the romance are clearly modeled on the kind we see in Graelent or Marie de France's Lanvcd. The supernatural character of these fairy mistresses authorizes their dominant role in the couple and justifies their power to impose a taboo on the merely human male lover. How does this compare with Melior's situation vis-a-vis Partonopeu? What stands out in Partonopeifs variations, once their common features have been recognized, is the "humanization" of the Celtic model, which yet retains and, in a sense, amplifies the reversed male/female hierarchy of the lais. Melior is Empress of the Byzantine Empire, an exotic and fabulous world to be sure and a source of considerable power with twenty-five kings under Melior's sovereignty; but it is not the Other World of Celtic myth, situated on a supernatural plane. As represented in the romance, Byzantium operates according to the same feudal hierarchies as contemporary France. Furthermore, Melior's powers are not the "natural" endowment of a fairy, but rather the result of hard work and years of study. They are developed in her, as she explains, because her father made sure his only heir received what might otherwise be considered a male education, the entire encyclopedia of twelfth-century learning (w. 4583-622). Melior has been prepared for and taken on the normally male role of ruling the body politic. She is, however, like any feudal king of the twelfth century, subject to the "advice and consent" of her noble barons—hence the initiation of her marriage strategy, which follows, at least in part, the sort of plot engineered by Celtic fairies. Melior's assumption of male prerogatives necessarily affects Partonopeu's characterization and development. Unlike the knights chosen by fairies in the Breton lais, Partonopeu at thirteen has not yet fully assumed his manhood. In some sense, Partonopeu's youth corresponds to Melior's status as empress: both characterizations exaggerate and rationalize their disparity in the social hierarchy. Partonopeu's youth thus justifies to a certain extent the passivity to which Melior's activity seems to relegate him.

124

Chapter 4

From the age of thirteen through fifteen, Partonopeu is a romance hero managed, in consecutive stages, by Melior, his mother, Melior's sister Urraque, and Gaudin, his companion at the tournament which dominates the second half of the romance. Is Partonopeu simply growing up, moving from a female to a male space? We might compare him to Perceval in Chretien's Conte du Graal, whose transition from his mother's tutelage to Arthur's court is routed through Gornemant de Gorhaut's training. That comparison and its puzzling ironies suggest, however, that romance "reality" is more complex than any simple linear trajectory. What we may observe in Partonopeu is not just a shift from female to male in the balance of power, but rather a dynamic interplay between the two, which varies considerably in the different stages of the romance plot, but consistently focuses on the interaction of male and female power, as it is revealed or concealed.33 In the first part, while Melior remains subject to her barons' marriage wishes, she nevertheless exercises and conceals her total control over Partonopeu and her own plans for arranging the desired marriage. The role of surrogate mother and tutor, part of the Celtic model that Melior's fairy ancestry offers, is amply exploited in this section.34 Since fairies are ageless, they may serve in some stories as foster mother and mistress with no contradiction or conflict, but this is clearly more problematic in human terms. Melior's authority certainly makes her seem older than Partonopeu. She tells him, in fact, that before the age of fifteen she had surpassed all her tutors (w. 4609-10). The situation suggests an older and more experienced Melior completing Partonopeu's higher education. But as the years of his sojourn begin to go by, and especially once he is ejected from Melior's protective invisibility, we tend to forget their theoretical age difference and the resonances of Melior's role as teacher. When Melior's magic powers as quasi-fairy mistress are stripped away by Partonopeu's betrayal, their relationship retains its reversed hierarchy in terms of another literary model, which until now has been operating concurrently with that from the lai: the male lover of troubadour and trouvere lyric who humbly serves his beloved as domna, vassal subordinated to lady. In one sense, we might say that Partonopeu's betrayal reduces Melior's power to that attributed by the male lover to his lady, though we should not forget that this metaphor of power is realized politically as well by Melior's role as empress. Not surprisingly Melior's descriptions of her pain and sorrow, whether in anticipation of Partonopeu's betrayal (w. 4229-64) or after (e.g., w. 4554-66, 4703-74), as well as

Gender and Genres in Partonopeu de Blois

125

Partonopeu's own love laments (e.g., w. 5203-58), echo the words and phrases of troubadour and trobairitz lovers. Melior's sufferings parallel the poet/lover's, although they are located on the opposite side of shared love, before and after.35 Her laments correspond to the desired image of the lady in love, although she also appears as the haughty, rejecting domna feared by the lover. This is the role she plays the morning after Partonopeu's betrayal and continues to play in public, despite her regrets and change of heart, in order to shield her true feelings from court gossip. Melior's monologue during the tournament shows her awareness of such role-playing, when she complains about the different standards for men and women in love, the latter forced to pretend indifference, the former able to declare their love openly (w. 9092-126). Already much earlier we may have noticed such role-playing as a function of the romancer's own exuberant and exaggerated use of lyric discourse, as when he has Melior exclaim about her loss of Partonopeu through an enumeration that lists anaphorically all he represents for her on the plus and minus side (w. 4735-66): "Vos est'ies tos mes delis, Mes cuers, mes pros et mes porfis, Et ma noblece et ma bobance, . . . (w. 4735-37) (You were my delight, my heart, my profit and my advantage, my nobility and my pride, . . . ) Concentrated in a few lines, such enumeration speaks with lyric charm, as we see, for example, in Chretien's first romance, when her father describes Enide to Erec (w. 543-45). Extended over thirty-two verses as in Melior's complaint, such an exaggerated litany may cause us to smile at the romancer's game, despite the seriousness of Melior's feelings. At such moments, we are explicitly reminded that the story of Partonopeu and Melior involves not only the characters and their history but the romance's representation through the interplay of roles and genres. In this crisscrossing of literary models, we should recall that Partonopeu's betrayal also corresponds to the typical romance pattern of crisis disrupting the initial harmony of the first part and setting a problem to be resolved in the second.36 As in Chretien's Train, that crisis is phrased here in terms of the lover's betrayal of his lady's command. Their reconciliation will be effected, again as in Tvain, through the good offices of a female

126

Chapter 4

intermediary.37 This typical romance plot resembles a narrative working out of what we often see in lyric tensos, when a lover asks another lady for help in seeking his beloved's pardon.38 In the face of Melior's adamant refusal to forgive Partonopeu, Urraque is forced to invent a false report ("fause noveile," v. 6055) of pardon, when she discovers Partonopeu dying of remorse in the Ardenne. She then arranges for his recuperation at Salence and his dubbing by Melior in Chef d'Oire with all the resources of secrecy at her disposal (having her own island is a big help). Her cover is considerably less effective, however, than Melior's previous power of invisibility—and gives rise to the dramatic near recognition of the dubbing scene, when Partonopeu ("hidden" by his helmet) is exposed to Melior's beauty. Nor can Urraque maintain her control over the plot any more effectively than Melior, when a contrary wind takes Partonopeu away from Salence and lands him in Armant's prison—from which Armant's wife will be called upon to release him for the tournament (an echo from Chretien's Charrete, which reminds us how much Partonopeu, like Lancelot, is a particularly attractive object for women helpers).

The Interplay of Male and Female Power, Male and Female Beauty With the typical romance event of the tournament, we move apparently into the domain of male power, where the prowess of the knight participants and the evaluation of the barons constitute the determining factors. We discover, nevertheless, that the shift from performance in combat to the non-physical act of judgment leaves ample room for Melior's manipulations. Once Partonopeu arrives at the tournament, having left behind all his female managers, his actions are subject primarily to male judgment, whether that of his companion Gaudin, who orchestrates all his entrances and exits, or that of the seven judges chosen to elect six to ten of the best knights as finalists. Although the conditions of the tournament give Melior the final choice, she exercises that choice in a deceptive manner, manipulating the judges who speak out for Partonopeu, while she claims to defer to their judgment, despite her own preference for the Sultan of Persia. Why, we may wonder, does Melior need to hide her real choice from the judges? There will be several kinds of motivation to explore on that point, but for now Melior's own reasons point to the typical situation of courtly lyric. In monologue, she complains that men have the better part

Gender and Genres in Partonopeu de Blois

127

in love, since they may declare their feelings, court actively by word or letter, while ladies must play hard to get and cover their desire with silence (w. 9091—114). This is a far cry from Melior's bold conduct in the first part of the romance, but the narrator reinforces her assumptions about the lyric lady's role, when he expostulates at length during the tournament against scandalmongers always ready to criticize ladies and turn whatever they do into something to blame (w. 8397-428). Rather than run such a risk just when her heart's desire is about to be realized, Melior prefers deception and subtle manipulation. We may remember in any case that Melior's original plan, as described during the first night at Chef d'Oire (w. 1476-78,1496-1506), already anticipated such careful control. She has finally contrived to have her barons choose Partonopeu de Blois as her husband. The predicted and predictable has been arranged in the unpredictable manner that typifies romance plots. Melior's exercise of power throughout the romance entails secrecy and manipulation, however these factors vary in object, degree, and kind, as they are manifested before and after Partonopeu's betrayal. Her loss of magic powers plays a significant role in differentiating the first and second halves of the romance, but Melior as Empress of Byzantium in her own name and as the object of pursuit by numerous suitors continues to operate from a nexus of power throughout the romance proper. The shifting back and forth between male and female controls or rather the varying combinations of both within the plot is doubled in the frame supplied by the narrator's own love story, which repeatedly functions as a commentary on the characters' story. We recall the Prologue's springtime opening initiated the romance's lyric connection through the narrator's own je. At that point, we learned that the narrator loved, perhaps like the oriole a distant love ("lointaine amor," w. 49—6o).39 This evocation of the lyric system recalls the balance of power between domna and poet lover, typically based on a set of reversed hierarchies. Thematically the poet puts his beloved in the position of the feudal lord and thus reverses his society's subordination of women to men. In the poem itself, however, real power is exercised by the poet as the manipulator of words and images. The feudal metaphors that apparently give power to the lady are also meant to control her response and restore order to the potentially disordering and unpredictable effects of love. This is precisely what we see in the romancer's lyric persona as well. Like the troubadour lover, the narrator exercises his own power in poetic display, as he places his enormous literary talents in the service of his beloved, uses poetry to praise

128

Chapter 4

ladies and win love. Over and over again, his interventions show him to be particularly inclined in favor of ladies, ever ready to leap to their defense and justify their conduct in love, although (again like the troubadour poets) he does not hesitate to use his control of the word to complain about his own lady's conduct or lecture her and the general public on what he believes to be proper conduct. Unlike Partonopeu, he will never betray his lady; unlike Melior, he promises always to pardon his beloved, no matter what she does (w. 4543-48). Parallels in positive and reverse patterns cross and crisscross the distribution of gender and hierarchy, but always the narrator claims to be in his lady's power. He thus makes her responsible for his choices as poet/romancer, whether in the decision to continue (w. 10607-24) or in the changes of meter and rhyme made in accordance with her will (w. 1469-73^.40 But despite his apparent submission, the narrator complains of how little he has received for his service. The lady's exercise of power does not always fit the image of his desire. In this respect, Partonopeu and the narrator are located on opposite trajectories of success and failure: Partonopeu has all the good, the narrator all the suffering (w. 1880-86). As he sums up Partonopeu's situation during his first year's sojourn at Chef d'Oire, the narrator contrasts it with his own: what one has, the other does not. What the narrator has, and has only, is the unobstructed vision of his lady. What he does not have is the pleasure of touch, the exclusive domain of Partonopeu's joy. We understand what lies behind his lament from our knowledge of courtly lyric and its commonplaces: the lady's beauty is the first cause of love, as it wounds the heart through the eyes. Such is the normal beginning of love, and the narrator remains uncomfortably stuck there, unable to proceed on to the next steps of the gradus amoris, until his lady accepts his suit. As Douglas Kelly points out, the conceit of Partonopeu's plot is to reverse the normal order of love by starting at the end and going backward to the beginning.41 Partonopeu falls in love with Melior because he has enjoyed with her the ultimate joi^ as he himself explains, when Melior fears he will scorn her as too quick to give her favors (w. 1427-40). Having felt her, he is now anxious to see her (w. 1441-42)—that is, to complete the usual gamut of a typical love relationship. We have already seen how Melior's prohibition delays that progress (backwards) and how Partonopeu's betrayal dramatizes and traumatizes the moment of sight by rushing it against Melior's will. What is remarkable in the narrator's presentation of that scene is the way he withholds from the romance public a description of Melior's

Gender and Genres in Partonopeu de Blois

129

beauty, even though it has now been fully revealed to Partonopeu by the light of a magic lantern. Partonopeu's vision of her beauty is quickly summarized: he sees her completely naked and has never seen anyone so beautiful (w. 4528-30). The narrator's discretion before Melior's naked body is all the more remarkable in view of his earlier unexpectedly detailed description of Partonopeu's sexual advances on the invisible woman's body (w. 1302—4). Instead of a description of Melior, however, we are offered the next morning an elaborate portrait of her sister Urraque, which includes a detailed description of her head and body, as well as her richly ornamented clothing (w. 4879—922). Only much later will we actually see Melior in two descriptions, one at the dubbing ceremony (where the effects of her beauty on Partonopeu and the other new knights leave us no doubts about her beauty's capacity to inspire love, w. 7453-514), one on the final day of the judgment (w. 9828-40). In both cases, the narrator devotes more verses to describing the viewer's reaction to Melior's beauty or summarizing it, than to an actual description of her physical charms. In fact, we learn more details about her clothing than about the shape of her head, nose, mouth, and so on. Presumably we can reconstruct these from our general knowledge of ideal feminine beauty in twelfth-century literature and from the specific example of Urraque's portrait. When it comes to descriptions of beauty in this romance, Partonopeu, not Melior, is the primary object of our agreeable contemplation. When morning brings discovery, once Melior's powers no longer conceal Partonopeu from her court, Melior's ladies first blame her for giving herself to a "vallet" (v. 4850). But their blame quickly gives way to admiration for Partonopeu's beauty (w. 4863-74): in the six repetitions and variations of the verb esgarder that proliferate in twelve lines, we see the ladies stare at him, more and more fascinated, absorbed by the vision of Partonopeu's beauty. The portrait of Urraque follows immediately and functions, in some sense, as a displaced crystallization of that beauty— indeed, the ideals of male and female beauty share a number of features, as detailed comparison of Urraque's and Partonopeu's portraits would reveal. Partonopeu's beauty, already described at great length, when he was first introduced in the romance's opening scene, is here merely stated and then amplified through the increasingly attentive stares of Melior's ladies. Their reaction anticipates that of the young knights mesmerized by Melior's beauty and reminds us that Melior herself first fell in love with Partonopeu because of his beauty. In Partonopeu de Blois, the gradus amoris is not only reversed in re-

130

Chapter 4

lation to the order of its steps; it is reversed in terms of its usually male point of view. From Melior's point of view, the steps of love proceed rapidly and directly from the inspiration of Partonopeu's beauty to the consummation of love's joy. As Melior explains to Partonopeu, once she heard the report about him, she immediately fell in love, set off on a secret visit to France to confirm the report, and then put into motion the events that would bring him to Chef d'Oire (w. 1369—90). We are confronted again with the play in crisscrossing gender roles that typifies this romance. Given that pattern, we may notice how the key events marking the successive stages of the romance plot are figured as a series of beauty contests. I mentioned earlier the strategy used by Melior to find a suitable husband based on beauty, not wealth. Her search resembles that of King Ahasuerus in the Book of Esther: Melior heard reports about many good men from the spies she sent out to search all over for a whole year, but those who had seen Partonopeu were the most joyous (w. 1356-64). Such is the first beauty contest, but since the winner is a young child (v. 1366), marriage must be deferred, with the results already described. The scene of Partonopeu's betrayal may be seen as the second variation on the beauty contest—this time with a normally female contestant—to the extent that it is Melior's beauty that proves to Partonopeu, as soon as he sees her, that she is not a devil. Devils, he has been warned, are identifiable by their ugliness (v. 4476). The third variation, in fact a reversal, occurs at the nadir of Partonopeu's life, when Urraque finds him in the Ardenne. At this point, we are offered a second portrait of Partonopeu (w. 5957-72), through which the extent of his fall can be measured by antithesis: his idealized beauty has been transformed into its opposite, the "idealized" ugliness of a wildman.42 Just as his beauty reflected Partonopeu's noble blood, so his extreme ugliness mirrors the fault of his betrayal. That fault, however, derives mostly from the bad counsel of his mother and the Bishop of Paris (cf. the "fils a vilain" theme), given during his two returns to France. Once he is placed under Urraque's care, the fundamental goodness and nobility of Partonopeu's character allows him to be quickly restored to his former beauty. Persewis, a young lady introduced in this episode to help care for Partonopeu, seems to have been invented chiefly to demonstrate once again the power of Partonopeu's beauty to inspire love in any susceptible female who sees him. During the preparation for the tournament, Persewis sees Partonopeu armed for the first time, a considerable enhancement of his beauty in her eyes and the occasion for increased love suffering for her. Her reaction

Gender and Genres in Partonopeu de Blois

131

reminds us that normally in romance it is female beauty that inspires love in the man, male prowess that inspires love in the woman. Partonopeu has by now demonstrated considerable prowess during the Sornegur episode, but Melior's earlier choice of him at age thirteen made his beauty the cause of her love and led to the complications of betrayal and the continued need to arrange a marriage to suit the demands of her barons. I suggested above that the transformation of the tournament into a kind of beauty contest has the effect of intertwining beauty and prowess, birth and individual performance, as complementery, rather than contradictory, values, equally necessary in a comprehensive social system. A closer look now at the tournament itself should clarify how this interlacing of values occurs.43 Through Melior's report to Urraque we learn the conditions set up for the tournament, as suggested by her chief advisor Ernol: there will be three days of fighting, with pagans and Christians evenly divided on both sides. Seven judges will choose six to ten of the best contestants, among whom Melior will make the final choice. Such are the initial terms, but there is some tension in their realization. Neither the Sultan nor Partonopeu is content to be among the finalists; each wants to win and be chosen as the winner of the tournament. And each one fears the other has won, when the final day of fighting draws to a close. Their fears seem not unwarranted, given the action of the judges themselves. Once Anfors has enumerated the six finalists, they begin to discuss who should be declared the best among them, Clarin choosing the Sultan, since he is the richest, six remaining silent, and Corsolt supporting ErnoPs choice of Partonopeu. To help resolve the dispute, Corsolt asks Melior to see them both unarmed and make her own choice: —"Par foi, fait Melior, ce quit, Que vos en aves le melz dit, Si ne ruis pas qu'il soit cele Que je tir molt a la belte: Molt vuel que cil soit belz et gens Qui de moi fera ses talens Et qui m'avra tote ma vie, Si ne vos en merveilles mie, Car si choisist chascuns de vos Quant doit prendre une de nos." (w. 10385-94) ("By my faith," says Melior, "I believe you have spoken the best. I don't seek to deny that I am much drawn to beauty. I want the one

132

Chapter 4 who will have his will of me all my life to be handsome and noble. This shouldn't amaze you, since each one of you chooses thus when you must take one of us.")

What is striking in Melior's comment is the way it acknowledges the typical model against which Partonopeu's reversals play—an acknowledgment that comes precisely at the point when demonstration of prowess, as judged in the course of the tournament, has not been sufficient to determine who Melior should marry. Prowess is followed by debate, but Partonopeu is chosen unanimously by all—as he was originally in Melior's first private beauty contest—once the standard of beauty is invoked and applied to the two outstanding finalists. Male performance in the public arena is essential, but beauty (with its network of associations in this romance) is still the higher factor that caps and encapsulates all the rest. The gender reversal here reminds us that it is normally female beauty that causes all the stir and male prowess that serves it. Women are contemplated and men usually choose them to enjoy their beauty, as Melior indicates. Men are loved for what they do, not for how they look, even if the romance hero's looks generally support the favorable impression of his prowess. In this romance, however, Melior chooses based on Partonopeu's beauty. The "disappearance" of her own beauty may be interpreted, in some sense, as the concomitant of her taking on the male role, just as the focus on Partonopeu's beauty suggests the feminization of his (cf. the emphasis on his youth in the first part of the romance). We are led to such a conclusion, in particular, if we consider the apparent arbitrariness of Melior's prohibition and her subsequent loss of magic powers, an arbitrariness that stands out with respect to the various literary models Partonopeu conjures up. Consider first the Cupid and Psyche story, which clearly furnishes the prohibition of sight, since the element of secrecy in all the Celtic tales of fairy mistresses involves rather a taboo against speech. The concealment of Cupid's physical appearance is motivated within the story's own terms by his desire to conceal his identity as a god. Indeed, as soon as Psyche sees him, she recognizes the beautiful god whose wings and bow and arrows confirm his identity. And while both Cupid and Psyche and Partonopeu de Blois may be properly located among the dozens of variants of the type "man or wife in search of their supernatural husband,"44 their common background serves to highlight the changes made by the twelfthcentury romancer, especially when we realize that it is not seeing Melior that suggests to Partonopeu that she is supernatural. Melior's beauty, once

Gender and Genres in Partonopeu de Blois

133

revealed, humanizes her, shows her to be all too human in comparison with a real fairy, since Melior is henceforth only a beautiful woman who can no longer exercise her magic powers, acquired through her male education.45 Neither Cupid nor the Celtic fairy mistresses find their powers contingent on their lovers5 behavior, but this is precisely, mysteriously, what we discover to be the case, once Partonopeu betrays Melior's prohibition of sight.

Invisibility, Power, and Knowledge Scholars have spoken of the "rationalization" of Melior's fairy ancestry, replaced by a kind of university education, but if we follow that implied logic we are then confronted by the irrational character of Melior's loss of power. Any explanation for this apparent arbitrariness with respect to the romance's models must be found in the coherence particular to Partonopeu de Blois, a coherence that is not necessarily limited to the psychology of a given character (at least if we judge human psychology by twentiethcentury notions). There is, however, another narrative model not yet discussed, which throws a helpful light on Partonopeu^ inner logic. Partonopeu himself signals that allusion when he compares his ejection from Chef d'Oire with that of Adam from Paradise (w. 5229-35). Like Adam, Partonopeu is ejected from a garden of delights through his own disobedience, but inasmuch as the role played by God in the Bible story is here assigned to the woman of the couple, Partonopeu loses not only his honor and fief but the lady herself.46 Partonopeu's comparison highlights the kind of role reversals we have already seen to be typical of the gender patterns in the romance. It may lead us further to consider an important nexus of elements brought together by Melior's prohibition: the link between invisibility, knowledge, and power. The story of Adam and Eve, particularly as represented in the twelfthcentury Jeu d*Adam^ concerns not only the hierarchy between lord and man (vassal), but also the bond between male and female: will it be hierarchical, as it mirrors that between God and man, or egalitarian, as befits two images of the same species, humankind? The prohibition not to eat from the Tree of Knowledge can be considered a kind of game, a serious one to be sure, a contest to see whose will prevails: God's? man's? or woman's? God (with)holds certain powers, as long as Adam and Eve do not eat the fruit of the Tree of Knowledge. Once that fruit has been eaten,

134

Chapter 4

however, the knowledge of good and evil belongs to man and woman as well, who immediately recognize their own shame and try to hide it from God. Knowledge opens the path to power and God must impose a new limit. Lest Adam and Eve eat from the Tree of Life and acquire the truly godlike power of immortality, they are exiled from Eden. It is a truism to state how important a role this passage has played for the inheritors of the Judeo-Christian tradition in general and the Middle Ages in particular. The story of Adam and Eve read as history and metaphor represents a potent cultural model for the organization of knowledge and power, sexuality and free will. As I suggested earlier, this felix culpa empowers us to this day, as we follow through on the Prologue's invitation to discern moral lessons from good and bad actions recorded in the romance. It also throws a revealing light on the kind of choices made by Melior and Partonopeu, and on their relationship as a series of experiments in shifting hierarchies. These experiments ultimately raise questions about the substitutability of different members of the same species or even the equality between them. Melior's double role as both woman in the garden and supernatural arbiter/creator of its delights situates her, at the moment of Partonopeu's betrayal, both inside and outside the circles of power and knowledge. As long as Melior remains invisible, she controls who knows what by controlling who sees whom. As long as Partonopeu cannot see Melior, the two lovers can enjoy their shared sexual knowledge, while Melior's fairylike powers keep Partonopeu and her subjects invisible to each other. They all occupy the same world, but respond separately to a daytime or a nighttime image of Melior. Partonopeu's betrayal destroys Melior's powers of invisibility, as it disseminates the knowledge of their sexual enjoyment. Melior is particularly pained by the shame that will result, as she explains to Partonopeu, and that is precisely what we see the next morning with the arrival first of Melior's ladies, then of Melior's knights.47 The process of dissemination, in fact, began earlier that night with Melior's explanations to Partonopeu about the source and limits of her powers. During Partonopeu's first night at Chef d'Oire, Melior had identified herself as the Empress of Byzantium, but she did not mention her magic talents, which operated secretly on Partonopeu and her court. If we consider this nexus of power and knowledge more specifically in terms of the reversal of roles that characterize Melior and Partonopeu's relationship, we may begin to see an equivalence—mutatis mutandis— between the female use of invisibility specific to Partonopeu and the male

Gender and Genres in Partonopeu de Blois

135

use of incognito, as we see it here, as well as in romance generally. Indeed, at first glance, Partonopeu's incognito at the three-day tournament appears as gratuitous as Melior's prohibition of sight seemed arbitrary. Partonopeu's initial desire for anonymity is motivated first by his desire to die undisturbed by any do-gooders in the Ardenne, then by his desire not to be known as Partonopeu in his present ugly state (w. 6163-68). Urraque needs Partonopeu's incognito to protect her own lies and to punish Melior's lack of pity, even when the lady herself has begun to regret her rigor. But once Partonopeu arrives at the tournament on his own, why does he want to do so incognito:1 According to Urraque's "fause noveile," Partonopeu believes that Melior has forgiven him. This is a tournament for Melior's hand in marriage: would it not seem logical for Partonopeu to declare his candidacy openly, as do the other combatants at the tournament? Of course, we could conjecture on the public shame already generated by his betrayal of Melior: this would seem to furnish motive enough for Partonopeu's anonymity and yet the romance itself remains silent on that issue. When Partonopeu hesitates to tell Gaudin who he is, he decides to reveal his name without adding any details beyond the immediate circumstances: he is Partonopeu from Tenedon (w. 7830—35). This is no lie, since Tenedon is where he was held captive by Armant, a piece of information not shared with Gaudin until the end of the tournament. Partonopeu tells no lies, but does not tell all. Yet when Anfors gives a biography and portrait for each of the six finalists, he seems to know all about Partonopeu, his lineage and his character, although the description makes no mention of his earlier liaison with Melior (w. 9963—92). Anfors offers no explanation for his source of information, whether to the public addressed in the romance or outside it. While we can thus enumerate a series of partial explanations for Partonopeu's incognito, depending on a number of characters' viewpoints, the real reason for his anonymity comes, I think, from the romance tradition itself. Alexandre Micha has described in his introduction to Cliges how Chretien's four-day tournament becomes an important model for later romances (pp. xvi-xiii). Partonopeu clearly alludes as well to the tournament in the Charrete. Partonopeu appears at the tournament as "cil a eel escu d'argent" (w. 7985, 8117, and passim: "the one with the silver shield") because the hero of romance typically uses incognito at tournaments to enhance and dramatize his extraordinary, heroic status. A knight's incognito raises our curiosity by creating a kind of tabula rasa that momentarily erases any previous reputation and dramatizes the performance

136

Chapter 4

of the unknown knight. A stage of incognito typically prepares for a new level of attainment for the romance hero, whether as with Cliges it first introduces him to Arthur's court, or as with Lancelot it crowns an already extraordinary career. We can see this very clearly in the pattern of entrances and exits carefully and consciously orchestrated by Gaudin for his companion and repeatedly commented on by the judges, who know immediately that he is someone worth watching (e.g., w. 7932—40). Indeed their commentary of the knights' performance—and of Partonopeu's in particular—is one of the most important aspects of this tournament cum beauty contest (cf. the role of spectators' comments during the tournament in the Charrete}. Enhanced by his use of incognito, Partonopeu's performance rivets their attention each day, as they watch to see if this unknown knight can live up to his previous feats, in order to become one of the finalists in contention for Melior's hand in marriage. Partonopeu's incognito is thus maintained identically from day to day (unlike Cliges's daily change in the color of his arms) to play out his drama not only before the judges, but also with respect to the ladies. In another allusion to the Charrete tournament, we can once again see the pattern of male/female reversals, when Partonopeu's cryptic words, "Mar vos vi" (v. 8365: "Woe to me that I saw you"), furnish the key first to Urraque and Persewis's recognition and then, with their help, to Melior's. Not only does Partonopeu's specific choice of words play off against Enide's fateful words in Erec et Enide (where the variations on mar fui, mar fus, mar vi, etc., become a kind of leit motif in the second half of the romance), but the word play itself recalls Guenevere's secret commands to Lancelot to fight "au noauz." Where Guenevere uses a pun to identify and then verify who the knight with the red arms really is, Partonopeu speaks the mysterious words whose meaning is clear only to those who know Partonopeu's past and present. Incognito as tool in a knight's establishment of his true identity is a kind of controlled visibility that conceals and reveals at the same time, ultimately revealing more thanks to the greater clarity bestowed by the momentary obscurity of anonymity. It is a typical romance ploy available for the hero and here used at great length through the amplification of the tournament episode, its preparations, execution, and consequences, which together constitute the more typically romance-like second part of Partonopeu de Blots. This same power to control and manipulate knowledge characterizes Melior's use of invisibility in the romance's first part. As the female equivalent to the knight's incognito, invisibility allows Melior to

Gender and Genres in Partonopeu de Blois

137

reshape her identity—at least partially, since she is obviously not interested in a complete sex change.48 In order to keep the male position of power, even in the embrace of her lover, Melior must cover her female beauty and thus deprive Partonopeu of the opportunity to fall in love with that beauty and follow the normal course of the gmdus amoris from the man's point of view. In some sense, we might say that Melior willingly foregoes the power exercised by her beauty—a passively female power that she leaves to Partonopeu. She sacrifices her beauty as the price required for sexual enjoyment seized before the appropriately public bestowal of such rights in the sacrament of marriage. Melior has already been accustomed to enjoying the exercise of her magic powers in the secrecy of private chambers. Where no sexual engagement was involved between magician and public, no risk in the loss of power was entailed. But now with her lover as "spectator," his participation in the pleasures is necessarily accompanied by a loss of vision, so that power and knowledge remain under Melior's control. Her political identity as Empress clearly does not suffice here; secrecy requires the magic powers acquired through her male education. If the act of love itself requires that each keep his or her own sex, their relationship as a whole will continue only as long as Melior remains in charge. Or at least the relationship as she defines it will continue only until Partonopeu's betrayal, at which point we follow first the lover's despair, then his conquest of Melior's love through the intervention of Urraque, the tournament, and so on—that is, the deployment of a male quest to develop and enhance the hero's identity, the factor conspicuously missing from the romance's opening moves, when Partonopeu did nothing per se to gain Melior's love, but only was the youth whose beauty promised the knight he would become. Melior's own identity and power are profoundly modified at the moment of crisis and must be reconstructed on the basis of a new relationship with a newly defined Partonopeu. Her previous identity as fairylike mistress in the romance's first part is completely dependent on her power of invisibility. As she explains to Partonopeu, his betrayal does not rob her of all the knowledge gained in her arduous years of study, but only of those magic talents that enabled her to conjure up beautiful spectacles in private chambers (w. 4661-71). Deprived if her magic talent, Melior reverses—or at least appears to reverse—her role in the balance of power. Where she formerly played to the hilt that of the bold, manipulating fairy mistress, she will now go to even greater lengths (through another 6,000 verses) to act out all the haughty rejection, as well as the fear, suffering, and hesi-

138

Chapter 4

tation, of the lyric lady, whose manipulations must, as a result of court scrutiny and male criticism, be ever so much more subtle and secretive. As she indicates to Partonopeu, she is a master of all the arts, including those of trickery ("Et tos engiens et tos baras," v. 4666). In the first moments after Partonopeu's betrayal, she may disdain the knowledge she still holds, but as the rest of the romance demonstrates, she will not disdain to use the power of engin to obtain her secret desire. Power still resides in the control of knowledge, even if that power is no longer magical. Although there is significant variation, to be sure, in the second half, as in the first, power is located throughout in the control of what is or is not seen. Thus Melior's deceptive hiding of her true feelings is the new form her invisibility takes. Even when she finally abandons the model of the lyric lady, even—and especially—when she stops hiding from herself the extent of her love for Partonopeu and fully endorses the pardon already given by Urraque, the Empress continues to conceal her true sentiment from the court, in order to guarantee their choice of Partonopeu through covert manipulation rather than any open declaration of her real preference. Indeed, she hides her desire for Partonopeu by claiming to prefer the Sultan, as if to obtain her goal through what we would nowadays label reverse psychology. She seems convinced that her barons will not give her the husband she desires unless she deceives them one way or another.

Deception, Judgment, and the Role of Gender Melior's reversals of role and identity are vital to the romancer's own play with literary gender and genre. In the already fabulous representation of a medieval Byzantium, they introduce another level of the fabulous, the deception of fiction that is the very lifeblood of medieval romance—and perhaps just as necessary in the realm of reality. Deceptions, sometimes even outright lies (as in Urraque's "fause noveile") are needed to keep not only the story but history going as well. Consider in this light once again the narrator's praise of engin as demonstrated in the genealogy by Eneas's mother. We might even go so far as to say, based on the story of Adam and Eve in the Garden of Eden, that history cannot start until deception takes us out of the innocent, but ignorant cycle of nature and opens the door to knowledge, sexuality, and the succession of generations. As in the Biblical story, the agent of change, whether for good or for evil, is fre-

Gender and Genres in Partonopeu de Blois

139

quently figured as a woman (cf. the medieval play on Eva/Ave linking Eve and the Virgin as the corresponding agents of sin and redemption).49 Thus Melior will remove Partonopeu from the forest of Ardenne through her magic deceptions, so that she can educate him, prepare him to be her husband, and enjoy considerable pleasure in the process. But Partonopeu, too, will have to learn how to make his own choices, discern good and evil counsel, and earn his (re)union with Melior. Education and experience are crucial to Partonopeu's (or anyone's) progress. The Prologue tells us that we should imitate good examples and avoid bad ones, but once we have the power to discern and choose, our choices may not always be so simple (cf. Lancelot's choices between competing goods in the Charrete). Deception in the wrong context may be evil—witness the betrayals of Anchises and Partonopeu. But truth may not always be the right choice either. We have seen that ladies, particularly ladies in love, are a special case when it comes to truth-telling, since any welcoming face ("bel semblant," v. 8401) they might show, even to a brother, is subject to people's penchant for gossip and scandalmongering. Deception is for them a necessary fiction to survive in a courtly world, as demonstrated by the romance's narrative and validated by the narrator's commentary. But even more problematic for both men and women is the shifting nature of the truth category itself, as described by the narrator in an elaborate description of love's paradoxes (w. 9199—250). Although his intervention is occasioned by a description of Melior's need to hide her true feelings from the judges (w. 9181-98), the narrator leaves far behind the specifics of her state, as he explores the conventional antitheses of love. The long list culminates in the observation that love's madness makes each one believe his lady to be the best. In this final twist, the vocabulary of good and evil, truth and deception, and most especially choice, is linked to the inseparable coupling of beauty and love. Se Ten preist par sens amors, Por bel[t\es ne por bones mors, Une seroit de tot amee, La plus bele et la plus senee, S'en sordroit malz, et par envie Ruistes meschies et grans folk. Por cest m'est vis qu'il est molt mielz Qu'on soit al choisir engeigniez

140

Chapter 4 Et qu'on les aint en tos parages, Beles, laides, foles et saiges. (w. 9233-42; emphasis added) (If one chose to love sensibly according to beauty and good character, one woman would be loved by all, the most beautiful and the wisest. From this would result evil and, through envy, violent mischief and great folly. So it seems to me that it's much better that a man is deceived in choosing and that women of all sorts are loved, beautiful, ugly, crazy and wise.)

When the narrator speaks here in terms of gender roles, we see that he has shifted to his own, more typically male point of view to describe how love makes the lover blind and thus usefully permits a variety of different ladies to be chosen as the object of love. Beauty being in the eyes of the beholder, each lover believes his own hyperboles, regardless of the lady's actual appearance. Ladies, love and truth-telling once again appear as a special case, but we are seeing it now specifically in relation to the male's choice. The narrator maintains that if one chose in love according to reason, then the basis for the choice would be determined by the lady's beltes and bones mors—that is, the very factors that determine Melior's choice of Partonopeu. Reason's choice would lead, moreover, to the troubling situation in which all lovers would love the same lady. Luckily deception usually solves the impasse, diversifies the choices blinded by love. Each lover, engeigniez, believes his own hyperboles, regardless of the lady's appearance to others. But then which comes first, beauty (as the lyric motif would have it) or love? And how do we get to an objective truth, the choice of reason, if each of us sees with the eyes of love? The passage seems to assure us that reason's choice would lead to a truth that certainly exists: there is one lady with more beauty and good character than all the rest who deserves the love of all, although it seems to lie beyond the power of human discernment to recognize who she is. But Love's deception may serve a good purpose. If we could all recognize that one lady, such a singular, if truthful, choice would inevitably cause great harm (cf. Melior's effect not only on Partonopeu but on the Sultan). The obstacle to absolute knowledge may be a blessing, since jealousy and hate would lead directly from unanimous choice to violence and folly. In the face of this subversive reflection, the narrator goes on to relate these generalities to his own situation: Ce doi je plus que nus valoir Que se li chois alaist al voir

Gender and Genres in Partonopeu de Blois

141

Par bones teches, par bonte, Par grant sens et par grant belte, Chascuns vorroit la moie amer, Car el est el siecle sens per De franchise et de cortoisie. Li veraiz Dex le beni'e! (w. 9243-50; emphasis added) (I should value this above all since, if the choice went truly according to virtues, goodness, great sense and great beauty, each would want to love my beloved, for she is without peer in the world for nobility and courtesy. May the true God bless her!) He benefits more than anyone else from love's blindness, since any reasonable choice would logically fall on his lady alone—may the true God bless her! His exclamation suggests a need, at this critical moment, to associate God's unquestionable truth with his own choice in love. But by his own reasoning, the narrator has already undercut the absolute truthfulness of his statement: can we take seriously the idea that his lady is the most beautiful? Or is this rather a proof of the sincerity and strength of his love, since it is love's blindness that exaggerates the lady's beauty? The problem of deception will not go away, as the narrator continues to praise his lady's perfection and the difficulties that causes for him when it comes to her choice in love: Will loving him subtract from her perfection? If only her mirror could lie to her a bit, he believes, then she might require less beseeching (w. 9251-66). The relation between truth and deception is here intimately tied to the issues of love and beauty, as they once again appear through the dual perspectives of male and female lovers. As elsewhere in the romance world, truth is intimately tied to point of view. This narratorial intervention has significant implications for the culminating event of the romance, that is, for the judgment that determines not simply who are the six finalists but who is the best among them, the only one who deserves Melior's hand in marriage. I have already described the anomaly that transforms the tournament into yet another form of beauty contest for male participants: once their prowess suitably demonstrated earns them the right to qualify (the equivalent of talent in today's Miss America pageant?), beauty revealed as their armor is removed becomes the final determinant to decide between the Sultan and Partonopeu. I quoted earlier Melior's justification of such a procedure, based on the way men typically choose their wives, but we should look further at

142

Chapter 4

the suggestion made by Corsolt, to which Melior is responding. Corsolt is certain that if Melior requires all the judges to give their opinion (six of them have remained silent so far), according to the truth as they see it, and uninfluenced by love, money, or fear, they will undoubtedly choose Partonopeu—or let Melior's own eyes see the truth of that choice by direct observation: "... par vos oilz sachies de voir

Le quel vos volrois melz avoir, Assez nos sunt andui loe, Mais veons en la verite" (w. 10381—84; emphasis added) (". . . with your own eyes, know truly which you would rather have. Both have been much praised, but let's see the truth of it.") Melior accepts both suggestions, first supporting by male example her willingness to compare the contestants' beauty, secondly requiring the judges to guide her choice, since she will follow "lor esgart" (v. 10400)— that is, their judgment as formulated by the act of seeing. Melior picks up Corsolt's invitation to judge the truth with her eyes, but she adds the corroboration of her heart: "cOd mon cuer jugeront mi ueil'"(v. 10398). She thus acknowledges the heart's role in such evaluations and echoes, however briefly, what the narrator has already told us about love's blind choices. Will beauty cause love or love beauty? Of course, we know that Melior has already loved Partonopeu, has already chosen him for his beauty, and we fully expect the happy outcome that follows. But what we notice in the romance's realization of that outcome is how much Melior insists on the judges' participation even in the final choice, although the conditions originally set up by Ernol gave her freedom of choice among the finalists. It is as if Melior insists (by implication) that her own subjective truth, conditioned by love, be protected—or is it verified?—by a more objective truth available to the judges. Public agreement should thus ratify private choice. The judges' access to truth here should devolve from their dispassionate stance. They are, in this context, unaffected by the blindness of love, but the persuasions of money (conjured up by the sultan's riches as described by Clarin) have apparently led six of them to prefer the Sultan, despite the barons' original instructions to Melior to choose according to

Gender and Genres in Partonopeu de Blois

143

other, more personal criteria, lest they find themselves selling their Empress to the highest bidder (cf. the ric ome debate in troubadour lyric). The judges' hesitation to choose is quickly solved by Partonopeu's appearance, as everyone acclaims his beauty superior to all others'. Apparently, male beauty is less contested than female beauty, more liable to lead to consensus (at least when the judges are themselves men). If we try to summarize and draw out further the implications of these events and their representation—since the Prologue invites us to recognize good and bad examples and make our choices accordingly— we might first see the romance as a resounding argument in favor of beauty as the tangible sign of birth and lineage. The blood line establishes value, passes it on from generation to generation, whether for good or for ill. One's proper place is determined by the dual coordinates of nature and society: the order of society should mirror and align itself with the order of nature. Hence the negative example of the "fils a vilain": you can make a big mistake, as Priam, Sornegur, and the Roman emperor did, if you forget a person's lineage and elevate the individual out of his place. The bad effects of low birth once dislocated will lead eventually to betrayal. On the other hand, if you choose only according to beauty as a reflection of lineage, you may be forgetting something equally important: Melior's choice of Partonopeu based on the sole criterion of beauty in the first half of the romance leads no less surely to his betrayal, even if that betrayal may be excused in part as the result of evil counsel. One's place may be marked out by the parallel orders of nature and society, but the slot thus opened must still be earned and secured by individual effort and achievement. The second half of the romance with the tournament as beauty contest demonstrates what is to be gained by combining lineage with individual worth as demonstrated through performance. Ultimately Partonopeu's beauty stands for both sources of value as inextricably linked in a more fully heroic identity. Although prowess alone is unable to single out a clear winner, individual performance is as necessary as lineage for choosing the best with the agreement of all. That is precisely what Partonopeu's own beauty comes to embody. If the narrator's overt commentary on the "fils a vilain" theme makes the argument for lineage a strong and insistent one, the complexly interwoven threads of the beauty theme implicitly encapsulate both birth and individual achievement as sources of worth: it is up to us as wise readers to unravel and understand the more comprehensive example of Partonopeu, as inscribed in the romance as a whole.50

144

Chapter 4

The Mobility of Categories If we continue to follow the good example of bees, whose ability to turn one element into another is no less dear to the author of this romance than it was to Montaigne, then we will remember that the preparation and selection of Partonopeu as winner of the tournament and husband for Melior was accomplished through a whole series of experiments in gender and genre crossing. These role reversals, shifts, and recombinations operate in the service of plot-building and character development, but they also have an important effect on the way we begin to think about all sorts of categories and their definitions. In the course of reading this romance we may come to believe that certain categories established by nature and society are not really as apriori nor as stable as the genealogy would imply51 They may be subject to disagreement, as we see explicitly in the debate between Partonopeu and Anselot on the equality of all versus the evils of the "fils a vilain" raised out of his normal place in society (w. 286-3ioc). The two argue briefly for and against the notion that we are all equal as sons of Adam and Eve—an equality that would imply social mobility open to all. Specific examples are used to argue both sides of the question until Anselot's reference to Partonopeu's own history (in the Sornegur episode) ends the debate, as argument is replaced by the friends' recognition. The treacheries of Mares seem to throw the weight of example against equality or social mobility for the "fils a vilain." But the implicit strategy of the romance, as well as some of the narrative commentary, seems designed to open and complicate this debate, at least if we enlarge its scope beyond the specific problem of the "fils a vilain." If, as the romance politics suggest, we stay within the social category of the nobility and its values, there still seems to be ample room for discussion about what belongs there and how the members of the category relate to each other. As indicated in the narrator's commentary on beauty and love, the way we determine and qualify certain categories is inevitably linked to the role of judgment, which in turn is related to point of view and our position inside or outside. This is as true in the story as in the "real" world represented by the narrator's frame story, the intermediary linking the worlds of romance and public. Our personal engagements change what we see, make us see differently from each other. As in the Charrete^ the diegetic and extradiegetic worlds mirror each other, offer the same model for the reader/listener. We might usefully remember here that among the romancer's innovations is his double stance, at once subjective and objec-

Gender and Genres in Partonopeu de Blois

145

tive, which constantly affects how he sees and represents the story of Melior and Partonopeu. He understands and teaches us about the effects of love, for example, because of his own experience in that domain. The mix of objective or didactic commentary and subjective identification makes us wonder to what extent we could ever disentangle the two, if indeed we would want to do so, since the romancer's insights are presented as linked to his involvement in love.52 Consider in this respect another long intervention from the narrator, this time on what he believes to be the proper conjunction between beauty and love, chastity ill-befitting a beautiful woman (w. 6261-90). According to the narrator's view, ugliness and beauty stand at the antipodes in a world where beauty not only describes physical appearance but embodies all the courtly values endorsed in the romance. To throw ugly chastity into such a world is to disrupt its "natural" affinities: "Car chaste femme o grant bealte, / Trop i a grant mal assemble" (w. 6261—62: "For a chaste woman with great beauty is a terrible mismatch"). Whereas chastity is avaricious, refuses to laugh and play, beauty loves sweet talk and welcoming smiles, nobility and generosity, prowess and honor, courtliness and courtesy. Chastees soit et noire et losche. La bele soit et blanche et bloie, Et vive toz jors en grant joie. (w. 6282-84) (Let chastity be black and ugly. Let the beautiful woman be white and blond and live always in great joy.) Although the narrator calls on God to keep the world well ordered so that beautiful women will not be chaste and chaste women will be too ugly to attract lovers, the clear separation of categories into black and white, chaste and beautiful, does not hold. The mismatch of beauty and chastity is precisely the disruption the narrator decries here in his lady's refusal of love, since she provides the example from which his generalizations flow. When he speaks low, she speaks loudly, ignores his sighs, and refuses to accept his tokens of love (w. 6285-90). Her "contrariness," always the opposite of his desire, is eloquent testimony to the disharmony caused by her inappropriate conjunction of what the narrator takes to be opposites, beauty and chastity. If we take this complaint out of its immediate context in the love story of the narrator, what it suggests more generally is that certain qualities

146

Chapter 4

should go together in the same category, but they sometimes do not. There is a problem in alignment here that upsets not only the narrator's personal feelings but what he considers to be the proper order of nature as expressed in the social interaction of man and woman. If there is so much play in the system that beauty, for example, may not always carry with it the proper constellation of values, then the category itself may be a mobile one. And if the categories move or shift in value, then we must accumulate the knowledge that will give us the power to recognize, evaluate, and choose properly. We thus return inevitably to the narrator's injunction to learn from examples. Beauty finds a fitting expression in Melior and Partonopeu, especially in their final successful and public union, achieved as soon as male beauty and prowess, lineage and individual worth, have been shown in proper alignment. But the series of switches in gender roles that characterize their relationship at different stages of development, as well as the multiplication of heroes that takes place in the tournament and Continuation, suggest we should raise questions about the appropriate order within the couple, just as we need to test the suitability of any individual to fulfill the role promised by his birth. If we consider the second part of this proposition first, we might notice how the tournament's arrangements and the selection of finalists imply that any number of candidates may make a suitable husband for Melior. This substitutability is fostered by the narrator's tendency, as he recounts the tournament, to present the Sultan as Partonopeu's counterpart. The degree of interest manifested in Margaris, not only by the narrator (e.g., w. 9553-58) but by Partonopeu himself, who constantly admires his opponent and fears that he may be the better knight (e.g., w. 9449-56, 9563-74), seems to prepare our interest in the Sultan as a major character, even a rival hero, in the Continuation. Just like Partonopeu, Margaris is motivated by love of Melior; the prowess of both is inspired by their love (w. 9553-58). Of course the tournament itself is designed to show the differences among the various contenders for Melior's hand, but once again the notion that Melior will choose from a group of six, or finally, two finalists introduces the idea of substitutability at some level.53 Although the exterior circumstances seem to allow a free choice among candidates, we know all along that Melior's choice is by no means free, or rather it has not been free since her original choice based on Partonopeu's beauty. As he describes Melior's reactions to the initial silence of the judges, as well as Clarin's argument for the Sultan, the narrator

Gender and Genres in Partonopeu de Blois

147

comments on her fidelity and, in general, the loyalty of ladies in love (w. 10109-16): For ce aim je dames com moi, Que en eles a si loial foi; N'est gaires nus horn ne changaist For coi amender se quidaist. (w. 10117-20) (For this reason, I love ladies as I love myself, for there is in them such loyal faith. There is scarcely a man who wouldn't change ladies, if he thought he could find a better one.) In the question of fidelity and choice of the beloved, the narrator's commentary distinguishes between men and women. The links between romance and frame supply suitable examples. From Melior's point of view, you cannot substitute the Sultan for Partonopeu; her choice is unshaken by any of Margaris's virtues, whether in prowess, riches, or personal attractions. But can we say the same about the narrator? To be sure, as lover, he claims to be ever faithful to his lady. But as narrator, he succumbs to the charms of the Sultan and displays the kind of novelerie he has just decried in men (cf. Thomas's narrator on human changeability). Such novelerie may be only hinted at in the narrative of the tournament; however, his desire for multiplication appears directly in the substitution of heroes announced in the first epilogue: there he offers to tell the love stories of Anselot, Gaudin, and the Sultan. In the actual Continuation, Melior and Fartonopeu will appear, but more as incidental figures than as principal characters; we hear instead the inconclusive stories of Anselot and Margaris, neither one furnishing much happy closure in comparison with Melior and Fartonopeu's love story. If from the point of view of the main characters, the choice of Fartonopeu as Melior's husband is the right one, from the perspective of the storyteller, love for Passe-Rose requires him to seek substitute heroes through whom he can continue to explore his own emotions, his prospects of success or failure in love. The analysis of individual substitutability required a crossover between story and frame; in complementery fashion, the complex balance of power between writer and beloved points us logically back to the story once again, so that we can analyze further how it explores the proper order within the couple. We might read Partonopeu as an example or rather a whole set of examples posed to answer the question: should the relation-

148

Chapter 4

ship between man and woman be hierarchical with all the power concentrated in one gender:1 The crisis of Partonopeu's betrayal demonstrates the instability of such an arrangement, at least if all power is held by the woman. Should we read the second half of the romance as an argument for stripping Melior of power, properly subordinating the woman to male rule, as required by St. Paul, for example? This same authority is cited in the Prologue to guarantee the edifying character of all writing, but if the romancer seems to stretch St. Paul's authority to give a sort of carte blanche to the pleasures of fiction, we may be equally authorized to stretch the romancer's play with gender and genre to suggest that the overall strategy of Partonopeu tends to argue—at least by implication—for a more equitable exercise of power between man and woman as two members of the same species, rather than for any simple reversal of Melior's initial hierarchy. We might remember in this context the equal exchange of "flowers" at the moment of their sexual initiation (w. 1305-8), which offers the model of a natural or biological equality for the sons and daughters of Adam and Eve that might be recreated in the social dimension. On the surface, however, we never see a clear statement for or state of equality between Melior and Partonopeu, even though the closing descriptions of their wedding and celebration establish a kind of parallel in their activities. Our last view of them in the romance proper is a short description of Partonopeu's joy now that he has "s'amie" (w. 10602-6). Throughout the preceding events, there is a continually moving crisscross in the balance of power between the Byzantine Empress and the French knight. The apparent retreat in Melior's use of power in the second half of the romance, as she moves from undercover magic to deceptive appearance, seems to be a necessary withdrawal, if Partonopeu is to assume any power at all in their relationship. In the feudal pyramid, no exercise of power is without the give-andtake of contractual bonds. The question is where in that structure will the male-female bond be located, on the vertical dimension between lord and vassal or the horizontal one between equal members of a given category? The romance offers no simple answer to that question, suggests rather that any given relationship may move around in the model, may occupy one site or another, depending on the stage of development, the point of view taken, or the aspect to be examined. Consider in this light what happens in the Continuation. On the one hand, we notice a "normalization" of Melior's role as woman in medieval society, since she is now clearly subordinated to Partonopeu as Emperor. When news of the Sultan's invasion

Gender and Genres in Partonopeu de Elois

149

arrives, he is the one who calls the barons together to solicit their advice. But it is worth noticing that Melior is consulted and respected among the advisors as one among equals. When she offers a plan to delay the impending attack, Melior is still using engin as a successful strategy with Margaris. If we observe a kind of normalization of the male/female balance of power, once Melior and Partonopeu are married, we should also take into account a shift in focus: we no longer have a view of the internal dynamics between Melior and Partonopeu, but see only their public roles as Emperor and Empress; their private interaction and the nature of their relationship per se is no longer the main focus of interest. For a focus on the inner dimension, we have to turn to other heroes and heroines in the Continuation and there we may see a multiplication and splintering of the elements earlier in play between Melior and Partonopeu, as the narrator continues to introduce women exercising power with varied results. The reversed hierarchy of lyric appears now in the Sultan's courtship of Melior, while the Anselot-Euglar relationship plays off against Melior's earlier persona from the lai as the outspoken lady who takes the initiative in love.54 The narrator's own lady is evoked more and more in a role that resembles Melior's vis-a-vis the Sultan. This may be a reflection of the way the narrator himself seems to take more and more of the limelight in the course of the Continuation—and he, unlike Partonopeu, has no happy ending to report for his own aspirations in love, which hover perhaps between Anselot's exile from a lady who loves him (he still hopes) and the Sultan's definitive rejection by Melior.55 The narrator's virtuoso performance in rhetorical play, his exploitation of different literary styles and genres, seems to displace our primary interest from the "what" to the "how" of the stories told. Of course, this is also true to a certain extent in the romance proper, where we are expected to follow the experiments with gender and genres, but the rate of switching seems to accelerate in the Continuation. In Anselot's meeting with Partonopeu we see the narrator playing with the form of school debates, while Anselot's story itself is a kind of lai told in the first person. When the narrator changes to rhymed alexandrines for the epic part of the Margaris episode, his intervention about the difficulties of such a form, willingly assumed to increase the beauty of the ending and respond to his lady's command (w. 1463—73), insistently calls attention to the writer's role and the story used as vehicle for his own love. We may understand in this light why Margaris becomes a writer in the Continuation, as if the narrator projects himself more and more into his character—or perhaps we should say it the other way

ISO

Chapter 4

around?56 It is not without significance that Margaris shows his clerkly talents more than his chivalric power when he returns as Melior's suitor. As the author of two poems, one concentrated in two couplets, the other a long salut d'amor (124 lines), Margaris is as concerned as the narrator himself about the quality of his product and of his public. Indeed, the narrator seems as proud of the poetic results as if he had written them himself (w. 3O46-58c). Just as the narrator figure's combination of lyric subjectivity and clerkly distance moves him closer to his characters by identifying his love story with theirs, so in the Continuation the Sultan Margaris's representation as clerkly lover moves him closer to the image of the narrator.57 This interaction between character and narrator serves to make the writer figure emerge as one of the heroes of Partonopeu^ the beauty of whose art, as finely polished as his verses, is the fitting mirror in which to see Melior's and Partonopeu's incomparable beauty. The substitutions, whether within the story or between story and frame, keep in motion the interplay between male and female power. The effects of the romancer's fusion of genres, his crisscrossing of categories, allows the constituent elements to mix and interact, to invent new patterns without effacing the old. Most importantly, his example invites us to think experimentally, to try out new configurations that may lead to new definitions and new choices.58 I do not intend to make this anonymous romancer into a feminist defined by twentieth-century standards, although the narrator does frequently picture himself in his interventions as identifying with ladies, their point of view, their actions, and their values. But I would suggest that his romance is less politically conservative than Fourrier argues, if not immediately in the choice of non-noble advisors as articulated by the narrator persona, at least by implication in the romance as a whole, if we consider the role played by the (noble) individual within the couple, or in the larger context of family, lineage, and the body politic—and this seems to apply to female individuals, as well as male. I would argue that the romancer's experiments in fiction, as inscribed in his text, are not limited to esthetic effects, that his play with the crossing of history and romance, truth and fable, gender and genre, leads, on the one hand, to the self-empowering of romance as its own source of authority and, on the other, to romance as a powerful tool for change in the (dis)course of history, as it shapes vernacular society's views of itself. In comparison with a romance like Chretien's Charrete, which asks questions about the ultimate paradoxes of human life in the Arthurian

Gender and Genres in Partonopeu de Blois

151

world, Partonopeu de Blois may strike us as more light-hearted fare, appropriately Byzantine in its love of beauty and the values of the visible world, even as they toy with invisibility. Here are no unresolved paradoxes; we enjoy rather the pleasure of momentary puzzles, which each in turn piques our curiosity, engages our sleuthing skills, and ultimately yields its solution before our appreciative and satisfied gaze. Compare in this respect the different outcomes of the tournament episodes in these two romances, the secrecy that continues to separate Lancelot's public and private identities, the disruption of marriage plans by hidden love, versus the publicity and social approval that attends Partonopeu's final triumph and choice as Melior's husband. Unlike Guenevere, who, in the Charrete's closing scene, must hide her desire to greet the returning Lancelot with all the joy of her love, at the end of Partonopeu, the Empress of Byzantium makes no scruple about kissing her beloved, even if a thousand knights witness the scene (w. 10523-30). In this sense, Partonopeu allies itself with the idyllic romances from the East, like Floire et Blancheflor, themselves a subset of what Dafydd Evans has designated as wishfulfilment romances.59 Unlike the tragic model of romance based on the Tristan type, wishfulfilment romances achieve their happy endings by combining love and marriage, despite the incompatibilities of the two within a feudal society that sees marriage rather as a tool for the political and social gains of the lineage. Evans identifies the use of fantastic adventure and the supernatural to reconcile personal desire and social exigency as clues that signal the unreality of such fictional solutions.60 But who knows to what extent such fictions, precisely because of fiction's greater freedom of invention, became models for different views on love and marriage or the relation between the individual and society? Romance is not merely a passive reflector of contemporary history, and Partonopeu seems designed at will to invent new fusions of history and romance, reality and fiction, as we enjoy the pleasures of both and the limits of neither.

The Poetics of Continuation We might compare Partonopeu in this respect to the erotic pleasures of love-making—a not inappropriate metaphor for a romance whose action begins with an unusually graphic description of its hero and heroine's initiation into love. Unlike the more theologically oriented sex-for-

152

Chapter 4

procreation (which we might locate here in the genealogy), the characteristic rhythm of sex-for-pleasure is the repeated and prolonged rise and fall of tension; the desired end is not ending at all, but rather momentary pause and prelude to a new beginning. There is, for example, a pleasurable rise in "tension" when Melior nearly recognizes Partonopeu during the dubbing ceremony before the tournament. The ultimate recognition, however, is deferred and then doubled during the tournament itself by the increase of participants, as first Urraque and her lady Persewis and then Melior herself finally recognize the unknown knight as Partonopeu. Partonopeu's gift for amplificatio as a way to prolong our pleasure is nowhere more evident than in the romance's transformation of what is essentially the simple plot of a te', whose kernel has been enormously elaborated by fusion with a diversity of materials and traditions. Despite its length, Partonopeu^ narrative structure retains the concentration of a lai—at least one like Guigemar, even if it does not demonstrate the singleness of action we see in a Laustic or a Chievrefoil. We might compare, for example, the relatively few episodes that make up its plot with the multiplicity of episodes in a romance by Chretien. It is the episode itself that expands here—the three-day tournament narrated over thousands of verses may be considered exemplary in this respect, as it occupies the entire second half of the romance and furnishes the equivalent of the knight's quest, usually elaborated in a series of different adventures.61 But Partonopeu does illustrate the usual romance pattern in its overall structure and, like Chretien's romances, it grows through a series of analogues as its different parts anticipate, repeat, and reflect on each other. Consider the returns to the Ardenne forest: the structure of the Continuation, which introduces the "digression" of Anselot's story within the narration of Margaris's invasion, takes us back to the moment when Partonopeu eluded Anselot in the forest—that is, it recalls the crisis episode in the middle of Partonopeu's story, when he returned to the very place where his adventure began in the opening hunt with King Clovis. There in the Ardenne Partonopeu's story starts again, once he is found by Urraque, and proceeds through the tournament to the happy ending. In a similar pattern, the Continuation introduces a loop in the narrative: Anselot's spin-off story, begun much earlier at the point where the friends part in the forest, is resumed when Partonopeu finds Anselot during a hunt near Chef d'Oire (a recall of his daily pleasures during the secret sojourn); through Anselofs own retelling, his story is now taken from the Ardenne to its unhappily suspended conclusion. The digression first inter-

Gender and Genres in Partonopeu de Blois

153

rupts and then returns us to the Sultan's invasion, as it replays Partonopeu's own story with a different ending. To the analeptic recapitulation of the main plot that we can thus discern in the Continuation corresponds the genealogy's proleptic variations, which anticipate crucial elements of the major section, as it introduces the themes of East-West movement, the benefits of engin, and the "fils a vilain" versus beauty and blood. These analogues, with their repetitions and variations, remind us again of the romance's erotic rhythm, which desires both ending and renewal of pleasure. But the rhythm of deferral does not only characterize the development of events within the romance; it evolves into what we might call a poetics of continuation. Partonopeu ends the first time with the marriage of its main characters. This "natural" endpoint for a romance that has already gone on for a considerable number of verses (a romance and a half by the standard measure of Chretien's) is not, however, to be taken as any indication that there is no more to tell. A romance whose geographic and moral center is occupied by Chef d'Oire clearly locates us in the more ample dimensions of the romans antiques. In a 49-verse epilogue, the narrator explicitly introduces the idea of continuing his romance. He even enumerates the possible strings of narrative to be picked up, events he briefly summarizes as the stories of three secondary characters (w. 10625— 42). With these short descriptions he no doubt hopes to entice his lady and lure her into further romance, as Melior once lured Partonopeu. If the lady agrees that his meritorious service in writing deserves a favorable recognition, a mere wink from her will set him to writing again, but for now his love pain is so great as to interfere with his power to work (w. 10609-24). While the link between love and composition is a commonplace of troubadour lyric, where lovers often complain that their suffering makes song impossible (and yet they sing on heroically), these comments are truly remarkable as a first for romance. A survey of twelfth-century romance epilogues yields not a single other announcement of the sort that promises there is more to the story as yet untold, until we come to Renaut de Beaujeu's Le Bel Inconnu (which may be late twelfth or early thirteenth century). On the contrary, the epilogue generally functions as an elaboration of the explicit and usually includes among its topics the affirmation that there is no more to tell. In Tvain Chretien even goes so far as to equate additions with lies (w. 6804-8).62 Although Chretien's own Conte du Graal already gives rise at the end of the twelfth century to the first of

154

Chapter 4

its many continuations, it does so by virtue of its unfinished and puzzling form, rather than by any explicit, authorized promise of more to tell. The innovative Partonopeu does not simply participate in the typical romance pattern of the twelfth century, where romances generally remain separate and discrete, and only implicitly acknowledge, through their own discontinuity, the romance tendency to proliferate and continue. Partonopeu boldly announces the possibility for continuing its story and thus prepares the way, along with Chretien's own experiments, for the combinations and cycles of romances that will become so popular in the thirteenth century. One facet of Partonopetfs innovation here can be grasped in the way the narrator ties the continuation to his own love story, once again intermingling story and frame. When he claims to know more and thus turns the marriage ending into a momentary suspension of action (w. 106078), the narrator uses a ploy dear to Melior: he withholds knowledge from his beloved to gain power and control as much as possible the other's conduct. While the narrator's insertion of his own love story into the matter of romance offers a model for the variations explored in LeEellnconnu and Florimont, he seems to go further than any of his followers in anticipating, on the one hand, the fusion of lyric and romance we will see in the Roman de la Rose, when the lyric "I" becomes the hero of his own romance and, on the other, the multiplication of heroes that characterizes the prose cycles and the verse romances of multiple quests.63 Melior provides the romancer with a precious model, as well as an indispensable tool, for shaping and prolonging his romance. Her manipulations of appearance and identity, however feminine and passive they seem after the crisis of betrayal, continue throughout the romance to supply the matter for yet more story, just as her power of invisibility furnished the starting point for a beautiful and marvelous adventure. In terms of building a story, the "problem" of Melior's invisibility—why does she prohibit Partonopeu from seeing her?—is clearly what keeps the plot going in the first half, despite the lovers' repeated enjoyment of ultimate joy (the usual ending for a love story, not its beginning). Likewise Melior's later refusal of pardon prolongs their story over many more verses before arriving at the predictable romance ending of marriage—already predicted by Melior herself in the first explanation to Partonopeu and reiterated with suitable variations during the scene when Urraque tries unsuccessfully to persuade Melior to forgive Partonopeu and go ahead with her marriage plans. But if Melior pardons Partonopeu right away there is no more to tell—or not much: already during his sojourn at Chef d'Oire most of the

Gender and Genres in Partonopeu de Blois

155

narrative reports time passing in great chunks of a year or six months, Melior's explanation and warnings about the prohibition furnishing most of the "story matter" for those passages. There needs to be some tension between the lovers to spin out a plot, and luckily for the romance public's pleasure a precipitous ending is avoided thanks to Melior's skills as a creator of fiction. This is equally true for the Continuation, since it is her deception that permits the Sultan's misreading of her feelings and motivates his later return as an invader and suitor. As soon as Melior stops giving ambiguous messages, the Sultan's story runs out of matter. In the Continuation of ms. T, after the digression of Anselot's tale, the Sultan's attack is first delayed by a message from Melior that seems to suggest that she still loves him and seeks an agreement that will bring them together (w. 1356-67^ i683-i704c). All the Sultan's hopes—and his story—stop quite abruptly, however, as soon as Melior reveals her true feelings. She has no more desire than Fenice to play the role of Iseut, shared by two men (w. 3894-96^. Once this truth is revealed, the Sultan departs and the narrator hastens to the epilogue with his insistence that there is no more to tell about his characters (w. 3908-9, 39i5c). In some sense then, Melior offers the romancer both a way to finish his romance and an excuse for continuing it, as the occasion warrants. Melior bases part of her original refusal to pardon Partonopeu on her horror before the prospect of an open-ended series of mesprison: if he falls again, after being pardoned the first time, Melior can but anticipate more and worse faults (w. 4987—92). But the romancer shows himself less frightened of such openness than his character. In fact, the second epilogue of ms. T closes with his announcement that he could write yet another book, this one about the lady herself (w. 3927-3oc). In his projection, which functions more as praise of Passe-Rose than real promise to write more, the narrator anticipates the moment when he and the lady could finally become the main characters of their own book. The romance thus ends again with the evocation of more—a possibility that is fulfilled in certain respects by the manuscript tradition itself: not only did scribes continue to rewrite and rework Partonopeu in French; medieval translators repeatedly imitated and multiplied the romancer's initial translatio from Latin into the vernacular, by making it available to readers across Europe, throughout the thirteenth and fourteenth centuries and even well into the age of printed books.64 Enormously popular, Partonopeu set a standard for beauty and pleasure, for experimentation in form and fusion, that became exemplary for

156

Chapter 4

the romancers who followed. It dazzles us still with its capacity for reinventions that both entertain and educate, as it opens the door to our own ability to recognize, evaluate, and choose. The impetus for continuation, so strongly set up and elaborately staged in Partonopeu, invites our own continued pursuit through analysis of the problems and possibilities represented in the romance's dynamic crisscrossing of gender and genres. As they are played out in the story of the narrator and his lady, no less than in that of Partonopeu and Melior, we may see how romance can motivate and put into motion the kinds of questions that may ultimately be continued in medieval society's own meditation on and transformation of categories, whether cultural or social, literary or philosophical.65

5. Textual Identity and the Name of a Collection: Marie de France's Lais

Marie de France was not the only twelfth-century author to have written Ms, even if she may have been the first; nor was she the only collector of Breton lays—the editor scribe of ms. S and the thirteenth-century Norse translator did the same.1 What is unique to Marie's project, and what radically changes its effect, is her combination of both roles. When she writes and assembles a group of twelve lais, presented explicitly in the General Prologue of Harley 978 as a collection formed under her own authorial control, she interwines significant characteristics and issues related to oral storytelling with some of the most fundamental problems, techniques, and resources identified with romance-writing. She thus asks us to read not only the individual lais but the collection itself as the product of her art. This achievement in creating a kind of mini-cycle of Ictis is unparalleled in vernacular writing of the twelfth century and anticipates, on the one hand, the combination of romances into cycles that we see, for example, in the thirteenth-century Lancelot-Grail, and on the other, collections of tales that appear much later, like Boccaccio's Decameron or Marguerite de Navarre's Heptameron, arranged by their author/collectors.2

Lai and Romance Marie's arrangement, as presented by the General Prologue and realized in the collection itself, asserts the kind of authorial control discussed in Chretien's Prologue to Erec et Enide: Chretien there criticizes the untidy and unartful habits of oral storytellers, whose actions are associated with verbs of destruction ("depecier et corronpre," v. 21: "take apart and corrupt"). At the same time, he implicitly praises his own artful ability to draw out from a story of adventure "une molt bele conjointure" (v. 14: "a very beautiful conjoining"). Marie naturally refrains from such a negative

158

Chapter 5

view of storytelling, the very tradition in which she explicitly places herself, as she retains certain of its discontinuities. But she does show a similar concern to bring together through her own art of verse-writing the same story materials described by Chretien, those offered before counts and kings (Erec, v. 20): Rime en ai e fait ditie, Soventes fiez en ai veillie! En 1'honur de vus, nobles reis, . . . M'entremis des lais assembler, Par rime faire e reconter. (P, w. 41-43, 47-48) (I have put them into verse, made poems from them and worked on them late into the night. In your honour, noble king, . . . did I set myself to assemble lays, to compose and to relate them in rhyme. P-4i) 3 Marie associates this project with a proper name, not the King's to whom she offers her collection, but rather her own. Just as Chretien proudly states his name in the Prologues to Erec., Gliges, and the Charrete to establish the link between author's name and textual tradition, the guarantee of auctoritas transferred from Latin to vernacular writing, so Marie places her name before her collection, assuring us that all the stories to follow have been shaped by her care and artistry. Marie's name, once associated with the Lais, becomes as evocative as the titles and proper names linked to the original adventures.4 Her mastery of prologue topics, like Chretien's, proclaims Marie's rhetorical competence, while her use of the patron topos—usually praise for the instigator of the work—demonstrates her own remarkable initiative. By citing the auctoritas of others, she makes claim by implication to her own, based on her knowledge of traditions, oral and written, Latin and vernacular. Marie's auctoritas brings to the collection a kind of closure based on the singularity of an author's name: all the diverse materials of the stories have been brought to us by a single memory, a distinctive narrating voice, whose ability to blend with those of her characters preserves their original difference and diversity while including them in the unity of the collection. Marie's literary art, sustained throughout the collection of twelve lais, joins her work to that of the philosopher poets, worthy of glossing and interpretation to locate their meaning (P, v. 16: sen and

Textual Identity and Marie de France's Lais

159

surplus). Any efforts to locate this surplus of meaning—or better, meanings—will have to take into account the intersection of perspectives offered, the plurality of voices and traditions filtered through Marie's narrative, whether at the level of individual lais or in the collection as a whole. The Berne and Oxford Folies Tristan are also short recits that operate in the context of romance, insofar as they retell within their brief form much of Tristan and Iseut's story. Marie's Lais, however, strike up a different kind of dialogue with romance. Marie tells us in the General Prologue that she will not translate a good story from Latin to "romaunz" (v. 30: the vernacular), because so many others have already distinguished themselves in that domain. Once we have read the whole collection, however, her statement strikes us as both true and not true (cf. the rhetorical figure of paralipsis or occultatio, where you claim to pass by what you are in the process of saying). As stated, she is not translating from Latin like the authors of the romans antiques—the romance tradition, the particular form of vernacular writing inevitably evoked by Marie's allusion. But the lais themselves will show that she is borrowing from Latin materials either directly or indirectly through Eneas and Wace's Brut.5 Moreover, her use of the matiere de Bretagne connects her to another written tradition of vernacular romance, to which she may refer later in Chievrefoil (v. 6): we cannot read her Lais without noticing how often she alludes explicitly or implicitly to the Tristan story, while Lanval refers us to Arthurian romance as well. Marie's denial and claim to do something different, not romance, paradoxically invites us to see how in fact Marie rivals with romancers by both playing at and reinventing their game through an intersection with the oral tradition of Breton lais. By adopting the romance technique of the learned prologue, Marie includes herself within its school mentality and processes—despite her disclaimer—and signals to her readers that we should not always take what she says at face value. This is just as true in relation to the storytelling tradition she claims to participate in, as it is in relation to that of romaunz, which she claims to reject. Although explicit interventions of the framing narratives call our attention to Marie's activities as author and narrator, they limit her role to the choice and presentation of traditional material passed along by Breton conteurs. But we know from research done on Marie's sources that her activity in reshaping them must have been considerable, even if we cannot always be sure exactly what she received and what she did with it.6 Anyone who compares the anonymous Graelent and Guingamor with Marie's Lanval, three different lais which rework a common underlying story, can gain a healthy re-

i6o

Chapter 5

spect for the degree of variation the oral tradition can generate (at least as reflected by the written versions that preserve it for us). Marie certainly does not limit herself to oral tradition alone, but combines and brings over into her own work both materia propinqua (already polished) from Latin sources and materia remota (still rough) from the Breton storytellers, in order to effect her own conjointure.7 Two types of "bone estoire" (P, v. 29: "good story55) appear in constant contrast in Marie's Lais in the interplay between conte and romance. On the one hand, the discontinuous nature of the collection, with each lai marked as a discrete entity, preserves Marie's link with oral storytelling; on the other, the continuous nature of the collection (unified by repetition of the framing prologues and epilogues8) situates her through the process of amplification within the context of romance writing. Each lai as a miniature recit demonstrates the concentration of abbreviation the collection as a whole, with its 5,786 verses, begins to take on the proportions and potential of romance. Ille et Galeron, for example (a romance either based on Marie's Eliduc or another version of the same story) has 5,835 verses; Floire et Blancheflor has 3,039, while Chretien's romances typically have six to seven thousand verses. Within Marie's collection, there is considerable difference between the shortest and longest lais, especially in the "point counterpoint"9 of the last two: the juxtaposition of Chievrefoil (118 verses) and Eliduc (1,184 verses—i.e., ten times longer) captures and recapitulates in the collection's closing moments the intertwining of conte and romance. The shortest concentrates in one brief adventure the quintessence of the Tristan legend—the stuff of numerous romances; the last and longest assumes by contrast a romance-like length (as in Guigemar^ the opening lai), complete with episodic narrative structure, conflicts in feudal loyalties, descriptions of combat and battle tactics (the latter not often enumerated in the Lais}. While Marie has been justly admired for her ability to concentrate a powerful story within the briefest of confines, she has received little recognition for the way she concomitantly achieves through her collection the kind of effects we associate with amplification and romance, as well as a level of complexity difficult, if not impossible to achieve within a single lai operating by itself. The great value assigned to amplificatio^ as witnessed by the amount of space its techniques occupy in medieval rhetorics, is echoed by modern scholarship, if we compare the enormous bibliography attached to Marie's Lais in comparison with the critical attention given the anonymous lais, either individually or as a group. Although the

Textual Identity and Marie de France's Lais

161

reader of ms. S's twenty-four lais might see some of the same effects we can follow in Marie's collection, there is no auctoritas, no named author, asking us to do so and guaranteeing through the density or "obscurity" of written textuality the significance of a continuous reading that links individual lais in intertextual dialogue within the collection as a whole. Indeed, such readings have so far been practiced only as a means of identifying the extent of Marie's authorship in relation to the anonymous lais. By naming herself in the Guigemar Prologue and by identifying her activities as author and collector in the General Prologue, Marie presents herself as the architect of a mini-cycle of lais in a way that recalls Jean Frappier's description of the hypothetical author/architect who orchestrates the ensemble of the Vulgate Cycle.10 While recent criticism has put into question the notion of a single, controlling author overseeing the plurality of voices and romances collected together in the thirteenth century prose cycle,11 we do see Marie realizing such an intertextual project in the Lais. In both works we follow a multiplicity of heroes, whose activities and characters are stamped with a family resemblance that constantly invites us to compare and contrast them. Although our actions as readers necessarily cross the formal boundaries of the lais, Marie does not anticipate the interlacing of her heroes' adventures, as Chretien begins to in his later romances. She retains the individual coherence of each te, just as most romance writers of the twelfth century focus their works on a single hero. But her juxtaposition of twelve tales in the collection invites both linear and non-linear readings that distinguish, as well as connect, the discontinuities of separate lais. By elaborating some of her tales through an episodic structure similar to romance (though without the quest as an overarching principle of organization), Marie includes within a single lai the kind of dispositio associative identified with romance, which invites us to interpret by superimposing typical blocks of narrative as they recur.12 By situating her lais in a collection, Marie further invites us to carry out the same interpretive gesture across the boundaries of individual lais. We thus see Marie engaged in a process of rewriting that characterizes both intra- and intertextuality in romance tradition, here transposed in the written textuality of "oral" storytelling. While the intertextual dialogue remains implicit in most twelfth-century romances, insofar as these authors present them separately and depend on the public to recognize their play, Marie intensifies and multiplies the dialogue by arranging her tales as a collection of twelve.13 Intertextuality is clearly located both within and without the col-

162

Chapter 5

lection. The sense of proliferation across the tales anticipates on a small scale our experience of the prose cycles, yet the author's exercise of control recalls the kind of conjointure that Chretien describes as the romancer's art of putting together the materials broken apart by oral storytelling. As I develop it below, this argument for reading Marie's assemblage as a significant unit does not require any particular order for all twelve of the lais, although it does depend on the frame established by the General Prologue and the two longest tales placed at the opening and close of the collection. Given the order in which the Harley manuscript presents the twelve tales, there are any number of analyses to be pursued that focus on the effects of "local" intertextuality, generated by the juxtaposition of specific lais. I would argue, however, that Marie's textual identity would not be completely changed by reordering her collection within the boundaries set up by the General Prologue, Guigemar^ and Eliduc. Although the immediate effects of juxtaposition would certainly shift, the network of intertextuality that is not tied to any linear ordering would still operate, as orchestrated by Marie's choice of twelve lais. In this respect, Marie's art differs considerably from Chretien's and romance in general, since she has invented a kind of structure that can integrate a degree of mouvance associated with the retellings of oral performance (or the varied patterns of rereading). We might compare this aspect of Marie's collection to the way different manuscripts sometimes reorder troubadour or trouvere poems: the stanza order may vary and generate different dramatic effects depending on the specific order chosen, but the poem, in some sense, remains itself in each of these particular and varied manifestations, especially since the first and final stanzas are more likely to remain in place. Similarly, Marie's Lais operate through the basic unit of the lai as each one is juxtaposed to those preceding and following. Every lai thus preserves its own coherence and deserves individual attention. But Marie's artistry also appears at the higher level of the arrangement: "M'entremis des lais assembler" (P, v. 47). This level, too, demands our attention—and here I would make an argument for the necessity of keeping the first and last lais in the positions they occupy, in order to maintain the collection's textual identity as assembled and presented by Marie in the General Prologue. With that frame in place, we are asked to read Marie's lais with a double focus, to see them as discontinuous and as participating in a larger unity that retains their diversity. While acknowledging that the complexity of Guigemar and Eliduc far exceeds the present concerns, I focus on them here most especially in terms of their role as markers, establishing important rhetorical boundaries for the collection as a whole.

Textual Identity and Marie de France's Lais

163

By virtue of their placement as the initial and final lais of Marie's collection, Guigemar and Eliduc implicitly ask us to read them, at least on one level, as statements about the way Marie has reinvented each lai and organized her collection, chosen from the many possibilities offered by the tradition of storytelling, as she invokes it in the General Prologue. In this sense, Guigemar and Eliduc represent in their narrative development two complementary processes at work throughout Marie's translatio: selection and substitution, Marie's specific variants on the general techniques of repetition and variation that characterize all writing in medieval literature.14 While Guigemar problematizes and dramatizes the process of selection, by showing the couple's effort to differentiate itself from others and recognize its own unity with respect to society, Eliduc situates the sorting-out process in a series of substitutions, through which the couple is finally relocated within a threesome. Unity and diversity are reconciled to establish a point of stability that furnishes an appropriate closing for the collection as a whole, since the inclusiveness of the threesome does not eliminate the distinctive configurations of its members in couple units. The same complementary strategies, based on selection and substitution, which describe the fundamental actions of characters within the stories, also delineate the activity of the author/narrator in relation to oral and written traditions of storytelling, as they exist before she begins to write and will continue after Eliduc ends. Within the intertextual network of stories circulating in written and oral forms, Marie's own process of selection and substitution is designed to culminate in our recognition of her textual identity. Detailed analyses of Guigemar and Eliduc will thus lead directly to an exploration of Marie's distinctive voice as articulated in her collection.

Selection and Substitution: Guigemar Although selection and substitution are constantly intertwined in the Lais, the problematic of Guigemar suggests that the first step of analysis should be taken along the route of selection, since identity is founded on difference operating against a common shared background. In the world of the lais, people are not sorted out in the right place, with the right people. Substitutions test their suitability until the right match can be found—and if found, they test further the possibility of maintaining the unit of the couple in the face of pressures for continued substitution. There is no

164

Chapter 5

selection without substitution, but the goal of Guigemar seems to give the establishment of difference the first priority. Detailed analysis of Guigemar shows that Marie develops a programmatic series of doubles throughout her narrative—epithets, actions, emblems, characters and plot lines, verse construction and renarration—to establish the correspondences that link together Guigemar and his lady and, at the same time, differentiate or separate the couple from those who pursue them.15 For Guigemar the initial problem involves his undifferentiated sexual identity expressed as a lack of interest in love—a fault that the narrator attributes to Nature herself (w. 57-58). This lack creates an imbalance in Guigemar's characterization, given the strength of his interests and attainments in chivalric pursuits, that leads to the accusation of perdition murmured by strangers and friends alike. We can distinguish in this initial situation a number of polarities that need to be sorted out and synchronized: the male/female difference must be firmly established, and Guigemar's perfectly symmetrical family unit—mother and father, daughter and son—must be dispersed and recreated in a new generation—a movement that involves Nature and society, lineage and marriage, past and present, all of which have to work together if feudal society is to continue into the future. A doe with stag horns, the magic beast whose appearance signals the beginning of Guigemar's adventure, aptly represents a confusion or fusion of male and female, which reflects both the opening problem and anticipates its eventual resolution.16 If we compare the initial situations of Guigemar and his lady, a set of parallels appear in mirror image: where Guigemar has nothing whatever to do with love, the lady has altogether too much all around her, since she has been enclosed by her jealous old husband in tower and garden, her room painted with pictures of Venus, consigning Ovid's Remedies Amoris to the flames and excommunicating all those who would read and follow that book's instructions. Of course, underlying that maximum of difference between their initial starting points is the essential similarity that already ties Guigemar to the lady and lays the groundwork for their future union: neither one knows true love, each one will experience concurrently—in the juxtaposed parallel descriptions of Marie's text—new love unfolding in the typically Ovidian mat d'amour that characterizes such scenes in twelfth-century narrative. As the lady's suivante tells Guigemar in order to encourage his open declaration of love, their love would be appropriate ("convenable," v. 451), as a reflection of their harmonious match: "Vus estes bels et ele est bele!" (v. 453: "You are handsome and she

Textual Identity and Marie de France's Lais

165

II

I

1

2

3

1

2

3

—

+

X

—

+

N

lovers together (in lady's chamber)

crisis: Guigemar's departure; Lady imprisoned by husband

lovers separate: Guigemar home; Lady imprisoned by Meriaduc

lovers recognize each other at Meriaduc's

nexus: lovers reunited through feudal means: love > prowess (combat), lady as prize

lovers separate

Figure i. Bipartition and tripartition in the narrative structure of Guigemar.

is beautiful"). The differences that link Guigemar and his lady are thus not just any unrelated differences, but rather precisely those oppositions, negative images, that Marie can reverse and coordinate in the machinations of her narrative. Guigemar can substitute for and successfully replace the old husband because he appears as the masculine equivalent to the lady and her qualities. Given this attraction of like-minded opposites, Guigemafs two separate plot lines are tied together, again separated, and finally reunited through a series of events designed to reduce the difference between Guigemar and the lady, while at the same time clearly defining each one's sexual identity—the irreducible kernel of difference that is the sine qua non of their union. Bipartition and tripartition not surprisingly are coordinated in this effort. While the plot is doubled in its general progression, each half has three distinct parts, the first two in direct correspondence from one half to the other, the third part in mirror image (see Figure i).17 The plot divides in half, when each character returns to a slightly, but significantly, altered version of his or her initial situation: Guigemar back in his own country is once again pursued without success by dames and puceles; the lady, now on the same side of the sea as Guigemar, is once again imprisoned and importuned by unwanted love, this time by Meriaduc, a neighbor and companion in arms of Guigemar. The process of substitution now threatens the unit of the couple, whose mutual selection is nevertheless guaranteed by their exchanged tokens: Guigemar's pleit, the knot in his shirt that only his lady can untie, the lady's ceinture (belt or girdle) that only Guigemar can unlock. These tokens of love and fidelity

166

Chapter 5

are exchanged when the lady, like the magic doe of the opening movement, prophesies separation and departure for Guigemar. That overwhelming narrative echo is supported in the very texture of Marie's verse by parallel constructions that link first Guigemar to his previous state (e.g., w. 44 and 643, 59-74 and 652-54, 63 and 645), and then Guigemar's current state with that of the lady (e.g., w. 644 and 718, 652 and 740-42). The stage is now set for a final replay that will remove the obstacles to the couple's unity. Knot and girdle, parallel objects that retain the essentials of sameness and difference that mark the male/female couple as a unit within society, play a key role in the elaboration and resolution offered by Marie in the second half of Guigemar. Meriaduc, who has seized the lady in a gesture of feudal possession, finds himself frustrated in his desire for sexual possession by the object and condition that echo the test of the knot already famous in Guigemar's country. When the lady explains that she will love only the man who can open the ceinture without breaking it, Meriaduc explicitly compares her test to that arranged by Guigemar, who has refused to marry any lady not able to untie the knot in his shirt using neither force nor knife. Meriaduc even guesses that she must be the one who made the knot (w. 722-34, cf. w. 643-54). Repetition is the key element that allows Meriaduc to discover the link already connecting the lovers. By marking the boundaries that set off the couple from those who pursue them, repetition and variation each have a role to play in the sorting-out process, each one identified in turn with selection and substitution. In the course of my analysis so far I have signaled in passing a good many of the doubles and series of doubles that lead programmatically to the final joyous moment when Guigemar leads away his amie. A closer textual analysis would support and elaborate the way in which Marie has saturated her entire text with such recurrences, modulating between repetition and difference, in order to define the lovers for each other, for themselves, and for society. The proof of their resemblance to each other and their difference from the rest of society is guaranteed by the multiplicity of those doubles that fit together like the pieces of a jigsaw puzzle. Why, we may ask, does Marie have to go to so much trouble!1 Why indeed in the story itself does Guigemar have such a hard time recognizing his lady? "Femmes se resemblent asez" (v. 779)—"Women look so much alike"—says Guigemar to himself and, while I am inclined to disagree with Guigemar's observation as applied to reality, it is nevertheless an accurate description of the textual world in which he lives. A number of scholars writing about Marie's Lais have commented on the similarities between

Textual Identity and Marie de France's Lais

167

her characters, all equally handsome and beautiful, noble, and courteous.18 To judge by their outward appearances, any one of these heroes and heroines could substitute for another. Moreover, Guigemar is not the only lai in Marie's collection where the pair of lovers is marked by parallel characterization. In this respect we might also think of Lanval, Deus Amanz, Yonec, Laustic, and Chiewefoil, though none of these elaborate that parallel to the extent that we have seen in Guigemar. Individual instances of repetition are insufficient to distinguish a given character: they mark rather a threshhold to be crossed, the minimum indication of belonging to the world of the Lais, beyond which men and women must go in the course of the narrative to establish their own identity and find their destined match. Recognitions occur repeatedly throughout the Lais and play an important role, whether literally or figuratively, in their plot construction. The importance of recognition as event mirrors its thematic significance: it is the very action that recognizes textually the interaction of selection and substitution as the major, complementary axes of the collection, considered internally in the construction of the twelve tales or externally in relation to the tradition. As such, Guigemar's representation of the recognition process, dramatized with particular care and insistence, sets up an important model for the entire collection. I would like to focus briefly on that scene and its components in order to show how, from Guigemar's point of view, sense perception, reason, and discourse all contribute to his efforts to recognize his lady. The narrative point of view throughout the scene follows closely the whole process of recognition, primarily identified with Guigemar's perspective, but occasionally with reference to those of Meriaduc and the lady. Marie's text thus moves back and forth between narration of actions and short descriptions of the inner states experienced by her three principal actors. The textual strategies thus give us a privileged place in the distribution of knowledge: we know who everyone really is, while at the same time we can follow the specific stages through which that knowledge becomes accessible to the characters involved. Since the lady has already heard about Guigemar's knot and heard his name pronounced at the beginning of their encounter (v. 765), her knowledge precedes, but waits expectantly, until Guigemar attains full recognition. Meriaduc has the unenviable role of being the one who triggers the very actions that will lead to recognition between the lovers and to elimination of the rival, Meriaduc himself. What is the process by which Guigemar recognizes his lady? What

i68

Chapter 5

are the signs that must be interpreted for the lovers to make contact across the gap that prevents any individual from direct apprehension of another, whether within or without literary texts?19 Guigemar first sees his lady; sense perception triggers the whole process—and Marie's verse construction uses doubling to insist on the moment of visual perception: "La dame vit e esgarda / E sun semblant e sa maniere" (w. 770-71: "He saw the lady and looked at her appearance and her manner"). That perception then leads to a pause in the narration, as he draws back for a moment (v. 772): an interior space opens for Guigemar's thoughts, reported in direct style. Guigemar's first response is an emotional one; the language of his desire speaks in lyrical repetition, recognizes his love: "Est ceo, fet il, ma duce amie, M'esperaunce, mun quor, ma vie, Ma bele dame ki nVama:"" (w. 773-75) ("Is this, he said, my sweet friend, my hope, my heart, my life, my beautiful lady who loved me:1") But already the interrogative mode adds an undercurrent of worry that soon displaces the emotional identification of the lady. Where does she come from:1 Who brought her:1 Reason asks questions, fears to be led astray by desire; rationality leads to doubt, identifies folie rather than the lady ("'Ore ai pense mut grant folie,5" v. 777: "cNow I have had a very foolish thought'"). Guigemar has earlier used the same term ("csi fis folie,'" v. 329: "CI committed a folly"'), when describing to the lady his entrance into the magic ship. In each case Guigemar's rational response is subsequently shown to be inaccurate, or at least inadequate to account for the reality of his experience. The verbal association with that earlier moment suggests where we should locate Guigemar's problems of perception. Earlier Guigemar could easily pass from one place to the other, from the reality of Brittany to that of the Other World; his task now is to perceive the intersection of the two: love, formerly reserved for that other place, has now to be located and identified "here." Reason alone does not function properly in the fused world of the Lais, where the marvelous and the human dimensions enjoy an equally compelling "reality" without canceling each other out. Guigemar's perception of resemblance, however illogical it may ap-

Textual Identity and Marie de France's Lais

169

pear to him, does furnish the impetus for further action, when he decides to speak to the lady for the sake of her double (w. 781-83). As always in the world of the Lais, repetition is at the root of narrative amplification. When the lady easily unties the knot, an amazed Guigemar recognizes his lady, but doubt still clouds that recognition (w. 812-14). One magic object calls for its mate: Guigemar asks to see and then touches the ceinture with his hands. Like reason, sense perception plays an important role here, as the narrative reintroduces it at critical moments in the process of recognition. Yet neither the information of the senses nor the introduction of the magic objects themselves close off that process. That moment of closure is reserved for the retelling of the past, when Guigemar asks his lady to tell what adventure has brought her to Meriaduc's castle (w. 822-35). Twice before at critical moments of the narrative—that is, when Guigemar arrived and departed from the lady's country (w. 308—58 and 602-9)—the characters told each other their stories, repeated what we in the audience had already learned by Marie's own narrative presentation. The separate strands of their histories are thus doubly entwined in the to, by their own discourse as well as the narrator's. Each time past actions are translated into discourse, which in turn generates future action. The lady's narrative of her past itself confirms the prophecy of the future uttered by the wounded doe: she has indeed suffered such pain for Guigemar and he for her that all lovers, present and future, will marvel (w. 119-21). The public of lovers projected by the doe moves constantly forward into the future, whether within the story or outside it. The tenses of the lady's retelling likewise move from past to present, from free indirect discourse to direct address (w. 825-36), the last verse significantly couched in the imperative, as the lady calls upon Guigemar to lead away his beloved: "Amis, menez en vostre druz!" (v. 836). Her command projects the future and furnishes the closing action of the story. The private love of the past moves from an intermediate stage, where public knowledge coincides with separation of the lovers, to the final stage of their union and social integration. As represented in the climactic scene of Guigemar, the process of recognition requires all the resources of sense perception, the special trials of magic objects, the faculties of reason and memory, and the crowning movement of the whole process—the transformation of action into discourse, recit. Through the combined effects of these varied representations, the totality of the characters' past is reactivated, moved wholly

170

Chapter 5

into the present of the current recit—that is, the lai we are now reading.20 For the public of the Lais, all is mediated through discourse, the retelling accomplished by Marie herself.

Recognition and Eliduc While Guigemar particularly valorizes the lovers5 successful match, their ability to differentiate themselves from all rivals, other couples in the Lais do not always achieve the right match, fail to maintain it, or maintain it only in death or metaphor. Variations on the common problem gradually inform the reader about the interplay of selection and substitution, as they bring into focus the importance of the second: it is the process of substitution that makes selection possible, but it does not always end when a selection has been made. The difference between success and failure, as illustrated by different sets of lovers in the Lais, points to the difficult, but necessary balance between the two fundamental operations of selection and substitution. Chaitivel serves as a kind of cautionary tale within the collection about the failure to achieve selection, as dramatized both by its plot and, most especially, by the discussion of its title. Recognition in Chaitivel, as in Guigemar, concerns a process of differentiation: four knights, all contenders for the lady's love, are unable to differentiate themselves without the lady's choice—and she, unwilling to single out one and lose three, encourages all four. Even after the tournament, with three dead and one wounded, the difference between life and death is neutralized as far as winning the lady's love is concerned. The choice between two titles, Quatre Dots ("Four Sorrows") or Le Chaitivel ("The Miserable One") not only dramatizes the problems of selection and difference but relates them to the question of gender and point of view. While the knight within the story convinces the lady to accept his title and thus single out at least nominally the only remaining suitor, the narrator's epilogue asserts the equal applicability of both titles: "Chescuns des nuns bien i afiert, / Kar la matire le requiert" (w. 235-36: "Each of the names fits well, for the story requires it"). The open-ended process of substitution seems to have kept the upperhand in this particular lai and so reminds us that reaching a point of equilibrium between selection and substitution is no easy feat. The most common substitution that appears in the collection is that of lover for husband, during which our sympathies may be directed either

Textual Identity and Marie de France's Lais

171

toward the former (e.g., Yonec, Miluri) or the latter (e.g., Equitan, Eisclavret). In some cases that initial substitution is overshadowed by further displacement when a symbolic object substitutes for the lovers' bond by representing it metaphorically (e.g., Laiistic, Chiewefoil). Less frequently, but with greater variety, Marie makes use of female character substitutions (Fresne, Lanval, and Eliduc). Many echoes between Fresne and Eliduc have been pointed out in previous scholarship. In a sense, the placement of Chiewefoil right before Eliduc rekindles our expectations about what may happen when a man has two wives—the story pattern that appears in Fresne and Eliduc, as well as in the Tristan story.21 The substitution of one sister for another is immediately corrected in Fresne, when the mother recognizes the blanket and ring she sent along with her baby daughter: the match that appeared to be wrong from the barons' point of view turns out to be right for both the lovers and society. The process of selection thus triumphs, but does so in a context that dramatizes not only the possibility, but the ease of substitution, through the invention of twin sisters, "twinly" named. Significantly, Fresne and Eliduc are the only lais in the collection that give names to the important female characters, names chosen as if to emphasize the process of substitution that relates each set of ladies. In the case of the twins, despite their separation at birth, both are named after trees, Fresne metonymically because she was found in an ash tree, La Coudre no doubt so that Gurun's barons may contrast metaphorically the fruitless ash with the fertile hazelnut tree (w. 335-40). In the case of Eliduc's two wives, repetition of the same set of consonants associates Guildeluec and Guilliadun as verbal echoes of each other, even before they are linked through association with Eliduc. A network of allusions and associations, both within and without the collection, thus prepares us to read Eliduc as the final movement in Marie's orchestrated "theme and variations" on the interplay of selection and substitution. In so doing, we may see it as a kind of counterpoint to the process opened up by Guigemar—indeed a number of textual elements invite us to line up and compare these two lais, not simply in terms of their interaction as two separate tales, but also in relation to their role as boundary markers for Marie's collection. By assembling the Lais as a unit, Marie inscribes each one in a complex web of intertextuality whose patterns elaborate and comment on the design of the whole. In certain respects, Eliduc appears as a kind of mirror image to Guigemar, their common characteristics highlighted by a number of significant reversals. Guigemar and Eliduc are the two longest lais in the

172

Chapter 5

two countries alternate

movement contained within one country

1

2

initial harmony at homew/K 1 , G1; disrupted by envy

E's departure; sojourn w/K 2 ;loveofG 2

3

4

return home to K l and G1; unhappiness

E's departure (return to K 2 ); elopement w/G2

5

6

return home to K l & G 1 (w/G 2 )

E's repeated departures to visit "dead" G 2

7

8

Cure of G 2 by G 1 brings G 2 to E's home

G1?s departure for convent

9

10

G 2 and E's married life

departure of E and G 2 to join G : in religious life

«

Figure 2. Episodic structure in Eliduc.

collection. They both demonstrate Marie's use of episodic structure, based on a pattern of departures and returns that invites us to superimpose the repeated units to see how their variations comment on each other. While Elidufs narrative structure is not identical to that of Guigemar as described earlier, it recalls the pattern of duplication that brings two different worlds—that of the hero and that of the beloved—into contact and ultimately into the same place.22 The crisis precipitated by moving both ladies into the same country continues the pattern of departures and returns, but now within a single geographical unit, until all three characters are joined in the common space of a religious life (see Figure 2, where Guildeluec = G1; Guilliadun = G2; King of Brittany = K1; King in Logres = K2).

Textual Identity and Marie de France's Lais

173

While the opening description of Eliduc (w. 27-33) recalls similar statements in Guigemar (w. 27-33), we soon learn that the same fundamental process of sorting out the right couple has been dramatically reversed: the multiplicity of men in the Guigemctr story is replaced by a multiplicity of women in Eliduc. But even given the opposite problem as regards gender, Eliduc's progress in the back and forth movement of the plot, at least until stage 7, suggests that his pursuit of the right match—Guilliadun—is similar to Guigemar's: selection is the ultimate goal and should take precedence over substitution, once the appropriate selection has been found and made through the process of substitution. The allusions to the Tristan story, which mark both specific actions in the plot, as well as the characters' speech, reinforce our expectation that such is the pattern to be realized in Eliduc, as it has been attempted, sometimes with great success, in all the previous lais. C£CVus estes ma vie e ma morz, / En vus est trestuz mis conforz'" (w. 671-72: ucYou are my life and my death. In you is all my comfort'"). We could easily mistake these words told by Eliduc to Guilliadun for a quotation from the Tristan story, even from Marie's version in Chievrefoil. But where the Tristan story moves inexorably to its tragic ending through the machinations of the second Iseut, the two women in Eliduc—whose names sound so much alike, but avoid the exact repetition of their models—rewrite the links between love and death, in order to furnish a happy ending to their story.23 Where Tristan's wife brings death, Eliduc's wife restores life—to her own rival. This critical difference requires us to look more closely at the moment Guilliadun is cured, in order to follow how Marie invents the unexpected ending of Eliduc by shifting the balance between selection and substitution. How should we understand Guildeluec's actions, when she uses the magic flower to awaken Guilliadun from her deathlike sleep? Her motivation is embedded in a series of substitutions that describe her position in the love triangle peculiar to Eliduc. Consider first the difference between Guildeluec as wife-to-be-replaced and the husbands of Guigemar, Yonec, Milun, and so on, the antagonists who are left behind or eliminated with no regrets, as our sympathies are identified with the lovers. There is no "bad guy" in the love triangle of Eliduc. Initially husband and wife loved each other loyally for a long time; their match was a good one, only disrupted by external circumstances that had nothing to do with the members of the couple per se (w. 9—16). The narrator prevents us from considering Eliduc in simplistic, negative terms by frequently reminding us of his good intentions to remain loyal to all those he loves—lord, wife, king, and be-

174

Chapter 5

loved (w. 83-84, 107-10, 322-26, 462-76, 585-618). Guilliadun's innocence is safely guaranteed by her ignorance, since Eliduc has omitted to tell her about his previous marriage. The sailor who blurts out that information during the storm and who speaks in the voice of normal morality ("Temme leal espuse . . . D e i . . . l e i . . . dreiture . . . fei,"5 w. 835, 837-38: "'Legally married wife . . . God . . . law . . . justice . . . faith"5), becomes a kind of lightning rod that brings upon himself the punishment he would have assigned to Guilliadun. His body substitutes for Guilliadun's, as Eliduc throws him into the sea and takes over the helm to guide their ship safely to shore. His is the real dead body; Guilliadun's has only the appearance of death.24 Since there is no easily rejected member in the love triangle, the solution to the conflict necessarily follows a different path of substitution. Guildeluec herself takes this path by identifying herself with (and thus in a sense substituting for) her husband. In a lai that gives so much attention to naming, we may see this process of identification foreshadowed in the spouses5 names, since all the letters of Eliduc are contained anagrammatically in that of Guildeluec. The actual process of substitution begins when Eliduc first returns to his country, as Guildeluec seeks an explanation for his obvious unhappiness (w. 721-26). Just as Eliduc was falsely accused and ready to defend himself before his king (w. 41-50), so Guildeluec assumes her position to be an analogous one vis-a-vis her husband and is ready (like Iseut) to prove her innocence. Her tendency to identify with Eliduc's point of view becomes even more striking in the scene when she discovers the "dead55 Guilliadun. When she has Eliduc followed to find out the destination of his repeated absences, she knows immediately that his grief cannot be explained by the hermit5s death (w. 1001—4). There must be something else at the hermitage to investigate, so she decides to visit it—secretly, as Eliduc did—to find out what can justify the depth of her husband's feelings. Seeing Guilliadun immediately reveals the truth and justifies Eliduc's conduct in her eyes: "Ceo est I5amie mun seignur Pur quei il meine tel dolur. Per fei, jeo ne m5en merveil mie, Quant si bele femme est perie. Tant par pitie, tant par amur, James n'avrai joie nul jur.55 (w. 1023-28)

Textual Identity and Marie de France's Lais

175

("This is my husband's beloved, for whom he laments so, and in faith, it is no wonder when such a beautiful woman has perished. Either pity or love will prevent me from ever knowing joy again." p. 124) His grief is now hers, just as his vow to found a monastery and become a monk—made when he left Guilliadun's body on the hermit's alter (w. 947—50)—will shortly become her own, mutatis mutandis, in order to give Guilliadun back to her ami (w. 1099—1102). Her convent will be founded on the very spot selected by Eliduc earlier (cf. w. 1122-44 and 895-902). In each of these reactions, we see Guildeluec voluntarily giving up her position as wife to assume the one identified with her husband—a displacement that opens the position of beloved (wife) for Guilliadun. The adventure, marvelous and unexpected, that allows Guildeluec to take the role of Eliduc and guarantee that his love bring not death, but life, to Guilliadun, introduces another series of substitutes, this time for the two ladies. The drama of the two weasels appears on cue as a mise en abyme that furnishes not only a model for bringing Guilliadun back to consciousness, but also anticipates the bond we shall see uniting the two women at the end of the tale. One weasel is struck dead for having passed over Guilliadun's body: that blow recalls (for the public) the one Eliduc aimed at the sailor—and while he was the physical victim, Eliduc's amie was no less felled by her lover's action. The weasel's grief for its companion (v. 1039) associates her with Eliduc's and Guildeluec's emotions for Guilliadun, so it seems a natural response for Guildeluec to imitate the weasel's miraculous cure with the red flower. In her subsequent conversation with Guilliadun, her words again emphasize the extent to which she has taken her husband's point of view, as she defends him against accusations of betrayal (w. 1085—92) and insists on uniting the lovers (w. 1099— 1104)—in effect placing Guilliadun in the position she has herself vacated by moving over to his "side." Each of the substitutions initiated or perceived by Guildeluec, first by taking her husband's point of view, then by playing out the drama of the two weasels as a scenario for the two women, seems designed to reinforce and valorize the substitution of Guilliadun for Guildeluec originally made by Eliduc's coup defoudre. However much he tried to avoid a choice between the two women, Eliduc's actions belied his intentions; his wife Guildeluec correctly divined and realized his choice by her own selfeffacement. If the story had ended at this point, with the lovers happily

176

Chapter 5

married and Guildeluec ensconced in her nunnery, we might still see Eliduc as another triumph for selection, which uses substitution for its own ends to delineate the couple and exclude the inappropriate substitutes. But Marie adds an event that transforms the final configuration and forces us to rethink the relationship between selection and substitution. After years of perfect love and acts of charity, Eliduc and his second wife convert: Eliduc builds a church, founds a monastery, which he enters, and places the second wife with the first, who receives Guilliadun as a sister (w. 1165-68). To what end does Marie include this last shift in the match? Some scholars have commented on the religious character of this ending, seeing a properly Christian cantos as the final ideal of love proposed by Marie's Lais.25 Others have stressed the amoralism of Marie's tales, where Christianity appears rather as part of the basic components of the characters' world without carrying any privileged message: Eliduc's conversion may appear then as a kind of apology for his weakness, an effort to raise him to the noble example set by Guildeluec.26 Without exploring the problem of Marie's or the Laifs spirituality, which is a difficult question that far exceeds the bounds set by the current analysis, we may nevertheless ask what this Christian ending offers in terms of the collection's overall design—that is, how does the figure of God help Marie realign the dynamic interaction of selection and substitution to provide a stable point of equilibrium that appropriately closes her collection of lais> What is striking about the final configuration is its inclusive, rather than exclusive nature. As anticipated by the two weasel companions, the two women are joined as sisters, with Guildeluec still taking the lead in the "man's role," when she teaches Guilliadun her order. Both women are designated in the closing description zsfemme, distinguished by qualifying phrases ("premiere . . . que tant ot chiere," w. 1165,1166: "first. . . that he held so dear"), in order to stress their identity as wives in relation to Eliduc. The normative demands of law, God, and Christianity—as interpreted earlier by the sailor and even by husband and wife (see w. 601-2, 1128-29)—required separation of the lovers, just as love's demands required separation of the spouses (despite Eliduc's unwillingness to face such choices). The final configuration shows the women as a couple vis-avis Eliduc: all three are united with reference to God, as the women pray for their beloved's salvation and he for theirs (w. 1171-80). Guilliadun's unorthodox withdrawal first allows the love match to prevail, then sets an example for all three to be joined in their service to God, within whose

Textual Identity and Marie de France's Lais

177

unity all diversity is integrated. God's universal aspect, which reconciles unity and multiplicity, provides Marie with a model to close off the game of selection and substitution as alternating modes, by replacing it with one that encompasses both in equal measure.27 The distinction that gives definition to the couple—the achievement of the right match, here based on the female gender of both—no longer excludes their double link to Eliduc. With Guigemar we saw the complementary processes of selection and substitution working alternately to create links of similarity that would bind together the male/female couple and guarantee differences to set them apart from further substitutes: selection triumphed over substitution, until the process started all over again in the next led. By changing the gender of the last couple in Eliduc^ Marie is able to contain in the final threesome both selection and substitution: the sisterly bond of Guildeluec and Guilliadun marks them as a couple and, by suggesting their equivalence or substitutability, joins them both to Eliduc through their spiritualized love of God.28 Unlike Guigemar and all the other lais where substitution is put outside the final couple, Eliduc ends with no one left out, no plot strings left hanging, no husbands deserted and forgotten, no children to start the process anew. Or rather, when we finish reading the last lai of the collection and recognize the boundary it establishes for Marie's assemblage^ any process of renewal now operates at another level, not that of narrative development, but rather in terms of the intertextual dialogue and the surplus de sen embedded in Marie's written text.

Marie's Textual Identity: Names and Titles It seems to me no accident that Eliduc appears as the last tale and only in the Harley manuscript that gives us Marie's Lais as a collection assembled by their author. If we consider the analogies between what her characters accomplish through the play of selection and substitution and what Marie as author/narrator working on traditional story matter accomplishes through the same means, we may recognize in the extra woman included in the final threesome a figure for Marie herself, now matched with the lovers who recur in each of her lais. This is one stage in a series of substitutions that involve Marie as author and narrator. As she presented herself in the General Prologue, Marie's task was to select from the body of Celtic tales already circulating a certain number, which she has now preserved in

178

Chapter 5

written form for a French-speaking audience. She thus substitutes her own version for the common body of tradition and, in presenting her collection to the King (with whom she makes the couple that gives birth to all those other couples included in the Lais], she establishes her own distinctive textual identity. The tradition of storytelling that precedes Marie will no doubt continue after her: it is because stories can be repeated, because one version can substitute for another, that the tradition itself exists and carries on. But if she is successful, Marie's selection will mark the common tradition with her difference: her written Ms will generate in the public a desire to gloss and elaborate their sen and thus set off a new chain of substitutions based on Marie's own particular narrative. While she tells the same stories others have told, Marie does so in a distinctive way, not only in the individual tales, but in their combination as a collection: when we recognize that textual difference, we form with Marie the couple represented in her Prologue by author and King, brought together by reception of the tales themselves. Within the collection the dual strategies of selection and substitution are constantly at work. Between the opening and closing lais the dynamic of both selection and substitution remains open, as we pass from one lai to the next, as Marie reinvents and elaborates her textual identity through their intertextual play. It is only when we arrive at the end of Eliduc, with that process complete, that Marie closes off her selection (at least on the level of narrative development). After establishing a closing equilibrium, she returns the stories, now given a new identity as a collection, to the common voice of tradition, typically expressed in the brief epilogue to Eliduc. The success of Marie's project is attested, not only by modern editions, translations, and bibliography, but by a contemporary witness like Denis Piramus, who describes her as the author of lais, written in verses, that delight courtly audiences who have them read out loud over and over again.29 This is precious testimony indeed of the recognition given Marie's art, but we may still wonder if this constitutes recognition of Marie's collection per se: Denis Piramus does refer to lais in the plural and so presents "dame Marie" as the author of a repertoire of lais. There is, however, no tide or term used to designate her assemblage. Given medieval usage, the existence or lack of a title may or may not be significant. While in general it may be correct to say that medieval works, with few exceptions, do not have titles in the modern sense, the lais seem to constitute one of the exceptions, at least on the level of the individual tale.30 The name given a particular lai and repeated by anyone performing it, whether

Textual Identity and Marie de France's Lais

179

orally or in written form, is indispensable to maintain the link claimed to connect the original adventure, the musical lai that commemorates it and the narrative retelling that explains it. Marie thus retains and repeats the titles of her lais, often calling attention to them as she translates them into different languages or includes discussion (by herself or the characters) of different titles proposed for the same lai. But she has not, as far as we know, proposed a title for the collection itself. In order to understand this authorial gesture, we need to see it in relation to a complex pattern of naming and not naming that characterizes Marie's Lais. As I have already suggested, Marie relies on the frame established in the Harley manuscript by the General Prologue, Guigemar, and Eliduc to set the boundaries and make known the identity of her collection.31 The manuscript tradition attests that Marie's lais circulated in different forms, some scattered here and there in miscellaneous anthologies, some gathered together in collections of Breton lays. When her frame is absent, Marie as collection does indeed disappear and she suffers the same indignities of dismemberment lamented by Chretien in the Erec Prologue. This is what we see in the miscellanies of mss. Q and C with only a single lai each, and even in mss. S and N, which not only mix Marie's tales among those of others, but also remove her "signature" in the Guigemar Prologue, as if to reclaim her tales for the anonymous storytellers of the oral tradition. Neither of these collections of lais include the General Prologue or Eliduc, although both place Guigemar as the first of their lais, introduced by w. 19—26 of the Prologue (preceded by the translator's Prologue in ms. N). Ms. P, which gives three of Marie's lais grouped together, retains Marie's name in the Guigemar Prologue, but she can no longer be identified as the author of a collection of tales on that basis alone. Interestingly, this manuscript also includes Marie's Fables, as does the Harley manuscript (a collection of various Latin and French texts). It is the Epilogue to the Fables that includes Marie's "signature," as it has furnished the name we give her today: "Marie ai num, si sui de France" (v. 4: "Marie is my name; I am from France"). But the Fables are separated from her Lais by quite a few folios (in Harley 978, folios 4oa-67b and n8a-i6oa respectively). Is this an accident of history, reflecting when the scribe was able to find other manuscripts for copying, or an indication of Marie's problematic status as author of more than one text (or perhaps something else entirely) ? Such uncertainties fuel the speculations of scholars like Richard Baum who question the identity of the author who signs Marie in the Prologues and Epilogues of these works.

i8o

Chapter 5

Although the historical identity of Marie as author does not concern us here, her textual identity as author of a collection of Ms does. As suggested by the manuscript tradition, that identity may be lost under certain conditions. But if we focus on the Lais as arranged in the Harley manuscript, we can continue to explore how Marie has inscribed her particular textual identity there through a set of strategies that serve to intertwine author, narrator, characters, and text. That new and specific textual identity, which differs from those made possible within the confines of a single led or even in a random sampling of lais, grows out of Marie's unique blending of features from oral storytelling and romance writing. Further analysis of her use of names and titles in the collection will lead us to focus on Marie's distinctive narrative voice. All the titles of the Lais are associated with names for the main characters. This usage places the author's name and the names of her lais on the same plane, tied to the proper name of a human agent, and facilitates the crossovers we can follow between the world of the author and that of her stories. The particular combination and contrast of Guigemar and Eliduc suggests the kind of play linking author/narrator and story matter. Consider first the chiasmic pattern that appears across the collection in the use of names and titles. Marie names herself in the so-called Prologue to Guigemar (v. 3)—which appears in ms. H as part of the General Prologue (marked only by the use of a capital letter at G, v. i), while the lai really begins (without a rubric) at v. 27.32 This is the only lai in the collection that Marie does not name in its prologue. Even w. 19-26, which use the topics of the individual framing prologues, make no specific reference to Guigemar and his story. The first lai of the collection receives its name only in the epilogue (v. 884). How should we understand these opening moves in Marie's strategy of naming? Although anonymity typically characterizes the oral storyteller, Marie does not remain anonymous, but proudly presents her name associated with a typical exordial topic that expresses concern for her art: Ki de bone mateire traite, Mult li peise si bien n'est faite. Oe'z, seignurs, ke dit Marie, Ki en sun tens pas ne s'oblie. (w. 21—28) (Whoever has good material for a story is grieved if the tale is not well told. Hear, my lords, the words of Marie, who, when she has the opportunity, does not squander her talents, p. 43)

Textual Identity and Marie de France's Lais

181

Even when she speaks directly to her public as any oral performer would, Marie aligns herself with the written tradition of romaunz, where authors typically name themselves among the topics of the prologue designed to guarantee the authority of their texts.33 But romance is not the only contemporary literary model in which authors' names are usually available and Marie has just alluded to that other tradition, that of troubadour and trouvere lyric, in the verses immediately preceding her name.34 When she offers her collection to an unnamed king in the hope of receiving great joy as her part of the exchange (P, w. 42-55), Marie evokes the same triangle of relationships operating in the aristocratic lyric, but she reverses the sex roles, since it is usually the male poet who offers the product of his art and the lady recipient who remains anonymous as (ma) dame. The King and Marie thus form a male/female unit that anticipates the many couples of her collection, even though within her stories Marie will return to the more typical pattern of naming in which the ladies typically remain unnamed. Marie includes her own name in the opening remarks, but not that of the King, whose description puts him on a par with any of the heroes we find later in the tales. Already in the General Prologue, Marie begins weaving together authorial and fictional worlds. Given these various contexts from which to consider Marie's acts of naming in the opening verses of her collection, how can we understand the placement of her name where we might otherwise have expected to see Guigemar's, either as character or title for the first lai> With Guigemar, Marie also forms a male/female couple, whose association is explicitly prepared in the Prologue when Marie describes herself in such a way as to anticipate the portrait of her male hero. Her obligation as author to speak out and treat her matter well (w. 1-4 quoted above) leads to the issue of praise and envy: while people should praise others who make themselves well spoken of, those envious of "Humme u femme de grant pris" (v. 8: "Men and women of great worth") often do just the opposite. Woman author and male character are both presented as objects of unjustified slander (cf. w. 65-68), despite their considerable accomplishments. Exceptionally, both Marie and Guigemar are named—but not at the same time and not in the same place. The displacement of names—Marie's where we might have expected Guigemar's, the delay of Guigemar as title to the closing moments of the epilogue—in some sense plays a variation on the pattern of naming—usually one member is named, the other anonymous—which at once preserves and violates it. Marie must announce her name if she is to establish her textual identity in the inter-

182

Chapter 5

section of traditions she evokes; she must also give us the title of her to to establish the link with Breton storytelling. By giving both, separated by the space of the to itself, Marie is able to satisfy both demands, while at the same time forming a "typical" couple with Guigemar as both character and tale. We can see another such variation across the worlds of author and text, when we consider the pattern of naming in the last to' of the collection. Marie's name does not reappear in the epilogue to Eliduc (quoted above), which closes the collection with no special fanfare, by briefly repeating the kind of topoi that have appeared throughout the collection to affirm the lai? links with a living body of oral tradition passed on by Breton storytellers. The absence of Marie's name at this strategic rhetorical boundary, as well as the absence of any mention of her collection per se, has occasioned some concern about both Marie's and the collection's identity.35 But once we have sufficiently understood Marie's use of names to connect and blend author, narrator, characters, and text, I believe we can read her absent name in the Prologue to the final to', generated in the verbal play between the two titles. The length and disposition of topics in this last Prologue seems to recall in echo the Prologue to Guigemar. After four verses that combine the non-specific topics that identify any Breton lai (cf. G, w. 23-6), followed by an elaborate summary of its plot (w. 5-20)—elaborate, that is, by the standards of brevity set in other individual prologues (cf. DeuxAmants^ w. 3—4; ITonec, w. 7—10), the Prologue to Eliduc offers an important discussion about the titles of the to'. D'eles deus ad li lais a nun Guildeluec ha Guilliadun. Elidus fu primes nomez, Mes ore est li nuns remuez, Kar des dames est avenu L'aventure dunt li lais fu Si cum avint vus cunterai, La verite vus en dirrai. (w. 21-28)36 (From these two the lay of Guildeluec and Guilliadun takes its name. It was first called Eliduc^ but now the name has been changed, because the adventure upon which the lay is based concerns the ladies. I shall relate to you the truth of it as it happened, p. in) If we look closely at the titles proposed for the last to of the collection, we may notice immediately that they call attention to the differ-

Textual Identity and Marie de France's Lais

183

ence between male and female points of view in a way that echoes the debate on ChaitiveFs title described above. We may then notice, as we look over the list of Marie's titles that, while they are all connected to the important characters involved in each adventure, Fresne offers the only other example of a woman's name as title, and the only other example of a named heroine, whose name and that of her sister, La Coudre, play a significant role in a story that anticipates the triangle of Eliduc. In a collection in which most of the heroines remained unnamed, it seems all the more important to notice those who are, especially when their names are singled out as designating the lafs title. Unfortunately neither medieval rubricators nor modern editors have followed Marie's instructions about the name change—so we continue to refer to Guildeluec ha Guilliadun by the (admittedly more convenient) title of Eliduc. What we miss in so doing is the way Marie has allowed two women characters to take the initiative both in the events of the lai and in a title that captures the essence of their aventure. When we are reading one of the few women authors of the period, we can ill afford to neglect one of the few instances where she not only chooses to reverse her general tendency to present heroines anonymously but even highlights their roles by making the heroines' link the title event. Most important, since this change occurs as the last moment of naming in the collection, we need to analyze it as a crucial move in Marie's strategy to establish her characteristic and very particular textual identity. By giving both titles, but this time choosing one over the other, Marie accomplishes several goals. First she alludes to and goes beyond the problematic set up in the discussion of ChaitiveFs title, where the narrator does not privilege one name over the other despite the characters' own declared acceptance of the lady's title alone. Secondly, Marie anticipates the inclusive three-part unit her characters will form at the end of their story, with the newly formed female/female couple valorized yet aligned with Eliduc in their mutual service to God. Finally she echoes through the repetitions of sound, for the first and only time, the name of Guigemar, her first hero, who sets up yet another threesome with the heroines of the last lai: Gz/zldeluec, G^/lliadun, and G//zgemar. I suggested earlier that the extra woman in the last threesome functions in some sense as a figure for Marie. The play of the names seems to lead in the same direction. The pattern of naming described earlier suggests that most courtly couples in this textual world consist of a named man and an unnamed woman. The first variation that both realizes and violates that pattern brings together the names of Marie and Guigemar, author/narrator and character/lai. The

184

Chapter 5

last variation presents us with a story in which all the characters are named and arranged in various couples, culminating in the final unit of two within three. Who is there to remain unnamed if not Marie, the author/ narrator who forms a couple with her work at its moment of completion? Even as the names of two female characters take the place of the woman author's name at the close of the collection, those names return us at the same time to the collection's opening—and to Marie's own name, anagrammatically inscribed in Guigemafs.37 If we return now to compare Marie's use of names and titles for author and Ms in the Prologue to Guigemar and in EliduclGuildeluec ha Guilliadun, we perceive a pattern of replacement or relay: the introduction of Marie's name and her authorial concerns delays the introduction of Guigemar's title until after the aventure has been told; on the contrary, where two womens' names appear as the most appropriate title for the adventure about to be told, Marie as author stays in the background, speaks only briefly through the conventional role of the oral storyteller. This chiasmic pattern sets up two "poles"—author's name and story's name—that are intertwined in the course of the collection, as Marie lends her voice to her characters, allowing them to speak through her, as she makes her identity as author and narrator speak through the story matter she reinvents. It may no longer seem so surprising then that, in the couple Marie forms with her work as a collection of twelve lais, it is she who is named, the collection unnamed. Each of its separate parts, of course, has a name, already supplied by the tradition of storytelling, but her assemblage can be named only by reference to the author's own name, a name identified, as Denis Piramus indicated, with the lais that so delight counts, barons, and knights—and especially ladies, for they are written according to their desire (u[q]u'il sunt sulum lur volente," v. 48).

Marie's Fusion of Voices Marie's apparent silence as writing author in the closing gestures of her collection is not the sign of an absent author, but rather an indication of how successfully she has fused her voice with that of her characters, how successfully she has taken on the character—in her writing—of the kind of storyteller who speaks to us directly, the mimic whose changing tones and inflections make different personalities come alive before us. Using the ambiguities of Old French grammar and the considerable resources of

Textual Identity and Marie de France's Lais

185

style indirect libre ^ Marie inscribes within her very language a series of fluid boundaries between her world and that of her characters, boundaries that are, in fact, passages through the narrator's voice that allow direct communication between these two worlds. Modern readers may be surprised to find a Flaubertian technique like free indirect discourse already put to sophisticated use in a twelfth-century text.38 It is surely one of Marie's most distinctive narratorial ploys, and though we can find other examples here and there in anonymous lais or in contemporary romances, she seems alone among twelfth-century writers in giving it a crucial role to play in defining the relationship between narrator and characters.39 Repeated examples show that Marie's narrator typically starts with indirect discourse and then moves gradually into style indirect libre (e.g., w. 48-51,122-28, 246-52), sometimes culminating that movement with a final remark reported in direct discourse.40 What is striking in these passages is the fluid boundary between speakers. A verbal sleight of hand, we recognize it only retrospectively when one voice or the other is clearly separated from the temporary fusion. The blended voices set up a channel of direct communication between narrator and character without suppressing the difference between them, as Jean Rychner has indicated in his analysis ofLanval: while the characters seem to have a life independent of the narrative, the narrator is able to present them through style indirect libre in a way that retains their "paroles d'origine, tout en les ouvrant, le cas echeant, a son affectivite."41 A specific example may serve to illustrate how style indirect libre also allows Marie an economy of style well suited to the abbreviated dimensions of the te, even in a lai as long as Eliduc. When Eliduc's lord sends for him to return, the narrator tells us: Ses sires 1'ot enveie quere Treis messages fors de la tere. Mut ere grevez e damagiez E encumbrez e empeiriez; Tuz ses chasteus alot perdant E tute sa tere guastant. Mut s'esteit sovent repentiz Que il de lui esteit partiz; . . . Par 1'ali'ance qu'il li fist Quant il 1'umage de lui prist, Que s'en venist pur lui aidier,

186

Chapter 5 Kar mut en aveit grant mestier. Elidus oi la novele; . . . (w. 551-58,567-71) (His lord had sent out three messengers to look for him. He was being set upon and injured. He was losing all his castles and all his land was being laid waste. He had often regretted that Eliduc had left him; . . . he summoned and required Eliduc by the promise he had made when he had accepted his homage to come and help him, for he stood in great need. Eliduc heard the news; . . . p. 118, slightly modified)

We learn a bit later (v. 595) from Eliduc's own remark that his lord sent him a letter. In the passage just quoted, a report of the letter's contents, given in style indirect libre and apparently offered by the narrator to her public, begins at line 553, after mention of the three messengers. As the lord's message is spoken through the voice of the narrator (w. 553-70), it is delivered simultaneously to us and to Eliduc, as we learn retrospectively at v. 571. Somewhere in the course of the description, the messengers found Eliduc and delivered their written message, which someone read out to Eliduc. All these actions are assumed and economically carried out through the contacts established by the narrator's voice, blended first with that of the seigneur, then with the messengers'. Where the substitution takes place we do not know, but the verbal magic accomplishes the passage between them and reaches Eliduc with the voice of the message speaking directly to him.42 After briefly describing Eliduc's anguished reaction to his lord's recall and the predicament he finds himself in, the narrator allows the character to speak directly, as he recapitulates in his monologue events that have led up to his present state and anticipates what can happen next (w. 585-618). Eliduc's direct discourse includes the kind of information-giving we elsewhere associate with the narrator's (re) telling of events, as he recounts for himself his love for Guilliadun, the letter from his lord, his wife, and so on (w. 588-97). We are thus reminded that the economy of Marie's tales does not exclude a significant amount of repetition or retelling of important events, recounted either by the narrator or by the characters. The common model of retelling is one of the major links between Marie and her characters; it constitutes the major parallel between them that both authorizes and invites the fusion of voices we can hear in the narrator's use of style indirect libre. They speak the same language and use it for the same purposes.

Textual Identity and Marie de France's Lais

187

The repeated use of renarration in the stories invites us over and over again to cross the boundaries between enonce and enonciation, as these worlds communicate directly through the narrator's speaking voice. By the time we read the last lai of the collection, Marie no longer needs to point out such parallels directly, as she does across the gender difference with her first male hero. As women who act and speak out, Guildeluec and Guilliadun, heroines worthy of giving their names to the title, embed those parallels implicitly in the working out of their story and thus ask us to play out their import, in order to understand how Marie has accomplished what she announced in the General Prologue. Indeed, the characters offer models for retelling to their author/narrator, as much as she does to them. And these models inform us about the problems involved in moving from adventure to discourse, whether for the characters within the fiction or for the narrator transforming their stories into writing. A comparison of Eliduc's renarrations with those of Guildeluec and Guilliadun suggests both the limits and the enormous potential to be realized through the act of retelling. Eliduc illustrates a negative model for renarration. His good intentions to the contrary notwithstanding, he is unable to tell Guilliadun the whole truth about his past, nor can he reveal the cause of his unhappy silence to his wife when he returns. Once Guildeluec takes matters into her own hands, however, she quickly arrives at the truth, first by seeing Guilliadun, then by questioning her about the past. With the women's exchange, discourse of the past and discourse of the truth now coincide: the tangled sets of relationships can now be resolved by the heroic action of Guildeluec. As in the Guigemar recognition scene, it is through the language of retelling that the past is moved into the present. If that retelling is truthful and complete, it can complete the process of selection and substitution, completing the story that will circulate among Breton storytellers until Marie, too, continues the tradition in a new language and form. Success or failure in the act of retelling, as represented in the last lai, coincides with a division along gender lines. If such a model suggests that Marie recognizes how gender affects the action of speaking out, the many examples of retelling included in the collection as a whole demonstrate that Marie's awareness of the complexity of this process is not ideologically predetermined. Not every woman is superior or honest in her retelling; nor is every man incapable of full and complete renarration carried out in the appropriate context. The examples of Tristan in Chiewefoil and the wife of Bisclavret come to mind as counterexamples to any simplistic di-

i88

Chapter 5

vision between the sexes.43 But it does seem significant that Marie, who offers the rare example of a woman writing in the vernacular in the twelfth century, closes her collection with a positive and emphatically female model for speaking out and retelling. All those characters who transform their past into discourse thus furnish Marie with both positive and negative models for her own project as narrator of the tales and author of the collection. Some of them, like Guigemar and Guildeluec/Guilliadun, retell their past only within the fabric of the story; some, like the lady of Chaitivel or Tristan in Chievrefoil, appear more directly as models for taking the next step, the transformation of speech into the artistic form of the lai. Of course, there are significant differences still to be noted in these parallel projects. Unlike Tristan, Marie is not the hero of her lafs adventure; like storytellers and narrators of romance, she tells the adventures of others and her truthfulness will be measured (at least in part) by her fidelity to the tradition she has heard and passed on in written form (cf. Thomas's famous comment in the Douce fragment, w. 835—84, on the diverse nature of the Tristan matter and its competing versions). On the other hand, we might say that Marie is the hero of a new and different adventure, insofar as she is the first to set the lais into writing and to gather them into a collection. This is the adventure she describes in the General Prologue and then portrays implicitly in the lais themselves. But there are still other facets to be recognized in the differences between the artists Marie and Tristan that mark both continuity and discontinuity. Tristan is a harper (v. 112), who in Chiewefoil makes a new lai to remember the words exchanged with the Queen.44 Marie's written translatio can refer to, but it cannot reproduce the music of Breton lays, played on harp and rote (G, v. 885). While her use of rhyme and verse (P, v. 41) may be considered an effort to transpose the musicality of the lais, we can no longer hear their original melodies. We may use the same word for both the musical compositions and Marie's narratives (although Marie herself seems to differentiate them in most instances); we cannot forget, however, that they are different kinds of artistic objects, related through their common bond to the original adventure. In the process of translation from one to the other, not everything can be said. The problems of retelling dramatized in Guildeluec ha Guilliadun/Eliduc will recur in the author/ narrator's relation to her material, but they will be transposed and multiplied by the problems of moving from the truth of the adventure to the obscurities of writing about it. This observation suggests a need to con-

Textual Identity and Marie de France's Lais

189

sider another aspect of Marie's narrative voice, as it fuses with that of her characters. To capture in writing the truth of their aventures, Marie lends her voice to preserve and substitute for the multiple voices of the past, now collected together and presented to the King in a form that also preserves them for future readers. She initiates a process of retelling that combines orality and writing, as it recreates and transforms the oral voice of the storyteller through the narrative voice of the written text.

Orality and Writing, Orality in Writing What are the characteristics Marie associates with oral storyteller and clerkly narrator? How does she acquire or transpose them into written discourse? Some of the features associated with writers—or philosopher poets, as they are identified in the General Prologue—have already been mentioned in describing how Marie establishes her identity and credentials with respect to romance tradition, that is, a tradition of writing carried out by school-trained cler.45 Marie sets up in her opening verses both her skills as cler, fully capable of deploying the rhetorical topoi of the prologue tradition, and her difference in relation to the matter chosen. She introduces herself as a writing author, but paradoxically an author who works on oral, not written materials. It is certainly appropriate that the initial topic in the General Prologue, even before Marie raises the issue of writing, refers to the acts of speaking and listening as essential to the circulation and flowering ofjjranz biens: Ki Deus ad dune esci'ence E de parler bone eloquence Ne s'en deit taisir ne celer, Ainz se deit voluntiers mustrer.46 Quant uns granz biens est mult oiz, Dune a primes est il fluriz, E quant loez est de plusurs, Dune ad espandues ses flurs. (w. 1-8) (Anyone who has received from God the gift of knowledge and true eloquence has a duty not to remain silent: rather should one be happy to reveal such talents. When a truly beneficial thing is heard by many people, it then enjoys its first blossom, but if it is widely praised its flowers are in full bloom, p. 41)

I9O

Chapter 5

Showing one's talents, a topic we can trace back to Biblical and Latin books, is here connected to the kind of exchange Marie evokes later in the Prologue when she describes her reception of the Breton lais: speaking and listening are the links that keep the lais circulating between oral storytellers and their public. If Marie appears pen in hand in the General Prologue, she quickly assumes the persona of the oral storyteller as well. Over and over again, she uses the verbs oir, cunter, dire (hear, tell, say), to describe her contact with the tradition, as well as with her own public. Like her Breton models, she does not want to forget (P, v. 40; Bis, v. 2; Chait, v. 1-2) these unforgettable tales, made for remembering (P, v. 36; Eq, w. 7-10; Bis, v. 317-18; Chievre, v. in). It is Marie's voice telling them that brings us into the process of remembering, brings their truth to our ears. The prologues and epilogues, like a repeated refrain, remind us of the storyteller's presence, as she speaks directly to us, ritually at each opening and closing, sporadically in the direct interventions that appear in the lais. Within the narrative of the tales, Marie blends her voice with that of her characters and seems to let the stories tell themselves, as if the characters now spoke to us directly. Such is the effect of those fluid boundaries described earlier between narrator and characters that we may mistake as an absence of art the transparency of the narrator's voice as she foregrounds the voices of her characters. That at least is frequently the reaction of my students reading Marie for the first time, especially if their expectations as readers have been based on the romance tradition Marie at once associates with and dissociates from her project. Unlike the witty narrator of Chretien, who likes to dazzle us with his rhetorical techniques in commentaries on love or descriptions of beds, or Thomas's narrator, who elaborates long discourses to follow the minute play of his character's inner feelings, Marie as storyteller strikes us as a discrete voice, always present, but never battering our ears with narrative commentary or extended rhetorical play, generally limiting her first-person remarks to short formulaic phrases ("ceo m'est avis," "tant vus di," "mun escient," etc.), as befits the brevitas of her chosen form. Marie tells the story as it happens to her characters, with little explicit commentary and only occasional anticipation or flashbacks. In quantitative terms, however, while the amount of narrative intervention practiced by different types of narrators does indeed differentiate Marie's persona from that of romance narrators, it suggests overall that she speaks out more frequently than most romancers in telling her characters'

Textual Identity and Marie de France's Lais

191

adventures. If we compare Marie's Lais with other contemporary lais, as well as with contemporary romances, there is a clear division between the two types of stories in the overall percentages of narratorial interventions (which individual lais and romances can and do cross over). Marie's first person, like the narrators of lais in general, is much more present as a storyteller relating her tale than is the typical romance narrator, whether we consider the use of prologues and epilogues or the number of interventions within the stories themselves. In this respect, Marie is careful to identify herself with the tradition of oral storytelling. But, of course, she does so in writing: the interplay between orality and written discourse recalls their problematic relationship in the Folk Tristan d'Oxford. Though Marie certainly does not seek to hide the written character of her work, she foregrounds the oral tradition she claims as its source, as well as its continuing vitality.47 In order to appreciate why Marie gives precedence to orality over (but in) writing, at least according to some of the explicit claims of her narrator, we need to formulate another set of questions, even as we continue focusing on the specific features associated with each type of storytelling: What is to be gained or lost by associating her project with oral or written traditions? How does Marie represent these two traditions and how does she explicitly distinguish them? They both share a process of transmission over time that establishes an unbroken connection between a point of origin in the past and the current recipients of the tradition, who will in turn pass it on to those who follow. While the point of origin for the written tradition, as well as the process of transmission, is textual, that of the oral tradition is a human experience, lived only once, but kept alive by an unbroken series of human voices who retell it pur remambrance (P, v. 35). Memory and voice are the agencies of oral transmission; writing and reading are the primary operations of a textual tradition. Here we have begun to reach significant differences that separate oral and written traditions against the background of their commonality.48 These may be played out further, by considering how Marie connects the oral tradition to the notion of truth, while the written tradition entails rather the problems of meaning, obscurity, and interpretation.49 If we compare Marie's statements about lais, contes, and aventures, as they appear in the General Prologue and in the framing narratives, we learn that truth is something to be associated with the original adventure. In the Prologue to Guigemar (w. 19—20)—which operates as part of a general introduction to the collection—Marie attributes truth status to the stories that inspire the lais

192

Chapter 5

of the Breton storytellers. The same connection is affirmed in the prologues or epilogues to three other lais—Eisdawet, w. 315-16; Chievrefoil, w. 3—4,117—18; Eliduc, v. 4—while within the stories themselves the truth topos appears frequently, invoked either by the narrator or the characters, to affirm the truth of events described or words exchanged.50 The originary status attributed to the characters' adventures makes them a locus of truth that must be maintained and passed on. Despite the repeated claim, however, the public hearing or reading these stories might be inclined to question the truthfulness of Marie's Ms (cf. Wace on the mixture of truth and lies in the tales about Arthur). Denis Piramus explicitly denied any truth to the lais: u[k]e ne sunt pas del tut verais" (v. 38). The fictionality of the Lais is an important feature, for the medieval public as well as for modern readers. But if for the moment we accept at face value the idea that an adventure truly occurred at a given point in the past, we can begin to appreciate why Marie assumes as much as possible the persona of the oral storyteller. The tales may indeed be fictional, but the hypothesis of their truth functions as a link between the lais and human experience, whether we locate it "before" the text in the characters' lives, or "after" in the audiences', or even "during" in the adventure of the author Marie.51 The claim to found truth on lived experience, which Marie makes for the Ms bretons, strikes an important chord in the medieval context, where the truth of eyewitness accounts is a standard topos of many different traditions. The authority of eyewitness testimony marks the Vie de Saint Alexis, with its letter written by Alexis himself, no less than the Roman de Troie, which leaves aside Homer's version and prefers Dares's and Dictys's reports, based on what they actually saw and experienced. Among the Lais, only Chaitivel and Chievrefoil describe how the characters' adventure is turned into a lai by the lovers who experience it. But the repeated topos of truth claims for all lais that their account of the aventures faithfully corroborates the truth of lived human experience. If this truth originates in a specific adventure of the past, then the duty of the storyteller who remembers and passes it along is at once both active and passive—actively lending a voice to make the characters and their story come alive in the presence of the listening public; passively preserving intact, without change, an original truth that is embedded in the adventure itself. According to the claim, that truth is established before the story is retold and must cross over unchanged through Marie's translatio, just as it did in the generations of oral storytellers that preceded her. The presence of Marie's voice makes the truth of the past present once again and allows us to

Textual Identity and Marie de France's Lais

193

recognize it, in a way that recalls the strategies of the Folie d'Oxford. Marie thus emphasizes the oral character of her narrative persona to carry out and affirm her unbroken continuation of the storytelling tradition that commemorates the original aventure. But she is, of course, not Tristan speaking in his own story or about his own adventure. Marie has established her voice in a written text. While her explicit claim to do something different, not romaunz, reinforces the orality of her project, Marie at the same time implicitly recognizes certain advantages associated with writing, advantages that must compensate for the inevitable change introduced when a truth associated with human experience and oral tradition is transposed into writing, a change recognized perhaps as much as it is denied by the reiteration of the truth topos. The question of writing takes us directly to the most problematic passage of the General Prologue—problematic, that is, for the tradition of scholarship that has tried to explain exactly what it is Marie is telling us in verses 9—22 about the issue of glossing and how it applies to her own work. The General Prologue to the Lais, and this passage in particular, has received enormous critical attention.521 do not expect to revolutionize our understanding of the issues here—much of what follows will be familiar to scholars or general readers who have lingered over the questions raised—but I think it useful to establish them in the present context, which aims to follow the dialogue between written and oral traditions, as it reflects the same kind of dynamic described earlier in terms of selection and substitution. Custume fu as anci'ens, Ceo testimoine Preci'ens Es livres ke jadis feseient, Assez oscurement diseient Pur ceus ki a venir esteient E ki aprendre les deveient, K'i peussent gloser la lettre E de lur sen le surplus mettre. Li philesophe le saveient, Par eus meismes entendeient, Cum plus trespassereit li tens, Plus serreient sutil de sens E plus se savreient garder De ceo k'i ert a trespasser. (P, w. 9-22)

194

Chapter 5 (It was customary for the ancients, in the books which they wrote [Priscian testifies to this], to express themselves very obscurely so that those in later generations, who had to learn them, could provide a gloss for the text and put the finishing touches to their meaning. Men of learning were aware of this and their experience had taught them that the more time they spent studying texts the more subtle would be their understanding of them and they would be better able to avoid future mistakes, p. 41.)53

The problems raised here concern the auctoritas of a written text and the location of its sen, complicated by obscurity and requiring interpretation to arrive at its surplus. The character of a written text's authority differs considerably from that of an oral tale based on lived experience. The word auctoritas itself calls attention to the auctores, the Latin authors like Priscian who guarantee the authority of the written word, anchored in Latin tradition.54 Marie's allusion to the anciens^ Priscian, and liphilesophe, attests her knowledge of such textual authority and thus associates it with her own written work. The later emphatic distinction she makes between her project and the proliferating translations of Latin texts may implicitly recognize a potential conflict within medieval culture between these two kinds of authority, textual and experiential. Medieval clerical authors showed significant skepticism about certain stories, not only the Ms scorned by Denis Piramus but also tales of saints and miracles not based on "unimpeachable written evidence."55 Given clerical writers' desire to keep their monopoly of literacy for themselves in Latin, they also tend to crystallize and align two sets of oppositions: clerical and lay and litterate and illiterate.56 While these distinctions are often invoked by modern scholars, as well as medieval clerics, the limits of their usefulness must be carefully defined, since the categories may not, in fact, cover all the complexities of what we actually find in the Middle Ages. Just as Julia Smith has demonstrated the overlapping of written texts and oral traditions in popular saints' cults, we need to consider how vernacular literature—and especially Marie de France in her Lais—cuts across the oppositions between folkloric culture and clerical culture.57 The Lais clearly represent an intersection of traditions and materials, just as Marie herself does. As a collector of oral tales in writing Marie already crosses over the boundaries implied by the oppositions just enumerated. As a woman, she adds yet another set by crossing over the usual distinction between male cleric and female layperson. Her identity as fe-

Textual Identity and Marie de France's Lais

195

male clerk receives little explicit attention in the Lais, although we might see it reflected in at least two areas: (i) When the problem of gender attached to Guigemar seems to surface discretely in Marie's text in relation to her own identity, she specifies "Humme u femme de grant pris" (v. 7) as the objects of envy. Her usage highlights by contrast the more typically male gesture of generalizing in the masculine mode. (2) Earlier in the General Prologue, Marie's description of her hard labor and sleepless nights associates her work with a quintessentially female action through the imagery of birth: Ki de vice se voelt defendre Estudier deit e entendre A grevose ovre comencier: Par ceo s'en puet plus esloignier E de grant dolur deliwer. . . . Rime en ai e fait ditie, Soventes fiez en ai veilliel (w. 23-27, 41-42; emphasis added)58 (Anyone who wants to guard against vice should study with the intention of beginning a weighty work. In this way one can keep far away from vice and give birth from great pain. . . . I have put [adventures heard] into rhyme and made poems from them, often staying awake through the night.) Marie represents this as a birth not of her own child/text, but as the rebirth of her characters' stories. Like other writers of the twelfth century—and the Middle Ages in general—she is not interested in a new conception, not a creation ex nihilo, but a reinvention of one story from another. That does not make her any less proud of her role as midwife to a textual rebirth; the General Prologue itself stands as ample testimony to the importance of authors in the textual process and the care with which she has considered and carried out her own role, located in the crisscross of oral and written traditions. The masculine forms used in w. 53 and 54 of the General Prologue by the scribe of the Harley manuscript—rejected readings in Rychner's edition—have been seen by some scholars, like Baum and Jean-Charles Huchet, as a clue suggesting there was no real woman author operating here. Most readers of the Lais, however, appear to assume that Marie's identity as female cler informs her writing, as it concerns her readers.

196

Chapter 5

Denis Piramus pointed out how her lais had special appeal for the female public, though they were enjoyed by both men and women at court. Given the impetus provided by feminism, more recent studies have tried to specify how Marie demonstrates a particularly feminine style or subject matter in the collection of lais.,59 What concerns us here more particularly is the way Marie's unusual combination of textual identities as both woman and cler provides another clue to understanding why she both claims and disclaims the authority of written tradition in a literary project that sets into writing the oral tales of the Breton conteurs. As woman, she figures more readily the realm of experience associated with the truth of the lais-, as der^ she brings to bear the auctortias of written texts and Latin tradition. As female clerk, Marie represents an unusual (and for some no doubt, an uneasy) intersection of both kinds of authority. But the authority of a written text does not eliminate the problems introduced by its obscurities. If we return to the passage quoted above and to the question about a text's meaning, we may notice that the ambiguity of syntax and Marie's use of Priscian as an authority allow us to read these verses in a number of different ways, but do not seem to give us one, authoritative reading. Is the obscurity intentionally left in books by the Ancients or is it the unintentional result of time passing by? Does the surplus belong to the Moderns or are they simply supplying what was already anticipated by the Ancients? Does each individual become more sutil de sens in the course of a single lifetime or do philosopher poets gain that advantage across generations? Tony Hunt has given a very convincing analysis of the exegetical tradition that must lie behind Marie's statements in the Prologue.60 If, according to Hunt (pp. 411-12), Marie may be alluding in w. 9-16 to a well-known idea of St. Augustine—namely, that Christian writers are intentionally obscure in order to exercise the wits of future exegetes—then Marie does seem to be speaking of intentional obscurity. The job of a Modern is then to unearth the hidden meaning planted in texts by the Ancients; the added gloss simply reveals what is already there inpotentm. On the other hand, Marie may be using Priscian as an example of someone who wrote obscurely without necessarily meaning to do so, a grammarian surpassed by the modern practitioners of the art.61 If so, then the surplus missing in the original text is supplied by the later glosses. Our understanding of the possibilities is certainly improved by knowing the exegetical background and the range of possible allusions. The alternatives are clarified, however, without clarifying the choice between them; neither one prevails, since both are equally plausible given

Textual Identity and Marie de France's Lais

197

the exegetical context and Marie's own written obscurity. The Gordian knot of complexity has not been cut. What strikes me when faced by such choices is that we need not choose one option over the other: both possibilities leave intact the necessity to gloss, interpret, add to previous texts. Moreover both offer fruitful ways to understand how Marie's text operates. The inherent circularity of this dilemma—scholars engaged in the activity of glossing those very verses whose subject and object is glossing itself—points out the selfreflexiveness of the Lais and reminds us that already in the Prologue Marie has entangled us in the kind of activity she herself engaged in to write the Lais, the same kind of activity in which future readers of the Lais must participate to perpetuate and commemorate the tradition.62 Such activity seems inevitably to entail the intersection and interdependence of categories and oppositions that in other cases may be clearly separated, but which are here constantly intertwined. This purposeful use of crossovers and ambiguity recalls the kind of fluid boundaries I described in Marie's use of style indirect libre.63 It serves her quite well here, if we consider the kind of counterpoint she sets up between truth (associated with an original adventure, but passed along and recreated through the voice of oral storytellers) and meaning (associated with the obscurity of a textual tradition that finds the surplus de sen at its point of origin and/or in successive generations of interpreters). The combination of meaning and truth thus yields one fixed pole, one changing, and reminds us of the complementary interplay of selection and substitution combined in Marie's translatio: selection seeks to get to the truth, find the right match, the unity of the couple; substitution explores the process of change, opens onto a world of diversity. Against the common background of the collection, we are thus invited to discern the difference, the truth of each individual adventure. But the inclusion of twelve lais in a written collection really makes the combination of truth and meaning more complex, inasmuch as it multiplies the truth by twelve. The "fixed pole" of truth offers a kind of multiplicity that recalls and anticipates the multiplicity of the textual tradition and its surplus of meanings. If, as I suggested above, we suspend our disbelief in order to play out the consequences of Marie's claim that truth is connected to the original adventure, then the truths Marie retells in her Lais are as specific as the unique human experience of each set of lovers. Each adventure included has its own, irreducible truth, which cannot be abstracted into philosophical or theological concepts; it does not refer

198

Chapter 5

metaphorically to another, transcendent level, but remains embedded in the working out of the story. The diversity of human truths represented in the collection of twelve tales cannot be reduced to any simple unity. Although there is clearly a general dimension to the lovers' experience—the recurrent elements of the Lais give them an identity as a whole—it is against that sort of unity that each couple strives to establish its own difference.64 Selection strives to overcome substitution, until the final point counterpoint of Eliduc finds a resolution that balances both processes. In the collection, as in each te', we watch the play between unity and diversity, as we seek the specific truths embodied in each aventure^ the specific character of Marie's own adventure as author and storyteller.65 The uni/verse represented in Marie's collection of Lais would have been impossible in a single lai. The insertion of the oral into writing parallels the insertion of the lai into a collection, as the two-dimensional link between oral tradition and faithful storyteller takes on the multifaceted dynamic of a retelling in written verse. In order to represent "[l]es contes que jo sai verrais" (G, v. 19), Marie needs the diversity of a collection that elaborates a broad, horizontal dimension for each of the tales through the interplay among them. The linear reading of the entire collection adds a syntagmatic level to the vertical, paradigmatic level described in terms of each la?s link back to its original aventure. The horizontal or syntagmatic level of the collection creates in turn a different kind of paradigmatic play: the recurrent elements of characters, situations, plots, and so on, like a set of themes and variations, suggest the general character of human experience, its substitutability, and at the same time highlights, against the common background of human stories, the uniqueness of each aventure^ its selection of a particular combination of shared features. Its placement within such a collection, where we can move back and forth among the twelve tales, allows each lai a kind of breadth and depth not available in a single, isolated lai—or available as a potential realizable only when the intertextual dialogue among lais has been supplied by the listener's experience. By locating that potential dialogue within her collection, Marie makes each voice speak with a wider range, a greater vibrancy, as it is simultaneously matched with and differentiated from the other voices included. Intertextual play is both intensified and multiplied: while her transposition from oral story to written text may complicate the translatio of the aventure^ original, spoken truth, by joining it to the obscurity of a textual tradition, Marie uses the capacity of writing to fix each lai in the presence of the others assembled, in order to magnify the potential of each

Textual Identity and Marie de France's Lais

199

lai to speak out across the collection as a whole. A plurality of voices from the past fixed in writing through the voice of the author/narrator and linked to the generations of readers and listeners who will remember their adventures and search for their meanings—such is the advantage Marie gains by joining the authority and permanence of a written tradition to the human experiences crystallized in the lais and commemorated by the oral storytellers. The amplification Marie achieves across the twelve tales of her collection recalls again the tradition of romance signaled in her General Prologue, and it points as well to the question of closure. The issue of glossing introduced in that Prologue pictures the obscurity of written discourse as an invitation to generate different interpretations in different generations, which may either be moving back to an original meaning located in the past, like the truths of the adventures retold, or moving on from surplus to surplus with the full meaning still projected toward the future. In either case, the completion of the written text's sen appears to be as open-ended as the process of storytelling itself, continually handed down from one generation of conteurs to the next. Given the multiple truths to be transposed into writing, which then multiplies the possible meanings of each aventure, what kind of closure operates in Marie's collection of lais> Obviously closure in the Lais cannot be treated as a simple choice between reading them as open or closed; it involves rather a complex understanding of how both categories apply, each in its own way.

Closing and Opening the Collection The question of closure is a complex one, as it operates on different levels of Marie's text. First, it is clearly amplified by the combination of individual lais, which multiplies the play between open-endedness and closure by repeating it twelve times. This is a recognizably canonical number for a Christian audience, although it need not necessarily carry a specifically Christian meaning for the collection itself. Like other Biblical numbers (seven, forty, etc.), it comes readily to mind in the medieval context and appears frequently in literary works: compare Roland and the twelve peers or the twelve boasts in the Voyage de Charlemagne. In terms of number symbolism, as a recall of the twelve tribes of Israel and the twelve disciples of Christ, the twelve lais of the collection may evoke the whole, as well as the diversity, of a people—a valuable reference for the human truths con-

zoo

Chapter 5

nected to the collection's adventures. But we should not forget that Marie has not only begun and ended twelve stories, she has also opened and closed the entire collection of lais. That dimension, too, requires investigation to see further what kind of closure characterizes her assemblage. Secondly, while Marie's Lais seem to enclose a whole world in miniature, the concept of closure may be seen working less metaphorically, in each lai as in the collection, on at least two levels: (i) the narrative in its linear organization (to what degree does the ending close or leave open the narrative syntagm?) and (2) the production of meaning (limited or not by the final configuration?). These two kinds of closure are necessarily related and may not always be so easily distinguished, whether in Marie's Lais or in the kinds of reactions they have provoked among different readers. Any discussion of closure necessarily entails a recognition not only of textual maneuvers, but also of the relationship between text and public.66 If we remain at the level of individual tales for the moment, even a cursory reading of the lai? endings reveals an enormous range of possibilities exploited to close the narrative syntagm. Here as in many other respects, Marie participates in the twelfth century's taste for artistic experimentation. The final event may reflect the ending of the story thematically by using one of society's rites of passage, standard moments of closure borrowed from human experience: reunion of the lovers in Guigemar and Lanval, marriage in Fresne and Milun, death in Equitan, Tonec, and Deux Amants. In other lais, the narrator's comments supply the ending: the description of the wife's noseless lineage in Bisclavret or the explanation about the titles in Chaitivel and Chievrefoil. But the events or comments that occur at the end can generate a sense of closure or, on the contrary, leave the reader with a notion of open-ended process.67 In fact, the very same lais sometimes give one reader a satisfied feeling of closure not shared by another: Philippe Menard considers most lais closed, except for Bisdavret, Chievrefoil, and Laustic. Paul Zumthor sees only Bisdavret as temporally unclosed, since the hero can still become a werewolf.68 The range of these comments suggests we need to focus not only on what occurs at the end of the story, but on whether or not we worry about what might occur next. Do we worry, for example, about the jealous, old husband reappearing to upset the union between Guigemar and his lady? Do we wonder about what happens to Milun and his lady's son, once his parents are united in marriage? We might imagine sequels to some of Marie's tales that would continue relating the lives of heroes and heroines, but it does not seem likely that Marie expects or inspires us to do so.

Textual Identity and Marie de France's Lais

201

In exploring the Lais' "triadic structure," Donald Maddox demonstrates how each lai achieves a firmly established narrative closure that contrasts with the indeterminate status of its meaning.69 This corroborates what Marie herself suggests in the General Prologue about her transposition of oral Breton tales into the medium of writing associated with Latin and romance. In her own terms, it is the truth of the aventure, embedded in the narrative itself, which generates a sense of ending, once the narrator assures us that she has faithfully passed it along. In its original particularity, the adventure is complete in itself as the lived experience of the lovers, successfully or unsuccessfully seeking union. Once that aventure is transported into story, first in an oral tradition and then, more significantly, in writing, its truth is both linked to the obscurity of a textual tradition and caught in the web of similar, but different truths embodied in the tales of the other lovers whose adventures make up the entire collection. Their meaning becomes an open-ended, elusive project that moves ever forward in time, as we seek to expand with our interpretations those concentrated, miniature recits, to identify their original truth and complete their (original:5) sen. Truth and meaning participate in the open-endedness of the lais and yet suggest that an elusive closure is at the same time articulated in the process of telling and retelling the adventures commemorated by the storyteller's art.70 Marie does not simply tell the adventures; she places each one within a frame, the short prologues and epilogues composed primarily of recognizable topics, used to associate the lais with a particular oral tradition: these typically include a selection of those repeated topics, expressed in a conventionalized vocabulary—cunter, aventure,, lai, remembrance, verrais, li Breton—and the title of the particular lai (sometimes translated into various languages). It is the correspondence, the perfect fit between the opening and closing segments of these frames that announces most emphatically and with great certainty the narrative closure of each lai. Marie thus reproduces in the written text the oral situation, in which "pour signifier sa fin, le message oral devra dire explicitement cette fin."71 Her usage exactly resembles what other storytellers do, either in oral performance or in other extant written versions that appear in the anonymous lais. These short prologues and epilogues announce not only the particular lais, Guigemar, Tyolet, and so on, but the narrative type: oral Breton tale. We have already seen that Marie introduces the opening topics of another narrative type, the learned Prologue of romance, before she opens the series of frames for the lais. Her work will obviously take advantage of the options available to writers (who can announce their endings in a variety of subtle

202

Chapter 5

ways), as well as participate in the greater explicitness associated with oral communication. As we follow Marie's strategies in this respect, we should keep in mind the specifics of a manuscript culture in the twelfth century, where most "readers" continue to be listeners following someone else's reading out loud. One thing that Marie does not announce explicitly in her General Prologue nor in any of the other prologues is the number of lots she has collected together and offered to the King. She speaks of her project several times in terms of lais in the plural (e.g., P, w. 33, 47, 51; G, v. 20; Tonec, v. i), but nowhere does she indicate how many she has assembled. How will we know when we have heard/read them all, unless Marie tells us? We may thus expect an epilogue to the collection, which would correspond to the General Prologue, and which would announce explicitly that (a) we have now reached the end of the assemblage and (b) the collection is now complete. There is no such explicit announcement, at least not at the level of the collection. There is, of course, the epilogue at the end of Eliduc, the second half of the frame which encloses the last "oral" tale. Can we or should we read it as an epilogue for the collection as well? That question again engages us in the distinction between oral and written traditions, the way each one typically ends its recits in the twelfth century, and the way Marie intertwines both types. You cannot properly close something unless it has been recognizably opened. As Frank Kermode has suggested, our sense of an ending will be created at least in part by the signals established in the moments of beginning.72 In that respect, our first impression is a sense of asymmetry between the opening and closing gestures of Marie's collection. In contrast to the kind of fanfare that marked the two beginning Prologues, the final epilogue, limited to only four verses, is indistinguishable from previous epilogues that close individual lais. Even if we hypothesize that Eliduc's epilogue has a double function, for itself and for the collection, we may be struck by its brevity. Again we need to verify such an impression with respect to contemporary practice, this time the practice of writing, since Marie's collection has emphatically situated itself as a written composition. The proliferation of Prologue materials, whether they reflect different moments of composition and/or the desire to inaugurate a work with special care,73 accords well with the kind of attention given to beginnings in the rhetorical tradition inherited and elaborated by the Middle Ages. We have already seen how Marie uses the topics of the exordium in the General Prologue to establish her mastery as both writer and innovator within the

Textual Identity and Marie de France's Lais

203

school tradition adopted and adapted by authors working in romaunz. Ernst Robert Curtius has pointed out that, while the exordial topics of medieval poetry could lean heavily on rhetoric, the topics of the conclusion attached to oral discourse (give a summary and make an emotional appeal to the public) did not so readily lend themselves to writing—hence, for example, the lack of conclusion in the Aeneid or the abrupt conclusion to Ovid's Ars amandi ("The game is over").74 We can find many more examples from romance, in which this kind of short epilogue functions essentially as an explicit (whether or not the scribe adds his own explicit to the manuscript). Floire et Blancheflor begins with a s6-verse Prologue and ends with a two-verse epilogue: "Ci faut li contes du roi Floire / Dieus nous mete toz en sa gloire" (w. 3038-39: "Here ends the story of King Floire. May God put all of us in his glory"). In Climes, although the Prologue's forty or more verses include an enumeration of all his works to date and an elaborate development of the translatio studi et imperii topos, Chretien's epilogue consists of a single verse: "Ci fenist 1'uevre Cresti'en" (v. 6664: "Here ends the work of Chretien"). Of course, we could also find numerous examples of more elaborate epilogues at the end of romances, which balance more or less the weight of their prologues (e.g., Ille et Galeron, Ipomedon, Partonopeu de Blois). The three romans antiques already anticipate the varied use of prologues and epilogues that will characterize romance tradition: Eneas has neither prologue nor epilogue; Thebeis thirty-two-verse prologue makes ample use of exordial topics, its 19 final verses draw out the moral of the civil war; and Benoit de Sainte More elaborates the Troie prologue through 144 verses, while concluding his thirty thousand plus verses with a sixteen-verse epilogue, brief in comparison with the romance's enormous length, ample in comparison with other epilogues. An abrupt epilogue, or even no epilogue in extreme cases, may not be unusual in the tradition of writing evoked by Marie in her General Prologue to the Lais. But we may still be uncomfortable that Marie ends her written text with an epilogue from the oral tradition of storytelling, and she does so in such a way that we might conceivably mistake the ending of her collection, if we are looking for the wrong kind of cues. We might believe, as Baum does, that it does not really exist, except as an anthology of lais, just as we have in ms. S a random collection, unordered by any author's hand, in which we will feel free to pick and choose what to read and when, without expecting any necessary internal correspondence shaping the group of lais bretons. We know from another work, that

204

Chapter 5

"Marie de France" knows quite well how to use the resources of an epilogue to close a collection in such a way that she guarantees recognition of the author and her work: Al finement de cest escrit, Que en romanz ai treite e dit, Me numerai pur remembrance: Marie ai num, si sui de France. Put eel estre que clerc plusur Prendreient sur eus mun labur. Ne voil que nul sur li le die! E il fet que fol ki sei ublie! (w. 1-8) (To end these tales I've narrated here / And into Romance tongue translated, / I'll give my name, for memory: / 1 am from France, my name's Marie. / And it may hap that many a clerk / Will claim as his what is my work. / But such pronouncements I want not! / It's folly to become forgot! Siegel, p. 257) The rest of the Epilogue to Marie's Fables (w. 9-22) includes the name of the patron for whom she undertook this book, information about the title, named for the man who translated the fables from Greek to Latin (Esope^ v. 13), King Alfred's translation into English, her verse translation into French, and a prayer that God may bring her to undertake a work that will lead to her salvation. If we look over these topics carefully, we recognize that most of them appear as well in the General Prologue to the Lais or in the so-called Prologue to Guigemar, including the concern about envious rivals and the naming of an author "[k]i en sun tens pas ne s'oblie" (G, v. 4). Marie does not need to repeat them in an epilogue to the Lais, "named" like the Esope for the author herself. As the earlier discussion of names and titles already indicated, Marie has given us what we need to identify the author and her collection in the interplay between the proper names given in the opening Prologues and in the final lai of her collection, as they connect adventures and lais, author and assemblage. That chiasmic pattern of naming furnishes the same kind of enclosing frame for the written collection that the short prologues and epilogues supply for the "oral" tales. If she has thus given us a way to identify the boundaries of her collection in a way that seems dangerously subtle, by comparison with the direct announcement we

Textual Identity and Marie de France's Lais 205 might expect in an oral reading of stories, Marie has warned us about the obscurities of a written tradition. We should remember that Marie is experimenting with the intersection of written and oral storytelling, mixing a plurality of voices with her own. In this sense the boundaries of Marie's text remain open to other voices of tradition, as much as they contribute to the distinctive character of her own voice.75 Marie's beginning is a beginning in medias res: even where she speaks to us directly as author, Marie does not speak in her own first person until v. 28; she places first sources, authorities, topics and proverbs, philosopher poets and Breton storytellers—through which, on which, she will work, making them speak through her voice, but allowing their voices to reach us as well. Likewise, Marie's closing statement is, in some sense, in medias res. In a curiously paradoxical twist, as she narrates the last led we see Marie appear very little in the oral guise of the storyteller, which she has cultivated so carefully throughout the collection. While telling the story of Eliduc, the narrator's first person intervenes as rarely as that of a romance writer. Marie thus implicitly affirms her identity as a writing author, just as she is about to end her role as "translator" and pass us back to the tradition of storytellers she describes one more time in the last epilogue of the collection: De 1'aventure de ces treis Li aunci'en Bretun curteis Firent le lai pur remember, Qu'hum nel deiist pas oblier. (w. 1181-84) (From the adventure of these three, the ancient and courteous Bretons made a lay to remember, so that it should not be forgotten.) The first line attaches this epilogue to the specific character of EliduclGuildeluec ha Guilliadun. The last three lines repeat the topics we can identify as the essential elements of a storytelling tradition upon which Marie has set the mark of her own textual identity, a long-standing tradition to which she now returns, so that the translatio may continue to move forward to future generations. The closure of Marie's text depends on the vitality of that open-ended process of transmission. But it depends as well on the kinds of boundaries she sets up in the Lais. It is obvious that we would not recognize Marie still today, if she had not reached us through the collection represented in

206

Chapter 5

the Harley manuscript. She would hardly have claimed so much attention if her voice had remained scattered among the anonymous voices of the Breton storytellers. From her work as author and collector, we have instead a fully articulated dialogue not only between Marie and the tradition of storytelling, but also among the twelve lais of her collection. Marie ends her collection on a point of internal equilibrium, even as she returns it, in medias res, to the ongoing process of traditional storytelling, now intertwined with the textual traditions of romance and the written text's invitation to gloss and interpret.76 Her tmnslatio across the boundaries of time and space, language and customs, brings the voices of the past into the present with their mysterious vitality intact. Through the particular voice of her own narrative presence, Marie allows us to recognize and remember not only the oral and written traditions of storytelling she has inherited but her own reinvention of both, as they are masterfully intertwined in her collection of Lais.

Conclusion

The implicit dialogue set up among the romance fictions analyzed in the preceding chapters invites some further articulation, in order to see how their common features, as well as the range of their differences, contribute to a better understanding of twelfth-century romance and its potential for reshaping. I offer by way of conclusion a comparison of the five texts, focused on two specific critical issues: first, the mirroring of diegetic and extradiegetic worlds, as they are connected through the triangular linking of author/narrator, story, and public, and secondly, the continuum between closure and open-endedness as they appear on a variety of formal and semantic levels. These two issues strike me as among the most significant for understanding how romance operates in the twelfth century. Their complexity requires the formulation of critical models sufficiently flexible to account for the repetitions and variations that characterize medieval rewriting, without distorting either the recognition of certain categories used by medieval writers (and still necessary for modern readers) or the process of transformation and intergeneric exchange that continually renews the models available. Medieval genres may not be easily defined according to modern generic notions, but verifiable differences do exist to distinguish different categories of medieval texts. There is no doubt that an epic hero like Roland would be as much out of place in romance as Lancelot would be in chanson degeste. On the other hand, it is equally clear that certain elements of romance may intersect with the epic world, and vice versa, as demonstrated by a romanesque epic hero like William of Orange (with his adventures as merchant in disguise or his love for Guiborc). Any generic models for medieval texts must recognize and account for these intersections, without necessarily labeling them as aberrant (as a model based on the Classical conception of genres would). It needs to account as well for the way some genres, like romance, might be more receptive than others to such a fusion of elements without necessarily falling into the categories of parody or burlesque. Hans Robert Jauss's analysis of the elements that constitute what he calls the dominant in epic, romance, and the novella,

208 Conclusion offers a valuable first step in breaking down the hegemony of a Classicallybased genre system, but it may be that the notion of the dominant itself needs further refinement with respect to the pleasures of "hybridizing" so frequently manifested in medieval texts.1 In some sense, we might say that the essence of romance is its protean adaptability, its experimental and hybrid nature (if we can imagine the term "hybrid" without negative implications) . The analysis offered here aims to highlight some of the most important features of romance, as well as the way those common elements allow constant play and revision. As a set of "theme and variations" on the two critical aspects selected, this summary may suggest the innovative potential characteristic of twelfth-century romance. Our working definitions of romance as genre generally separate the point of view of the author/narrator and that of the story matter, a separation unavailable in the context of chansons de geste, where the voice of the narrator expresses communal identification and participation.2 The romance narrator typically assumes a didactic stance; he is a der who translates his story, which is not his in any personal way, for a vernacular public eager for the exotic riches of Antiquity or the marvelous otherness of Brittany. Numerous prologues from the twelfth century corroborate this view of romance and the contract it establishes with a courtly audience.3 But once we move from abstract model to textual manifestations, we can also see how many romances both evoke and play with or against this generic potential, which in some sense sets a probability rather than a limit.4 Indeed, the group of works analyzed here suggests once again that variation is as essential to the definition of the model as repetition; if the latter establishes the links between different manifestations and gives them their family resemblance, the former just as significantly brings the model to life, at once affirming and contesting its heuristic value for writers and consumers of romance, who are constantly mvriting and rereading. The split between author and narrator in romance fiction invites all sorts of play with the romancer's typical stance: romances may both retain and suppress distance, as they complicate our response and channel it now through the author's, now through the narrator's engagement and/or disengagement. We are sometimes consecutively, sometimes simultaneously, disconnected from and connected to the world of romance, as the "real" world of the historical figures involved in romance production and reception (authors, patrons, public) is mirrored and transformed in that of the implied author/narrator and public (Gerard Genette's extradiegetic level) or that of the characters (Genette's diegetic level, which in romance must

Conclusion

209

at various times include the figure of the narrator as well).5 This mirroring passes through the variety of roles attributed to authors and narrators of romance, and their varying links to the story matter. Echoes between these different worlds arise from parallels and contrasts established either in terms of structure or in terms of content common to any set of two or more. If at various times the split between author and narrator is reduced to the point where we can speak of convergence, at other moments they move apart, especially as the fictional character of the narrator is elaborated.6 In the Prologue to the Charrete^ for example, we may easily conflate author and narrator, although the potential for a difference between them is already set up by the introduction of the name "Cresti'ens" (v. 25)—whose third person il controls the last five (or in Wendelin Foerster's edition, six) verses of the Prologue—after the first person narrator has spoken to us directly for twenty-four verses. We may catch glimpses of that difference in the claims of the narrator not to know whose comb Lancelot has spied (v. 1350), despite the Hospitable Damsel's own, truthful identification of the Queen as owner that follows shortly in his narrative (v. 1408—12). But we see it most forcefully in the parallels analyzed between Lancelot serving the Queen and Chretien as author first in service to his patron, Marie de Champagne, and then served in turn by the second author/narrator, Godefroi de Leigni. The parallels here are first based on structure, rather than content—love and prowess are the business of Lancelot, writing and shaping that of Chretien and Godefroi—but we can see the model of service operating in each of the couples' relationships, even if its domain shifts from chevalerie to dergie.7 And most particularly, we have seen how the problem of signs and their multiple interpretations bridges the differences between these worlds. Chretien as author and narrator may remain uninvolved in the story narrated per se, but his manner of narrating, as well as his role as author, involve his public in the same problems that occupy his characters. Thomas's narrator also tells us he does not know certain things—about love in particular—despite his appearance of omniscience in describing his characters' thoughts and feelings, as well as their actions. His claim to ignorance simultaneously accepts the equation between experience, knowledge, and identity, which operates for the lovers, and uses it to create a negative contrast between them: without personal experience of love, the narrator cannot judge the lovers and invites his public of lovers to do so in his stead. The gap created by the contradiction between the narrator's explicitly stated ignorance and his implicit demonstration of

2io

Conclusion

knowledge, shared with the public in the process of telling the lovers' story, sends ambivalent and ambiguous signals to the listener/reader, who is both distanced and involved by the narrator's example, on the one hand, and by his invitations to judge the lovers and find solace in their sufferings, on the other. This play with "aesthetic distance" finds further ramifications, if we consider Thomas as author in his intertextual relationship with other versions of the Tristan story. Thomas as narrator may not be a lover, but in his love of the Tristan legend, the author Thomas shows the same kind of ermnce and novelerie that he associates with Tristan the Lover: Thomas's gift for doubling, like Tristan and Iseut's drive to duplicate the experience of the other in the couple, is a principle that establishes both the unity or coherence of his version, as well as its diversity. The anonymous romancer of Partonopeu de Blois suppresses the distance between author and narrator by suppressing his name, even as he retains the duality of that fictional split and relocates it in his dual stance as clerkly narrator and lyric lover. His experiments in fusion do not erase the boundaries of romance and lyric, but rather play on their intersections, their transformations and redefinitions. The parallels between the diegetic and extradiegetic worlds are based here not only on a common structure of relationships, but on a common content as well: Partonopeu and Melior are lovers who function as positive and negative models for the author/ narrator and the lady for whom he writes. Although the Prologue signals the existence of a "real" author and (male) patron operating outside the fiction, and provides evidence for the narrator's persona as learned clerk, other comments both in the Prologue and especially in the increasing number of narratorial interventions throughout the romance and continuation tend to make the portrait of the narrator as lover converge more and more with that of his characters (and vice versa, if we consider the development of the Sultan as writing lover in the Continuation): he tells their story along with his own, as a means of winning the love of his lady. "Aesthetic distance" has been evoked and collapsed in this play of narrative personae, as the romancer of Partonopeu seems to respond to the kind of variations already in play in the Charrete and Thomas's Tristan^ and then "go them one better" in his own virtuoso performance. All three of these works illustrate how romance sends mixed signals that both distance and involve its public, not only in relation to the story told but with respect to the acts of writing and narrating carried out by the author/narrator. Romances are not mimetic in any simple way; they are careful to situate their stories at some distance from the public,

Conclusion

211

whether through the act of narration itself or in the choice of time and place for the stories recounted. But romances do represent real and complex problems of contemporary feudal society, which they invite their public to identify and tackle with the same inventive experimentation with which romance itself represents and transforms them through its fictional license.8 One of the chief ways in which romance achieves a sense of complexity is through the suppleness of the split between author and narrator and the range of possibilities represented in their relation to the story matter. The multiplication of layers opens a gap between story and public, while at the same time inventing a variety of bridges that bring them together through the personae of the author or the narrator or a combination of both. This multiplicity is further enhanced by romance's receptivity to the elements, types, or characteristics of other genres. If we consider the "I" who sings in a chanson degeste, the still singing but very different lyric "I" of a chanson courtois, or the oral and usually anonymous storyteller who speaks in the first person in short stories or lais, we recognize that each one has a particular persona created by the speaker's relation to the matter told or sung and generally fixed within certain narrow constraints. On the contrary, as a written genre read out loud, romance introduces the distinction between an author (usually but not always named) and a (fictional) narrator, who may take on a variety of different narrative stances that may include those available in the other genres just mentioned, along with the didactic persona of the cler most typically associated with romance narration.9 The anonymous author/narrator of Partonopeu, no less than the anonymous poet of the Folie Tristan d'Oxford and the named author Marie de France, know how to play back and forth across the "borders" of what is and is not romance, just as they know how those borders can be transformed and transgressed. Reshaping, whether on formal or semantic levels, is the essence of romance, as it puts into question itself and the world around it. If our view of romance through the author/narrator split is thus complex and multi-layered, what about the other two poles of the triangle set up with romance's public and matter? The implied audience of romance fictions (as well as the real public we can document through the names of patrons and authors) is first of all a vernacular one, courtly and aristocratic: the rise of romance (as the spread of troubadour lyric) is linked to the importance of some royal courts, but especially to those of the great seigneurs: among the patrons we remember not only Eleanor of Aquitaine

212 Conclusion and Henry II of England, but also Henry the Liberal, Count of Champagne, and his wife Marie, Philip of Flanders and Baudouin, Count of Hainault, Thibaut of Champagne and Beatrice of Burgundy, "[q]ui est de Rome empereris."10 No doubt many a smaller court participated in the enjoyment of romance, as we can see in Chretien's Chevalier au Lion, when Yvain arrives at the castle of Pesme-Aventure and discovers the lord's daughter reading a romance out loud to her father and mother (w. 5358-64). Maurice Keen has pointed out the equalizing effects of life at court in the Middle Ages: the great and the humble mixed together, once prowess demonstrated gave the deserving their entry into court; the common bond of chivalry linked together higher and lower levels of the nobility.11 If the same families supply both the ders who author romances and the public whose exploits they mirror (however directly or indirectly), it is not so surprising to see the link commonly made in them between clergie and chevalerie.12 In the aristocratic society of the feudal courts, education is as necessary as prowess for the development of the chivalric ideal. The romance audience we construct from the romances themselves suggests a public of connoisseurs, who can expect to enhance their enjoyment and pleasure to the extent that they recognize and play (along) with generic conventions and intertextual allusions, whether explicit or implicit, direct or indirect.13 Romance demands an active public: our reception is constantly involved in interpretation. Rereading parallels rewriting as the constant activity of medieval textuality in general and twelfth-century romance in particular. As connoisseurs we are necessarily distanced from the romance world, and yet we are drawn into that world to the extent that we go through the same processes and share many of the same concerns (linguistic, philosophical, cultural, and literary) as the author/ narrator and the characters. The stories of romance strikes us as both distant and close for a medieval audience. Jean BodePs well-known distinction of three matters, centered on Rome, France, and Brittany, appears in his epic Chanson des Saisnes, but it makes an interesting comment on romance eclecticism to recognize that all three are well represented and combined in various ways in romances from the earliest romans antiques on through idyllic romances, adventure romances, and so on.14 We might more usefully rephrase these three matters as Antique, contemporary, and foreign (Roman, Greek, French, German, Byzantine, Celtic, etc.), that is, a combination of materials from home and abroad, distanced by time or geography, but also as close as current events and the reports of travelers returned from exotic

Conclusion

213

ports, whether as pilgrims, crusaders, merchants, or scholars. Fiction is in some sense the glue that holds together all these diverse matters, allows them to communicate freely with each other, as the marvelous pleasantly removes some of the constraints of reality and introduces the freedom to experiment. We have seen how Partonopeu de Blois is particularly emblematic of the romance genre in this respect, as it combines the fall of Troy and French history going backward from the Merovingians, contemporary politics concerning nonnoble advisors, the splendors of Constantinople as described in twelfth-century travel reports, the Greek story of Cupid and Psyche rewritten as Celtic fairy mistress in love with a human boy, school debates and bestiary lore, pagan versus Christian conflicts to the north and the east, the Biblical model of Adam and Eve in the Garden of Eden, and the springtime renewal of lyric poetry, indeed vernacular literary materials of all sorts (from lyric, epic, romance, lais, and so on), borrowed and transformed in the splendid fusions of this twelfth-century summa. The intersections of history and romance as we see them in the twelfth century redefine the boundaries between truth and lies, fiction and truth. Or rather those boundaries are defined differently in medieval terms easily misconstrued by modern readers who simply assume that our current habits of thought are shared by the Middle Ages. Romances may scramble oppositions, but do not eliminate distinctions: the categories of fiction, truth, history, and narrative, remain essential to their effective functioning in non-disjunctive oppositions. These serve to represent the complexity, the ambiguities, and the contradictions that necessarily arise when human beings try to translate into practice the ideals and conflicting value systems they represent in terms of King Arthur, a historical figure elaborated and transformed by the power of fiction in order to experiment with the present through the "model" of a fictive, yet historical past.15 Such a conception of romance seems designed to persuade its public that the problems raised in the pursuit of a secular ideal are real in human and universal terms. They both require and deserve our most imaginative and informed efforts to explore them. Even if such experimentation may or may not yield possible solutions, we will have gained the pleasures of romance, the delights of adventure along the twists and turns of the "droit chemin." In order to explore some of the real problems fictionalized in romance, we need to focus on the second critical issue introduced for discussion, the continuum between closure and open-endedness as it involves not only the formal shaping of romance but also the shape of the questions

214

Conclusion

it raises. We can identify three loci for measuring closure: (i) in the text itself (where we need to consider not only the plot or narrative syntagm but also rhetorical and semantic aspects that may reach across the borders of the individual text), (2) in the context (provided by literary practice, social conditions, contemporary culture, ideologies, etc.), and (3) in the reader/listener (whose habits are no doubt themselves conditioned by a multitude of factors depending on location in history, literary training, esthetic orientation, cultural and social experience, etc.).16 The recognition that a text may be closed or open with respect to several different, sometimes overlapping domains, as well as a variety of different levels, leads to the further perception that these two "poles" are more usefully conceived as the two ends of a continuum, where one shades naturally or imperceptibly into the other. Closure and open-endedness, like so many other oppositions in romance, are not really incompatible; they work in tandem like non-disjunctive oppositions. Perhaps rather than speak of a continuum, we should find a three-dimensional image to convey the way romance fictions may be simultaneously closed and open, depending on what factors we consider. In terms of the narrative syntagm, we have seen how certain moments in human experience furnish a sense of closure—marriage or death, for example—or how that sense of closure may be induced in the reader, who recognizes a return to an earlier point of departure or the fulfillment of an event announced. Rhetorical strategies may reinforce such "natural" closure, or they may do just the opposite, either by presenting an otherwise unsatisfactory openness or irresolution on the narrative level as a satisfactory ending or by making us ask if the ending so insistently proclaimed as such really is all there is to say. Chretien's epilogues are notorious for opening up as many questions as they claim to resolve, but his play merely emphasizes the common practice of storytellers and romancers who exploit the topics of prologue and epilogue to provide rhetorical frames within which to circumscribe their work, establish their credentials, and set up the contract that links author/ narrator, story, and public. Once we move to the semantic level, the question of closure is immeasurably complicated by the problem of signs and their multiplicity of meanings. Among all the different kinds of literary genres we can distinguish in the medieval context, romance seems to focus most particularly on this problem. This is not to say that romance alone sees it in the Middle Ages; investigations into sign theory from Augustine to Abelard demonstrate how "modern" medieval writers and intellectuals were in this re-

Conclusion 215 spect.17 But among vernacular works, romance offers the problem of signs and their interpretation not only a privileged place in terms of plot—consider its love of puzzles and enigmas, its frequent use of disguises, incognito, and recognition scenes, and so on—but also in terms of its own operations as discourse. Any definition of a closed or open text depends in part on what we take to be a coherent whole. As Paul Zumthor reminds us, the site of totality or unity may not be the responsibility of the medieval "texte-fragment"; wholeness may be located elsewhere. If we are dealing with a troubadour or trouvere poem, for example, it is the ensemble of the "grand chant courtois" that supplies unity, not the individual poem: the continuity of the tradition contrasts with the discontinuity of performance.18 On the other hand, medieval texts can demonstrate in various ways a desire for and even a pursuit of architectural wholeness, as well as a sense of narrative closure, if we consider romance epilogues like those of Chretien's Tvain or Renaut de Beaujeu's Le Bel Inconnu.19 As Peter Haidu argues, our twentieth-century notion of "text" has to encompass within the medieval context "entities that are both fragmentary and yet frequent and normalized, as well as other entities that look strongly finished, syntagmatically completed, and that constitute coherent wholes of signification" (p. 84). Given the nature of medieval inventio, which so frequently assembles wholes out of ready-made and recognizable parts, we should also expect to see such parts subjected to the play of repetition and variation as poets, storytellers, and writers take the already-said and make it say something different, something new with respect to the tradition invoked and modified.20 Our reception may sometimes privilege the intertextual links that lie outside the text over the integration of the part into a textual whole. This play in medieval textuality between the whole and the fragment, the continuous and the discontinuous, clearly affects in a multiplicity of ways the operation and perception of closure. As evidence of the fragmentary character of many medieval texts, what remains of Thomas's Tristan stands as a kind of "cas-limite." Perhaps accidental, such fragmentation may be no less meaningful for having its cause outside the romance. JeanCharles Payen, for example, has interpreted the fragmentation of Thomas's romance as evidence of the subversive character of the Tristan story.21 Emmanuele Baumgartner reminds us that our view of Thomas differs profoundly from that of any medieval public who knew Thomas's complete romance, especially given the extant fragments' concentration on what

216

Conclusion

Baumgartner characterizes as "le versant douloureux" of Tristan and Iseut's story.22 According to his own claims, Thomas wrote a complete and unified romance; however diverse the story matter, he gave in his version the truth as promised. Limited by our ignorance of the vast majority of Thomas's text, any modern reading of his Tristan is necessarily tentative and openended, even if we can compensate for the loss to some extent by a narrative reconstruction based on other versions and translations of his romance. On the other hand, our reconstruction of the habits of reading and rewriting that characterize medieval textuality in the twelfth century can significantly improve our efforts to analyze Thomas's fragments as signs that point to the sense of the whole. This is particularly true given the episodic character of twelfth-century romance and its tendency to use thematic analogues, narrative blocks, and so on, whose repetitions and variations supply not only the links that connect the various parts but also an implicit commentary that emerges out of the interplay to shape the direction or meaning expressed in the romance narrative.23 The patterns of doubling illustrated in the extant fragments seem to offer ample testimony to this process as it must have extended throughout Thomas's romance; such a pattern leads to and corroborates the interpretation of the ending presented above, as it insists both on the closure of the narrative syntagm and the open-endedness of Thomas's truth, his ongoing dialogue with a dual public of lovers and storytellers. In some sense, we might say that Thomas's whole romance, just like the single episode of the Folie Tristan d^Oxford^ is itself a fragment of that larger tradition. From such a perspective, each of the extant versions of the Tristan and Iseut story is a fragment with respect to the whole legend; its identity can be measured in relation to that whole. We can thus compare the intertextual dialogue between different versions and the legend of Tristan and Iseut to that described by Zumthor between individual poems and the ensemble of the "grand chant courtois." In the romance context, we can identify a number of stories that operate like magnets for the medieval imagination, attracting .writers and public not only to the tales of the fatal lovers but to those of King Arthur and the Round Table and the quest for the Grail. Each of these great legends forms a growing reservoir of tradition within which to situate individual versions as fragments: the independent romances these stories generate in the twelfth century give way to the proliferation of cycles in the thirteenth; the efforts to totalize continue, as the stories themselves continue to multiply and connect.

Conclusion

217

Each version retains its own autonomy, to be compared not only with the abstract model of the tradition as a whole but with the particular characteristics of other versions, other fragments. Just so, the Oxford Folly inscribes its allusions to Thomas's romance into the order of its renarrations (and thus invites recognition of specific intertextual play by an initiated public), while at the same time its retelling of the lovers' story includes traces of other versions: the play of intertextuality includes the general frame of the legend as a whole. But the Folie d^Oxford itself retells a complete story and ultimately can stand on its own as a whole, if greatly compacted narration of the Tristan and Iseut legend: the fragment has its own coherence. If the description here seems to oscillate between the fragment and the whole, even though we are focusing on the same textual entity, it should not surprise us: the definitions of whole and fragment necessarily point to the shifting questions of point of view and context, whether taken by the analyst or supplied by the medieval text. Nor should it surprise us that the interplay between individual texts and tradition recasts the question of closure and open-endedness. If medieval invention invites the reuse of the already-said and the return to favorite stories that continue to exert their charms over a never satisfied public, then we can appreciate to what extent the twin processes of selection and substitution operating in Marie de France's Lais reflect the basic, underlying drive for continuation and renewal that characterizes medieval textuality in general and romance writing in particular. Just as selection tends to favor closure of a given lai and substitution the open-ended series that connects one led to the next, the assemblage—like the romance writing it explicitly rejects, but implicitly claims—requires both processes to constitute and put together its parts. Once its own textual identity has been mapped out internally, the same two processes characterize its further dialogue with the tradition from which it sprang: the public may follow the intertextual play set up between specific texts or with the abstract model of the tradition. The linearity of this description may reflect more the practice of reading than the actual functioning of romance itself: closure and open-endedness necessarily go hand in hand in the intra- and intertextual dialogue; our reception of the text should be as multi-layered as the multi-dimensional process of rewriting the tradition. Marie de France's Lais dramatize the interaction between the continuities and discontinuities of tradition and individual manifestations. They offer an alternative that rivals the bele conjointure of romance: the separateness and autonomy of each tale is preserved and yet all the lais are inte-

218

Conclusion

grated into the intertextual network inscribed in the written assemblage, where they become parts of a larger whole. The twelve tales play off each other like the thematic analogues of romance, but they introduce a higher level of plurality, a greater independence for each fragment, as they reverse the relative weight given to the discontinuous over the continuous. We might compare the give-and-take between unity and diversity that characterizes the Lais with the interplay described by Thomas between the diversity of the Tristan matter and the coherence of his own version, unified by his vers (Douce, w. 835—40). If we make such distinctions in degree, we recognize at the same time that neither romances nor Marie's collection of Lais can be understood except with respect to both continuity and discontinuity, whether we focus on the intra- or intertextual dynamic of their production and reception. The use of continuations may further complicate these issues. When Chretien passes his role as author/narrator on to Godefroi de Leigni, he raises questions about the romance's identity as one text or two: continuation as double, a fragment or an integrated part of the first author's text? Double authorship argues for autonomy (perhaps more from a modern point of view than a medieval one), but the claims of Godefroi's epilogue, as well as the play among the different parts of the romance, and the impossibility of delineating exactly where the first part ends and the second begins, all argue for the continuities operating throughout the romance. Partonopeu de Blois makes a much clearer break between romance and continuation, framing each one with prologue and epilogue. Both parts claim, however, to be the work of the same author, writing for love of the same lady, even though modern editors have failed to agree on whether or not we should believe the claim or count it among the many fictions of a romance particularly inclined to mix truth and fable, as it scrambles and rethinks the boundaries between them. In this case, we can identify the narrative completion of the first part, satisfactorily achieved in the marriage of the long-separated lovers, while the continuation appears more as a series of fragments—battles recounted in epic style, Anselot's unhappy love story, Margaris's writing efforts—strung together loosely by the narrative frame of the Sultan's invasion, but motivated primarily by the "narrative" of the author/narrator's own love story. As we consider the shifting textual boundaries in each of these examples we discover again that the contrast between continuities and discontinuities is equally located within the romance text itself. The kind of mouvance we tend to associate with oral traditions is, in

Conclusion

219

the manuscript culture of the Middle Ages, just as important within the context of writing and just as unsettling for our modern sense of what makes a text closed or open. At first glance, we might assume that writing favors closure, orality open-endedness, and thus associate written romance with the former and oral tales with the latter. But then what happens when the oral tale is transposed into writing, one by itself or a dozen combined? What happens when different romances rewrite the story of Tristan and Iseut or the tales of King Arthur? Each romance fiction operates as a new experiment on the already-said, crossing back and forth between oral and written traditions: one gives a definitive form and resolution by foregrounding the voice of orality (the Oxford Folly); another keeps the process of interpretation open despite the definitive closure of the narrative syntagm (Thomas's Tristan); one enlarges the romance through a continuation authorized by the narrator's lyric engagement with his beloved (Partonopeu), another disrupts the continuity of authorship by announcing the intervention of a second author who emerges only in the retroactive shadow cast by his epilogue (Charrete). The question of closure is continually reopened, as it operates on every level of the romance product and reappears on every level of romance production (composition, transmission, and reception). If we focus on the semantic level, we might reverse the earlier statement and associate romance with open-endedness (semiosis without end) and contes or lai with closure, resolution, and achievement of meaning. But here again exceptions and variations abound in the Protean transformations of romance fictions. The Folie confirms such a generic model just as strongly as the indeterminate meanings of Marie's lais dispute it. Each of the romances studied here complicates the interplay between closure and open-endedness by locating aspects of both either on different levels or in combination with different elements operating at a given level. On the other hand, the romance genre generally highlights through its variations the problematic nature of signs. As such, it tends to emphasize and play out the open-ended process of interpretation required, even if individual characters and the romances themselves find appropriate stopping places, or better resting places, before continuing the pursuit of truth and meaning inscribed in all the multiple guises of fiction. The issues of closure and intertextuality necessarily overlap with the question of meaning and thus lead us inevitably to focus on the role played by signs in determining how and when a given romance may be semantically closed or open or a combination of both.

22O

Conclusion

In the romance world signs are problematic: ambiguous, contradictory, misleading, and deceptive. Subject to changes as multiple as the multiplicity of viewpoints from which they are perceived, signs can be manipulated, voluntarily and involuntarily. And yet they can lead to some kinds of knowledge, some kinds of truth; their interpretation can yield meanings that may or may not be validated by further interpretations inside and outside the romance world. In such a context, knowledge is no longer securely located in and retrievable from an objective reality, but emerges from the interplay between observer and observed; subjectivity and interiority function as part of the interpretation of reality. Signs are uncertain not only from the point of view of the observer but also in their relation to referents. This is the process we see operating in the Charrete^ when Lancelot's actions at the tournament are subjected to the instructions "au noauz" and uau mialz," creating for the characters first contradictory and then complementery resonances with the herald's cry, "Or est venuz qui Paunera!" The disjunctions between words and performance, between words and other words, even between words as sound and words as verbal signifier (as in the Oxford Folly), each of these gaps problematizes the link between signs and their meanings, undermines the oppositions that organize our experience of reality—truth/lies, appearance/reality, fiction/truth—and reclassifies them as non-disjunctive oppositions, not only in the romance world but in that of the real author and public. As we saw earlier, the problem of signs furnishes one of the major bridges that pass between our world and the world of romance fiction.24 Signs in romance are as varied as those encountered in human experience: they are actions, words, visual images and sounds, letters and oral reports, bodily reactions and objects exchanged—but, of course, all of them come to us as verbal signs, the direct and indirect discourses of the characters and especially the discourse of the author/narrator who filters and frames all that happens in romance through the act of writing. If romance tends to make us focus on the problem of signs and the acts of interpretation they generate, it is not surprising that scenes of recognition are a favorite topos of romance fictions and figure prominently in this study, inasmuch as they dramatize the process of cognition itself as it functions in the medieval context. Nor is it surprising that words—speech, renarration, discourse—play a particularly prominent role in the return to knowledge held earlier and now regained and enriched after a period of loss. The recognition scenes in the Folie ^Oxford and Guipfemar were particularly helpful in their detailed representation of the process, as it calls

Conclusion

221

upon sense perception, reason, memory, and belief, images and intellect, working together in hierarchical order to bring about the recognition of the separated lovers. Discourse as renarration of the past plays a privileged role in both stories, although its success as the ultimate stage in Guigemar's recognition of the lady contrasts with its failure in the Folie until Tristan returns to his true voice and authenticates the source of his renar rations. The prominence of language in the process of interpretation, as in the model of recognition, suggests that in the romance world it makes no difference whether we consider the relation between a verbal sign and an event or that between two signs already represented as discourse—all is mediated for us by language. This is a view shared by twelfth-century thinkers like Abelard, whose theory of language is summarized by Jean Jolivet in the following passage: [Le langage est] un interprete infidele des choses, mais un bon temoin des activites de 1'esprit; . . . son rapport mediat au reel, et son rapport immediat a 1'esprit, nous interdisent de partir des mots et des phrases pour comprendre le monde. . . . la formation des mots doit etre comprise comme le resultat d'une certaine activite mentale plutot que comme la traduction parfaitement adequate de la nature des choses; . . . [le langage] est le fruit d'une institution, non un fait naturel; aussi reflete-t-il d'abord des structures mentales, et beaucoup moins nettement des structures reelles. Du sein du langage, nous ne pouvons pas deboucher directement sur le monde; il faut d'abord traverser la sphere de 1'esprit.25

Not only does language interpose itself between us and the world, it functions more as screen than transparent bridge, more as an indication of what and how we think than as a direct or faithful representation of the world. Such a formulation recalls the way knowledge and perception operate in the romance context, where the interaction of observer and observed together produces a representation, which is already an interpretation, of reality.26 With its emphasis on the verbal, the use of signs and their problematic interpretations, romance participates in philosophical and theological debates on language and signs that occupy medieval clerical culture. Romance's participation may be indirect and much less abstract, given the vernacular context of storytelling, but it is no less significant for the development of ideas and models for medieval society.27 The different intentions that direct the exploration of signs in clerical and vernacular writings may be characterized grosso modo in spatial terms: the vertical axis of theological inquiry has been given a decidedly horizon-

222

Conclusion

tal thrust in romance's attention to human concerns and problems, human truths and meanings.28 As a language user, human beings are in God's image; we share the power to create through language. Indeed, the parallel between the Creator as artist and the artist as creator is a favorite topos of the twelfth century.29 But that parallel ultimately reveals the essential difference between the unlimited power of the Logos and the limited power of human language. In human hands the powerful tool of language may be as dangerous as it is exhilarating and must be used with special care.30 If the power to deceive characterizes both signs in general and language in particular, in romance such deceptions are not necessarily judged to be evil. Meleagant's lying words are thoroughly condemned, but as a kind of personification of evil he may be an unusually one-dimensional character in romance. Heroes and heroines favorably presented in romance fictions are generally treated with more ambivalence, when it comes to manipulating signs and using the resources of engirt. If we take Partonopeu's genealogy as a model, human invention plays a necessary role in the development of history, which may even require a bit of lying and deception here and there.31 Nor are such deceptions always voluntary; they may be the result of our human limitations. Here we return again to the problem of multiple points of view and the unavoidable interaction between observer and observed. If the beauty of the beloved is a function of the loving eye, then the truth of who is the most beautiful is caught in the circle of subjective judgment. From God's point of view, there may be only one who is the most beautiful in truth and by reason, but if we remain within the circumscribed horizontal view of human lovers, each will proclaim his or her beloved the most beautiful—hence the proliferation in romance of Sparrowhawk Contests and tournaments to resolve the choice of one above all. We should note further that you cannot arrive in romance at a single or whole truth simply by adding up all the partial views, the multiple points of view included. The competing signs of romance cannot be totalized by mere addition. How can we put together all the different and contradictory reputations and accomplishments of Lancelot? Or, conversely, how could we arrive at a whole and truthful understanding of Lancelot's heroism, if we try to suppress or leave out the plurality of views offered:1 It may very well be that in the human context, if all must past through the filter of language, then some aspects of reality may refuse expression except through contradictions and ambiguities, or they may simply remain outside our power to express them. This notion appears

Conclusion

223

directly in the inexpressibility topos, used significantly by narrators and characters in the Charrete to describe Lancelot's feats, impossible but nevertheless achieved.32 It appears as well in narrators' comments on the indescribability of love's joy, the surplus that escapes language's grasp, although words can sometimes point to what they cannot say.33 It may be that romance is one of the places in medieval culture where the limits of human experience are explored, where the limits of language strain to see and express what lies beyond them. Perhaps this is the major function of twelfth-century romance as fiction. The "make believe" of romance in some sense enlarges the "whole" of reality, because it does not simply content itself with reproducing history, the stories of the past, but transforms them with its playful inventions. Of course, it does not and cannot act directly on reality, just as language in Abelard's view cannot have an immediate link to reality, but only to the activities of the human mind. Romance does act directly and indirectly on the human side of the equation: its contribution to the extension of reality is all in the mind, whether that of the author or that of the reader/listener. Romance's wild inventions, its marvelous deceptions, its absurdities and silliness, its play with multiplying different views of the same to produce something different, all contribute to expanding the human notion of what it is possible to conceive in language (or even almost conceive), and then perhaps to perceive or make happen in reality outside the human mind. Reality, in this view, is not a fixed quantity: human interaction with the outside world is liable to make changes on both sides of the dynamic. In this sense, we can understand the pleasure romance takes in craftiness and trickery, its own as well as the characters', the delight and approval with which it views certain deceptions—not all certainly: it presents the dangers of lying, the treachery of deceit, but it cannot help but recognize that its own feats are those of lying fiction, spinning deceitful tales that delight and enthrall. Romance's use of "aesthetic distance" serves as a warning to its public not to forget such deceptions, but to use them as lessons. As the Prologue to Partonofeu argues, the games of fiction are more profitable for those who play them than other pastimes of courtly society. While the time spent playing checkers or dice is lost, the pleasure the author/narrator takes in writing his romance will be shared by all those who learn from its examples how to discern good from evil and exercise their ability to choose between them (w. 77-93). Twelfth-century romance vociferously denies the idea that it offers lying fictions, which is not to say that it denies its fictionality—on the contrary.34 What twelfth-

224

Conclusion

century romance denies, implicitly if not explicitly, is that we can assume fiction lies or that there is a necessary correlation between lying and fiction. In the romance subversion of oppositions like truth and lies, or rather in the reformulation of them as non-disjunctive oppositions, romance preserves such distinctions, but redefines them to reflect and represent (and perhaps ultimately to reshape) the complexities of human experience that often escape the reduction and schematization of human reason. Romance lies in order to tell and even discover truths that human history creates in the transformations of language. The logic of romance thus understood may explain why truth as a category appears so frequently within its fictional domain. Like the category of meaning, to which it is repeatedly associated in romance fictions, truth is as indispensable as it is problematized in romance. On the one hand, we might say that truth claims are there as a kind of window-dressing, a claim to authority or credibility that, like the captatio benevolentiae among exordial topoi, invites the public to treat its fictions with some seriousness, even as they are enjoyed with pleasure. More profoundly, the notion of truth and the interpretation of meaning are both subjected to the same kind of "mobilization of categories" that we followed in the crisscrossing patterns of Partonopetfs gender and genre transformations. In romance, truth and meaning are human constructs, constructed by and through human experience. The constructions of romance are themselves both the result of such experience and the impetus to continue exploring and reshaping it. Once again, as the Prologue to Partonopeu reminds us, romance claims an ethical dimension to its esthetic pleasures, an epistemological adventure accompanies the rough and tumble of its chivalric adventures. Or, in the terms of romance writing, clergie and chevcderie serve each other, go hand in hand in the courtly society that produced and consumed romances. If the title of this study speaks of shape in romance, it is because that word catches both the formal and semantic dimensions of romance as they are tied together. Vernacular literature plays an important role in twelfthcentury humanism; romance in particular provides an outlet for transposing the riches of Antiquity, the exotica of the East, and the treasures of Celtic myth, for a vernacular public eager to expand its horizons. The incredible gift for experimentation we see in the formal shapes invented by romance fictions during this period is matched by their free-wheeling experimentation with the shapes of a secular society trying to translate conflicting ideals into practice. The formal reshaping of romance is indeed

Conclusion

225

another face of the reshaping of ideas, the experiments with meaning and truth, the questions raised if not always answered by romance fictions. In some sense, we might say that the shape of romance is as open-ended as the series of questions it opens for its contemporary public, as well as for a modern public still avid to explore the nature of the human world, the possibilities and accomplishments of the human imagination. Romance itself suggests that stories, ostensibly taken from the past but significantly transformed by present fiction, offer a privileged view of human reality, not simply as a reflection of what is outside us but as a tool that shapes and recreates what we see. Our twentieth-century present can still benefit by such an invigorating if fictional interrogation of the past.

This page intentionally left blank

Notes

Introduction 1. Cf. Jean-Charles Payen, "Le lai narratif," Typologie des sources du may en age occidental Fasc. 13 (Turnhout: Brepols, 1975), pp. 52-55; and Hans Robert Jauss, "Theory of Genres and Medieval Literature," in Toward an Aesthetic of Reception, tr. Timothy Bahti, Theory and Literature, vol. 2 (Minneapolis: University of Minnesota Press, 1982), p. 86: "To be sure, the romance follows the fictional principle of the fairy tale, that in the adventure no occurrence may be like reality; and yet the courtly narrator presents the claim of discovering a sensus moralis in the res ficta" 2. On the twelfth-century Renaissance, see Charles H. Raskins, The Renaissance of the Twelfth Century (Cambridge, Mass.: Harvard University Press, 1927, 1955); Robert L. Benson and Giles Constable with Carol D. Lanham, ed. Renaissance and Renewal in the Twelfth Century (Cambridge, Mass.: Harvard University Press, 1982); Martin Stevens, "The Performing Self in Twelfth-Century Culture," Viator 9 (1978): 194-212; Marie-Dominique Chenu, La theologie au douzieme siecle, Etudes de Philosophic Medievale 45 (Paris: Vrin, 1957) or selections in English in Nature, Man, and Society in the Twelfth Century: Essays on New Theological Perspectives in the Latin West, ed. and tr. Jerome Taylor and Lester K. Little (Chicago: University of Chicago Press, 1968); Stephen G. Nichols, Jr., Romanesque Signs: Early Medieval Narrative and Iconography (New Haven, Conn.: Yale University Press, 1983), pp. 12-13; R. W. Southern, Medieval Humanism and Other Studies (New York: Harper & Row, 1970), p. 31 (quoted by Stevens, "Performing Self," p. 209); Beryl Smalley, The Study of the Bible in the Middle Ages (Oxford: Blackwell, 1952). 3. Cf. Douglas Kelly, The An of Medieval French Romance (Madison: University of Wisconsin Press, 1992), p. n: "The genre of romance, as a heuristic referent, is not monolithic, which accounts for the originality and dynamism of medieval romance as it emerged." 4. Julia Kristeva uses the terms disjunctive and non-disjunctive oppositions in her analysis of a semiotic shift from symbol to sign: Semeiotike: recherches pour une semanalyse (extraits), Collection Points (Paris: Seuil, 1969), pp. 55-62. Cf. also Jacques Derrida's logic of the "supplement" (De lagrammatologie [Paris: Minuit, 1967], pp. 203-34). 5. Cf. Kelly, Art of Medieval French Romance, p. n. 6. Paul Zumthor, Essai de poetique medievale (Paris: Seuil, 1972), reviewed by Peter Haidu, "Making It (New) in the Middle Ages: Towards a Problematics

228

Notes to Pages 3—4

of Alterity," Diacritics 4 (1974): 1-12; and Hans Robert Jauss, "The Alterity and Modernity of Medieval Literature," New Literary History 10 (1979): 181-230, and "Literary History as a Challenge to Literary Theory" and "Theory of Genres," in Toward an Aesthetic of Reception, pp. 3—45, 76—109. 7. Anyone who has seen George Lucas's "Star Wars" trilogy, its popularity, and the order in which the three films were produced undoubtedly has increased insight into the appeal and growth of the Prose Lancelot and the Vulgate Cycle, as these stories multiplied from the middle out to the end and back to the beginning. 8. We can see, for example, both the usefulness and the limitations of Gerard Genette's theoretical model for continuations, as described in Palimpsestes, when that model is applied to medieval texts. See Bruckner, "Intertextuality," in The Legacy of Chretien de Troyes, ed. Norris J. Lacy, Douglas Kelly, and Keith Busby, 2 vols. (Amsterdam: Rodopi NV, 1987, 1988), vol. i, pp. 223-24, 245-65; and Kevin Brownlee, "Transformations of the Charrete: Godefroi de Leigni Rewrites Chretien de Troyes," Stanford French Review 14 (1990): 162-64,173. 9. D. W. Robertson, Jr., A Preface to Chaucer: Studies in Medieval Perspectives (Princeton, N.J.: Princeton University Press, 1962), pp. 89-90; Franz Bauml, "Varieties and Consequences of Medieval Literacy and Illiteracy," Speculum 55 (1980): 255-57. On the more positive side, Maurice Keen, Chivalry (New Haven, Conn, and London: Yale University Press, 1984), Chapter II, includes a number of references to the way books of chivalry and family histories cite with approbation the models offered by Roland, Arthur, Tristan, Palamedes, etc. 10. Cf. Paul Zumthor, Parler du moyen age (Paris: Minuit, 1980); Lionel Gossman, Medievalism and the World of the Enlightenment: The World and Work of Lacurne de Sainte-Palaye (Baltimore: Johns Hopkins University Press, 1968); and R. Howard Bloch, "Signs and the Times," University Publishing 9 (1980): 20. n. See, for example, the many contributions of Georges Duby and the Annales school in France. Timothy J. Reiss, The Discourse of Modernism (Ithaca, N. Y.: Cornell University Press, 1982), includes a chapter on the dominant discourse of the Middle Ages, "Questions of Medieval Discursive Practice," pp. 55-107, in which he recognizes the possibility for alternate discourses, but downplays them in the portrait of a dominant against which he can then define modernity. 12. Keen, Chivalry, p. 3. 13. Cf. Eugene Vance, "Signs of the City: Medieval Poetry as Detour," New Literary History 4 (1973): 557-74, especially pp. 571-73, where he borrows Roman Jakobson's distinction between code and context to argue for reading the interaction between twelfth-century romance and a changing feudal society, not in terms of context but on the level of code. 14. Cf. Peter Haidu, "The Hermit's Pottage: Deconstruction and History in Yvain" in The Sower and His Seed: Essays on Chretien de Troyes, ed. Rupert T. Pickens (Lexington, Ky.: French Forum Pub., 1983), pp. 127-45. For an exploration of the links between vernacular literature and philosophical inquiry in the Middle Ages, see, for example, Tony Hunt, "Abelardian Ethics and BerouPs Tristan" Romania 98 (1977): 501-40, and "Aristotle, Dialectic, and Courtly Literature," Viator 10 (1979): 95-129. 15. Florence McCullough uses this term to signal the juxtaposition of long

Notes to Pages 4-13 229 and short lais in Marie de France's collection: "Length, Recitation, and Meaning of the Lais of Marie de France," Kentucky Romance Quarterly 25 (1978): 257-68. 16. Andre Jolles, Les formes simples, tr. Antoine Marie Buguet (Paris: Seuil, 1972), pp. 137-5717. Marie-Louise Oilier, "Le discours en abyme ou la narration equivoque," Medioevo Romanzo i (1974): 351-64. Renate Blumenfeld-Kosinski analyses the frequent rhyme couple songelmensonge, in "Remarques sur songelmensonge" Romania IO2 (1980): 385-90.

18. Cf. Mary Carruthers's discussion of the heart as the seat of memory, which in the medieval context is the faculty that gives rise to judgment: The Book of Memory: A Study of Memory in Medieval Culture (Cambridge: Cambridge University Press, 1990), pp. 48-49, 68-69. 19. Cf. Gail McMurray Gibson, The Theater ofDevotion: East Anglian Drama and Society in the Late Middle Ages (Chicago and London: University of Chicago Press, 1989), especially her eloquent argument in the introduction for not treating the medievals as children (pp. 2-6). Although Gibson is addressing the particular biases against the fifteenth century, her defense applies with equal astuteness to earlier periods and their misrepresentations in modern views. 20. On the problems of dating Thomas and Chretien in relation to each other, see Bartina Wind's Introduction, pp. 14-17, in Thomas, Les fragments du Roman de Tristan: poeme duXIP siecle^ ed. Bartina Wind, Textes Litteraires Frangais 92 (Geneva: Droz, 1960). The debates on dating Marie's works are tied to the inconclusive research on her identity. See, for example, Joseph Bedier, "Les Lais de Marie de France," Revue des Deux Mondes 107 (1891): 835-63; John C. Fox, "Marie de France," English Historical Review 25 (1910): 303-6; Urban Tigner Holmes, Jr., A History of Old French Literature from the Origins to 1300 (New York: Crofts, 1937; rev. ed., New York: Russell & Russell, 1962).

Chapter i 1. Joseph Bedier lists six versions in his edition, Les deux poemes de la Folie Tristan (Paris: Firmin-Didot, 1907), pp. ii—iii: two independent episodes in French verse (Oxford and Berne), an episode included in the French prose Tristan of the thirteenth century, an episode in Eilhart's Tristan (end of the twelfth century), and episodes in two continuations of Gottfried (Ulrich of Tiirheim's, c. 1240, and Heinrich of Freyberg's, c. 1290). 2. Ernest Hoepffner, ed., La Folie Tristan d'Oxford (Paris: Belles Lettres, 1938), p. 5. All references and quotations will refer to this edition, unless otherwise specified. My translations represent an effort to help the reader follow the language of the original text. 3. This aspect of the Folie d'Oxford places it squarely in the Thomas tradition, though as we shall see later there are equally significant differences between the two poets. 4. See, for example, w. 366-73 or w. 464-71 for Iseut's aggressive ex-

230

Notes to Pages 13-22

pressions of distress. The timidity and fear that in the Oxford version characterize most of the Queen's private behavior with the fool appears in the Folie de Berne only when Husdent joyously recognizes Tristan (w. 517—19). 5. See w. 285, 295, 3", 381, 389, 414, 499, 500, 503, 533-

6. Cf. J. Duncan Robertson, "Literary Tradition and Poetic Realization in the FoliesTristan" Ph.D. diss., Princeton University, 1972, p. 75. 7. R. Howard Bloch, "Tristan, the Myth of the State and the Language of the Self," Tale French Studies 51 (1974): 61-81, esp. pp. 77-81; quotations on p. 81. 8. In Thomas the ordeal (Chapter XXIV in Bedier's reconstruction, Roman de Tristan) immediately follows the flour episode. See Hoepffner's introduction (p. 7) for a list of these changes in order between the two versions. My analysis of Thomas here and elsewhere will be based only on what is available in the extant fragments. 9. Beroul, Le Roman de Tristan: poeme du XIIe siecle, ed. Ernest Muret (4th ed. L. M. Defourques), Classiques Francois du Moyen Age 12 (Paris: Champion, 1970), w. 3244-4247, for the same exploitation of ambiguous words, with a slightly different action to support them. The Folie d'Oxford does not mention the ordeal that follows in Thomas's version of this episode, but relies solely on the verbal aspect of contemporary legal procedures. See Ernest C. York, "Isolt's Trial in Beroul and La Folie Tristan d'Oxford" Medievalia et Humanistica 6 (1975): 157-61. The evocation of the trial by oath (compurgation) accords well with the importance of the spoken voice in the Oxford version of Tristan's disguise. 10. Cf. Denyse Delcourt's discussion of the Aristotelian distinction between accidents and substance, explored in the writings of the School of Chartres and in romance of the twelfth century: Uethique du changement dans le roman du XIIe siecle (Geneva: Droz, 1990). n. Cf. Brigitte Gazelles, "Alexis et Tristan: les effets de 1'enlaidissement," Stanford French Review 5 (1981): 85—95. 12. Augustine had already demonstrated long before the twelfth century that reasoning correctly through a process of deduction is not enough; you also have to verify the correctness of your premises. See, for example, De Doctrina Christiana, tr. D. W. Robertson, Jr. (New York: Library of Liberal Arts, 1958), 2.xxxi.48-xxxvii.55 (especially xxxi.52). 13. On the importance of memory in medieval culture, see Mary Carruthers, The Book of Memory: A Study of Memory in Medieval Culture (Cambridge: Cambridge University Press, 1990). 14. The verb parler in Old French enjoys the same double meaning as the English "converse." For a discussion of this Old French usage, see Mary Blakely Speer's review of T. B. W. Reid, The Tristan of Beroul: A Textual Commentary, in Romance Philology 30 (May 1977): 639. 15. Faussete and dissimulation are the two words Ernest Hoepffner gives in his glossary to define feintise (p. 153). 16. Cf. Jacqueline T. Schaefer, "Tristan's Folly: Feigned or Real?" Tristania 3 (Nov. 1977): 3-I517. This adjective is absent from the extant Thomas (Hoepffner, ed., La Folie Tristan d'Oxford, p. 35), a significant omission that contributes to the different

Notes to Pages 22-26

231

thematic forms of these two works (see discussion below). Interestingly, this is the adjective that the narrator uses to describe Lancelot's heart in the Charrete, ed. Mario Roques, Classiques Francois du Moyen Age 86 (Paris: Champion, 1967), v. 4667. 18. Cf. Walter Ong's discussion of recognition in Interfaces of the Word: Studies in the Evolution of Consciousness and Culture (Ithaca, N.Y.: Cornell University Press, 1977), pp. 247-48. 19. Walter J. Ong, S.J., The Presence of the Word (New Haven, Conn.: Yale University Press, 1967), pp. 130, 113. My debt to Ong's work will be apparent throughout the following analysis. 20. Cf. Ong's discussion of truth and sound (Presence, pp. 148-61) and the quotation from Ambrose's Commentary on St. Luke (IV.s): "'Everything we believe, we believe either through sight or through hearing . . . sight is often deceived, hearing serves as a guarantee'" (pp. 52-53). Derrida's critique of logocentrism in the Western cultural and philosophical tradition comes inevitably to mind in this context, but it is useful to point out here the particular feature that distinguishes the Oxford Folly's version of logocentrism: it is not just speech, which is privileged in this text, but sound itself, which operates almost independently of the verbal signifier. 21. Tristan uses the same formula ("e go a dreit," v. 435) to verify and approve Iseut's past recognition of him as the slayer of the Morholt. 22. Cf. Ong, Interfaces (p. 242), where he describes all literary enjoyment as retrospective, rather than anticipatory or even participatory, and more generally, Roland Barthes's exploration ofjouissance in Leplaisir du texte (Paris: Seuil, 1973). 23. Guillaume in the Charroi de Nimes and Floire in Floire et Blancheflor both disguise themselves as merchants. Chretien's heroes use incognito either for tactical advantage (e.g., Erec and Cliges) or for more strategic, long term goals (e.g., Lancelot and Yvain). Any reader of medieval French literature can supply abundant examples of visual disguises. Even stories that clearly imitate the Tristan model—\ikcAucassin etNicolette^ for example, when Nicolette disguises herself as a jongleur and returns to Aucassin's court—do not necessarily borrow the oral/aural aspect of the Folie. 24. Le Roman de Renart, ed. Mario Roques, 6 vols., Classiques Francois du Moyen Age 78 (Paris: Champion, 1971), I, v. 2392. The entire episode "Renart teinturier" (w. 2321-3256) shares a number of motifs with the folly episode, but of course uses them for its own variations and ends. On the shared characteristics of Renart and Tristan, see Nancy Freeman Regalado, "Tristan and Renart: Two Tricksters," UEsprit Createur 16 (1976): 30-38. 25. Mario Roques discusses in his introduction the systematic representation of Renart's English accent (pp. xx-xxi). 26. That perfect refuge the fool imagines for himself and the Queen—a crystal chamber in the sky where all is brightness, clarity, stillness, and transparency—does it not summarize just those qualities singularly lacking in the reality of their occasional refuges, whether in cave or garden? We might also associate the tiger's milk that nourished the fool with the equally unnatural drink that nourished Tristan and Iseut's love, especially since the love potion (baivre] is prominently

232

Notes to Pages 26-29

recalled in the renarrations, first with Iseut in front of the court (w. 461-76), later with Brengain in private (w. 639-59). 27. See w. 322, 387, 409, 457-60. 28. See w. 681-84, 709-12, 835-48, 919-20, 939-40, 961-68. 29. This kind of reaction probably underlies some of the charges leveled against the Folie d'Oxford, either for the invraisemblance of its orderly renarrations—see Bedier's edition of the two Folies (pp. vi—vii), Robert Guiette's review of Hoepffner's Oxford edition (Revue Beige de Philologie et d'Histoire 18 [1939]: 997)—or for the cruelty of Tristan's treatment of the Queen—see Renee L. Curtis, "The Humble and Cruel Tristan: A New Look at the Two Poems of the Folie Tristan" Tristania 2 (1976): 3—11; Jean-Charles Payen, ed. and tr., Tristan et Iseut (Paris: Gamier, 1974), p. xvii. 30. I have suggested above the ambiguous sincerity of these complaints. They may, in fact, be narratively equivalent to the foolish inventions of the first part, this time accenting the real folie of Tristan's love-suffer ing (or the feigned echo of those true feelings). 31. Cf. Edmond Faral's discussion of the "dramatic monologue" in Mimes fran$ais du XIIP siecle: contribution a I'bistoire du theatre comique au moyen age (Paris: Champion, 1910), pp. viii-xiv. He also refers us to Geoffroi de Vinsauf's Poetria Nova for contemporary advice on oral delivery, w. 2031-65 in Faral's edition, Les arts poetiques du XIIf et du XIIP siecle (Paris: Champion, 1971), pp. 259-60. That discussion is particularly interesting in relation to the Folie d'Oxford, since Geoffroi deals specifically with the necessary harmony between voice and subject, with the importance of recognizable signs, and with the "death and life ... powers of the tongue" (PoetriaNova of Geoffrey ofVinsauf, tr. Margaret F. Nims (Toronto: Pontifical Institute of Mediaeval Studies, 1967), pp. 90-91, v. 2060). 32. Cf. Peter Haidu, "Text, Pretextuality, and Myth in the Folie Tristan d'Oxford" Modern Language Notes 88 (1973): 713-14. 33. See Sigmund Eisner's The Tristan Legend: A Study in Sources (Evanston, 111.: Northwestern University Press, 1969). 34. The concatenation of mises en abyme with its sense of endless returns to the past recalls on another level the inexhaustibility of sign interpretation described earlier. The tendency to use every opportunity to retell the lovers' story is characteristic of the Tristan material as a whole and can be seen elsewhere in Marie de France's Chievrefoil or in Beroul's version of their separation after the Morrois forest episode. Cf. Regalado, "Tristan," pp. 37-38. 35. On the connection between truth and oral storytelling, as well as the interplay between anonymous and named storytellers/authors, cf. the discussion in the chapter on Marie de France's Lais. 36. Cf. Maurice Delbouille, "Le fragment de Cambridge et la genese des Tolies Tristan,'" Travaux de linguistique et de litterature 16 (1978): 119-22. 37. See Beroul, Le Roman de Tristan, w. 1265—70; Thomas, Roman de Tristan, Douce, w. 835-84. 38. Deuxpoemes, p. 2, and his edition of Thomas, Roman de Tristan, II, p. 89. While HoepfFner notes the change in order between the two, his study of the language and versification agrees with Bedier's (Deuxpoemes, p. 13) and leads him

Notes to Pages 29-32 233 to emphasize the similarities between the Folie d'Oxford and Thomas over any difference (see his edition, pp. 2—3, 7—12, 21—39). 39. But see E. D. Hirsch's The Validity in Interpretation (New Haven, Conn.: Yale University Press, 1967) for a defense of reasonable inquiry in that domain. 40. See Ong, Presence^ p. 60 et passim. In addition, see Franz Bauml, "Varieties and Consequences of Medieval Literacy and Illiteracy," Speculum 55 (1980): 237-65; Joseph J. Duggan, The Song of Roland: Formulaic Style and Poetic Craft (Berkeley: University of California Press, 1973); Pierre Gallais, "Recherches sur les mentalites des romanciers fran^ais du moyen age," Cahiers de Civilisation Medievale 7 (1964): 479—93, and 13 (1970): 333—47; Brian Stock, Implications of Literacy: Written Language and Models of Interpretation in the Eleventh and Twelfth Centuries (Princeton, N.J.: Princeton University Press, 1983); Eugene Vance, "Roland and the Poetics of Memory," in Textual Strategies: Perspectives in Post-Structuralist Criticism^ ed. J. V. Harari (Ithaca, N.Y.: Cornell University Press, 1979), pp. 374-403; Paul Zumthor, Lapoesie et la voix dans la civilisation medievale (Paris: Presses Universitaires de France, 1984) • 41. Ong, Presence, pp. 114—15. 42. Vance, "Roland," p. 403. 43. Cf. Peter Haidu's discussion of the ending in "Text, Pretextuality and Myth," p. 71744. This relationship between tradition and texts, as well as between individual texts, undoubtedly plays a role in the difficulty of assigning texts to particular sources. See W. Lutoslawski, "Les Folies de Tristan," Romania 15 (1886): 511-33; Bedier, Deux poemes, p. vi; Maurice Delbouille, "Le fragment," pp. 117—29; Hoepffner's edition of La Folie Tristan de Berne (Paris: Belles Lettres, 1949), pp. 6-12; Robertson, "Literary Tradition," pp. 10-16. On multiple intertextual recalls in Tristan stories, see Curtis, "Humble and Cruel Tristan," pp. 3-11; Ruth Dean and Elspeth Kennedy, "Un fragment anglo-normand de la Folie Tristan de Berne," Moyen Age 79 (1973): 66 n. 5 (cf. p. 72, n. 57 and its reference to Thomas). 45. La Folie Tristan ^Oxford, p. 30. 46. The underlying optimism of Marie's Chievrefoil can include the lovers' death by stressing their final, inseparable union (the metaphor evoked in the title and explained in Tristan's message) and by focusing our attention on what remains after death, the lai composed by Tristan and recorded in writing by Marie. Cf. Duncan Robertson's discussion of the Folly episodes' timelessness ("Literary Tradition," pp. 22-23). 47. The changeability of human love does surface briefly in Folie d'Oxford, when the fool suggests that Mark is bored and offers to exchange his sister for Iseut, w. 289-94. 48. The striking difference between these two texts motivates a recurring aspect of criticism on the Folies: the defense of one author's style against the other. Bedier (Deux poemes, pp. iv-v) and Hoepffner (La Folie Tristan de Berne, pp. 10ii) prefer the Oxford version. M. J. Horrent, "La composition de la Folie Tristan de Berne" Revue Beige de Philohgie et d'Histoire 25 (1946-47): 21-38, and Krystyna Kasprzyk, "Fonction et technique du souvenir dans la Folk Tristan (Berne 354)," Etudes Felix Lecoy (Paris: Champion, 1973), PP- 261-70, defend the Berne version.

234

Notes to Pages 32-37

49. Cf. Delbouille, "Le fragment," p. 118, and Horrent, "La composition," P-37. 50. "F0, malgre les qualites de son style, n'est qu'un aide-memoire, qu'une 'table des matieres'" ("Le fragment," p. 127). Delbouille agrees in this respect with Horrent, "La composition," pp. 26, 32. 51. Such is BerouPs version of that farewell (w. 2700-2702). The Folk ^Oxford makes a significant change here: it attaches the dog and the ring (w. 943-56) to separate incidents in the past. 52. Cf. Stephen G. Nichols, Jr., "Realistic Perception in Twelfth-Century French Literature," L'Esprit Createur 5 (1965): 215-16. Kasprzyk sees a similar concern in the Folie de Berne with the materiality of human existence ("Fonction," pp. 268-69). It is useful to recall here the medieval use of cors to mean one's self, and the intense association between physical presence and the concept of personhood implied in such a usage. See Jean Stefanini, La voix pronominale en ancien et en moyen frangais. Publication des Annales de la Faculte des Lettres, N.S. 31 (Aixen Provence: Ophyrs, 1962), pp. 331-51. See also Sharon A. Farmer, "Softening the Hearts of Men: Women, Embodiment, and Persuasion in the Thirteenth Century," in Embodied Love: Sensuality and Relationship as Feminist Values, ed. Paula M. Cooey, Sharon A. Farmer, and Mary Ellen Ross (San Francisco: Harper & Row, 1983), pp. H5-3353. Cf. Douglas Kelly on the medieval concept of gradualism in "Courtly Love in Perspective: The Hierarchy of Love in Andreas Capellanus," Traditio 24 (1960): 119-4754. In The Mirror of Language: A Study in the Medieval Theory of Knowledge (New Haven, Conn.: Yale University Press, 1968), Marcia Colish offers us a useful overview of medieval epistemology, essentially based on St. Augustine's theory of signs. My brief description of that theory is gratefully indebted to her work. 55. I would like to thank Gail Berkeley Sherman, Reed College, for this insight.

Chapter 2 i. Those who identify Thomas with a glorification offin'amor include Jean Frappier, "Structure et sens du Tristan: version commune, version courtoise," Cahiers de Civilisation Medievale 6 (1963): 255-80, 441-54; Emmanuele Baumgartner and R. L. Wagner, "cAs enveisiez e as purvers': commentaire sur les vers 3125—3129 du Roman de Tristan de Thomas," Romania 88 (1967): 535; and Susan Dannen baum, "Doubling and Fine Amor in Thomas' Tristan" Tristania 5 (1979): 7 and n. Those who stress the negative in Thomas's point of view include Pierre Jonin, Les personnages feminins dans les romans frangais de Tristan au XIIe siecle (Aix-enProvence: Ophrys, 1958); Omer Jodogne, "Comment Thomas d'Angleterre a compris 1'amour de Tristan et d'Iseut," Lettres Romanes 19 (1965): 119; Evelyn Birge Vitz, "Desire and Causality in Medieval Narrative," Romanic Review 71 (1980): 229-33. A few readers of Thomas have integrated a dual perspective: Pierre Le

Notes to Pages 37—41 235 Gentil, "Sur I'e^ilogue de Tristan de Thomas," in Melanges Jeanne Lods (Paris: Collection de 1'Ecole Normale Superieure de Jeunes Filles no. 10, 1978), vol. I, pp. 365-70; Larry M. Sklute, "The Ambiguity of Ethical Norms in Courtly Romance," Genre n (1978): 315-32. 2. See Dannenbaum, "Doubling;" Joan Ferrante, "Artist Figures in the Tristan Stories," Tristania 4 (1979): 25-35; John Grigsby, "L'empire des signes chez Beroul et Thomas: 'Le sigle est tut neir,'" Melanges FoulonlMarche Komane 30 (1979): 118—22; Jodogne, "Comment Thomas"; Ruthmarie H. Mitsch, "The Monologues of Tristan in the Tristan of Thomas," Tristania 2 (1977): 29-39; Ann Trindade, "The Enemies of Tristan," Medium Aevum 43 (1974): 6-21. 3. Chantal Verchere, "Peripherie et croisement: aspects du nain dans la litterature medievale," Senefiance 5 (1978): 261-62. 4. Thomas, Les Fragments du Roman de Tristan^ ed. Bartina H. Wind, Textes Litteraires Fran^ais, 92 (Geneva: Droz, 1960), D, w. 977-84, 987-96; emphasis added. The page numbers given with this and subsequent translations refer to Gottfried von Strassburg, Tristan, tr. A. T Hatto, (New York: Penguin Books, 1975), which I have occasionally modified as indicated to follow more closely the Old French text. In general, I limit my analysis of Thomas's narrative techniques to what can be learned from the extant fragments alone, since the reconstructions of his text cannot reveal nuances of his style, the exact nature of his interventions, and so on, even if they may serve as some indication of the narrative events included in his version. 5. Cf. the nexus of (sexual) knowledge, identity, and power explored in the chapter on Partonopeu de Blois. 6. See also Sn1, w. 172, 197, 209, 211. Cf. Sn1, w. 285, 289; T1, v. 141; D, v. 500. Esprover (Sn1, v. 163; T1, v. 147) and proveir (Sn1, v. 595) function as synonyms for the same idea. 7. Cf. also D, w. 1195-96, when Tristan sends greetings to Iseut through Kaherdin: "Dites li saluz de ma part / Que nule en moi senz li n'a part." 8. This passage does not appear in Hatto's translation. See also Sn1, w. 54, 55,168, 453, 802; D, w. 1120,1238,1240,1245. Cf. departir (Sn1, v. 409) and departie (C, v. 29; Sn1, v. 82). 9. The text delights in playing with phrases that alternately distinguish and conflate the two Iseuts (I'altre Tsolt vs. ceste Tsolt or la meschine vs. la reine: e.g., Sn1, w. 305—68) and the two Tristans (D, w. 1010,1037,1046,1052). 10. See also Sn1, w. 216, 300 (cf. 335-42), 350; T1, v. 123; D, v. 79,1037,1290. 11. Cf. Freud's association of the repetition complex and the death instinct in Beyond the Pleasure Principle^ tr. and ed. James Strachey (New York. Norton, 1961), with Tristan and Iseut's desire for unity through doubling as it leads ultimately to their deaths. Peter Brooks, "Freud's Masterplot: Questions of Narrative," Tale French Studies 55—56 (1977): 280—300, offers an evocative analogue of Thomas's own storytelling habits, traits he shares with romancers in general (see below the chapters on Chretien's Charrete and Partonopeu). In the extant fragments alone, the pattern of retelling occurs with significant insistence: e.g., C, w. 41—42; D, w. 741-52,1214-48. Repetitions, as doubles of the past, sustain memory and hope to turn the past into a new (reenacted) present. Tristan's play on the opposition

236

Notes to Pages 41-51

oblierlmembrer (Sneyd1, w. 23, 47, 96, 97, 104, 106, 118, 119, 125, 126, 153, 154, 176, 178, and the narrator's commentary, w. 190,191, 244, 407; Douce, w. 14,118,1214, 1227,1243,1468,1473,1682; Sn2, v. 798) dramatizes how much his ermnces have put into question the power of their past fidelity to guarantee present love. Cf. the analysis of the Oxford Folly presented in Chapter i. 12. Cf. Joan Ferrante, "Artist Figures," pp. 27-28. 13. Franchise Barteau, Les romans de Tristan et Tseut: introduction a une lecture plurielle (Paris: Larousse, 1972), p. 222. 14. Both Jean Frappier ("Sur le mot 'raison' dans le Tristan de Thomas d'Angleterre," in Linguistic and Literary Studies in Honor of Helmut A. Hatzfeld, ed. A. S. Crisafulli [Washington, D.C.: Catholic University of America Press, 1964], pp. 174—75, n. 46) and Omer Jodogne ("Comment Thomas," pp. 104—5) discuss Thomas's use of desir, poeir^ and voleir. While I wholeheartedly agree with their insistence on the changes of meaning attributed to these terms, it does not seem to me to be a lack of precision on Thomas's part, so much as a sign of his penetration into the world of his character's own ambiguity and ambivalence. 15. Cf. Peter Haidu's discussion of possible meanings for the medieval mentality generated by configurations of two and three, in "The Hermit's Pottage: Deconstruction and History in Yvain" in the The Sower and His Seed: Essays on Chretien de Troyes, ed. Rupert T. Pickens (Lexington, Ky.: French Forum Pub., 1982), p. 142. 16. Mitsch, "Monologues," p. 31. 17. Hatto translates this line: "Truly I do not know what to say," p. 305. The subordinate clause introduced by que might equally well be translated "what I am saying." 18. In her edition of the Tristan fragments, Bartina Wind's explanation of "dire son bon" recalls and applies Foerster's definition of that expression as it appears in Climes: "dire ce qu'on a sur le cceur" (Thomas, Fragments^ p. 140). 19. Mitsch, "Monologues," pp. 35, 39 n. 13 20. Cf. the Charrete, ed. Mario Roques, Classiques Francois du Moyen Age 86 (Paris: Champion, 1967), w. 4550—51: "Bien poez entendre et gloser, / vos qui avez fet autretel" ("You can well understand and gloss, you who have done the same"). Thomas invents his own particular use of a common romance device. Cf. Robert Harming, "The Audience as Co-Creator of the First Chivalric Romances," Yearbook of English Studies n (1981): 1-28; and Michel Zink, "Chretien et ses contemporains," in The Legacy of Chretien de Troyes, ed. Nor r is J. Lacy, Douglas Kelly, and Keith Busby, Vol. I (Amsterdam: Rodopi, 1987), pp. 20-21. 21. Based on Brother Robert's translation, Merritt Blakeslee argues in Love's Masks: Identity, Intertextuality and Meaning in the Old French Tristan Poems (Cambridge: D. S. Brewer, 1989), that Thomas's version originally included an episode of intertwining trees growing out of the lovers' graves. 22. Le Gentil, "Sur Pepilogue," p. 367. Jodogne, "Comment Thomas," p. 112, points out that amor never appears in the fragments to represent a personified god of love, but only refers to the passion of the human participants. 23. The length of Thomas's epilogue stands out as unusual in the context of twelfth-century romance and calls attention to the gesture of closing (cf. Emmanuele Baumgartner, Tristan et Iseut: de la legende au recit en vers [Paris: Presses

Notes to Pages 51—60

237

Universitaires de France, 1987], p. 76). See further discussion of epilogues in contemporary practice and in specific works analyzed here in Chapters 3, 4, and 5 on the Charrete, Partonopeu de Blois, and Marie de France's Lais. 24. Baumgartner and Wagner, "'As enveisiez,'" pp. 527-39. 25. Cf. Douglas Kelly's "Topical Invention in Medieval French Literature," in Medieval Eloquence: Studies in the Theory and Practice of Medieval Rhetoric, ed. James J. Murphy (Berkeley: University of California Press, 1978), pp. 231-51. Kelly stresses the role of the author who uses rhetoric for the purpose of persuasion, but the question "to what end" is still open, given rhetoric's power to argue pro and con (see especially the discussion of the praeexercitamina, pp. 235-36). Tony Hunt, "Aristotle, Dialectic, and Courtly Literature," Viator 10 (1979): 95—129, discusses argumentation in Thomas and Chretien. 26. Douglas Kelly, "'En uni dire' (Tristan, Douce, 839) and the Composition of Thomas' Tristan" Modern Philology 67 (1969—70): 9—17. 27. Cf. the Frappier article on Thomas's use of the word "raison" (already cited) and Hunt's response, "The Significance of Thomas' Tristan" Reading Medieval Studies 7 (1981): 41-61. 28. We might notice what Thomas himself does not specifically say about the version he chooses to follow: it furnishes a better double for Tristan's situation if the lady is Tristan le Nain's bele amie, rather than his wife, as in the rejected version. 29. Pierre Gallais, "Bleheri, la cour de Poitiers et la diffusion des recks arthuriens sur le continent," Actes du VIP congres national de litterature comparee: Poitiers 27-29 mai 1965: moyen age et litterature comparee (Paris: Didier, 1967), pp. 47—79. 30. See the earlier analysis of that whole passage to appreciate once again how reason operates with the testimony of personal experience: the argument by sympathy convinces by its logical demonstration. Cf. also D, w. 1811-12. 31. Frappier, "Sur le mot 'raison.'" 32. See, for example, Erec, w. 2016 and 5208, or Guigemar, v. 523, for pleasures left undescribed. Brewster Fitz's discussion of Marie's Prologue focuses on the links between the surplus, gloss as interpretation (whether "extra" or "the rest"), and desire (talenz): "The Prologue to the Lais of Marie de France and the Parable of the Talents: Gloss and Monetary Metaphor," Modern Language Notes 90 (1975): 558-6433. Bedier's reconstruction of Thomas's plot suggests that a number of episodes were left out: e.g., episodes with the hermit Ogrin and the Folly episode (Roman de Tristan, 2 vols. [Paris: Firmin-Didot, 1902,1905], vol. 2). 34. On the fictional character of the narrator, see Rainer Warning, "Pour une pragmatique du discours fictionnel," Poetique 38 (1979): 221-337.

Chapter 3 i. Chretien de Troyes, Les Romans de Chretien de Troyes: III. Le Chevalier de la Charrete, ed. Mario Roques, Classiques Francois du Moyen Age 86 (Paris: Champion, 1967), w. 5563, 5564, 5571, 5963 (cf. w. 5617, 5618). The translation is

238

Notes to Pages 60-64

from Chretien de Troyes, Lancelot or, The Knight of the Cart (Le Chevalier de la Charrete), ed. and tr. William W. Kibler, Garland Library of Medieval Literature series A, vol. I (New York: Garland, 1981), v. 5563 (verse numbers generally correspond between the translation and Roques's edition). Subsequent translations will be either Kibler's (in verse) or my own (in prose), and aim in all cases to give an English equivalent that allows the reader to follow the play of the Old French. 2. See, for example, Norris J. Lacy, Douglas Kelly, and Keith Busby, eds., The Legacy of Chretien de Troyes, 2 vols. (Amsterdam: Rodopi NV, 1987, 1988); Beate Schmolke-Hasselmann, Der arthurische Versroman von Chrestien bis Froissart: Zur Geschichte einer Gattung, Beihefte zur Zeitschrift fur romanische Philologie 177 (Tubingen: Max Niemeyer, 1980). 3. Cf. David J. Shirt, "Godefroi de Lagny et la composition de la 'Charrete,'" Romania 96 (1975): 27-52. 4. Lucien Dallenbach, Le recit speculaire: essai sur la mise en abyme (Paris: Seuil, 1977), p. 175. This can be represented schematically as: Ai + Ni + X + Az + N2 (where A = approach, N = nexus, X = crisis). See Haidu's "Narrative Structure in Floire et Blancheflor: A Comparison with Two Romances of Chretien de Troyes," Romance Notes 14 (1972): 383-86, and "Narrativity and Language in Some Twelfth Century Romances," Tale French Studies 51 (1974): 133-46; Bruckner, Narrative Invention in Twelfth-Century French Romance: The Convention of Hospitality (1160-1200) (Lexington, Ky.: French Forum Pub., 1980), pp. 172-75; and Donald Maddox, "Trois sur deux: theories de bipartition et de tripartition des oeuvres de Chretien de Troyes," CEuvres et Critiques 5 (1980-81): 91-102. Wolfgang Brand's discussion of bipartite structure in the Charrete identifies the tournament episode as the pivotal moment when the romance changes direction: Chretien de Troyes: zur Dichtungstechnik seiner Romane, Freiburger Schriften zur romanischen Philologie 19 (Munich: Wilhelm Fink, 1972), pp. 55-62. 6. For example, does the first battle with Meleagant mark the first nexus, even though that stage usually occurs in the first third of the romance and we are here already in the middle? Or does the even later night of love with the Queen constitute the first nexus? The final events of the romance are equally problematic: they may achieve a resounding harmony on the level of combat, given Lancelot's final victory over Meleagant, but they do not seem to represent a higher level of harmony with respect to the love relationship between Lancelot and the Queen. 7. See w. 5367, 5394, 54O3, 5432, 5451, 5588, 5728, 5765.

8. Dallenbach's definition of mise en abyme requires it to operate at the level of diegesis, assumed by a character included in the narrative (Le recit speculaire, pp. 70-73), as in Tvain, for example, when Calogrenant recounts his own adventure; cf. Marie-Louise Oilier, "Le discours en abyme ou la narration equivoque," Medioevo Romanzo i (1974): 351-64. While the Charrete tournament may not qualify as mise en abyme according to this definition, Chretien's heavy reliance on the characters' representation of events brings the episode so close that the concept of mise en abyme offers a useful model for analysis. 9. As early as the thirteenth century, real tournaments used literary tournaments as their models. See Ruth Huff Cline, "The Influence of Romances on

Notes to Pages 64-72 239 Tournaments of the Middle Ages," Speculum 20 (1945): 204-11; and Maurice Keen, Chivalry (New Haven, Conn, and London: Yale University Press, 1984), pp. 92-94. The dynamic interaction of romance and history is neither simple nor one-sided. 10. See Sidney Painter, French Chivalry: Chivalric Ideas and Practices in Medieval France (Baltimore: Johns Hopkins University Press, 1940), pp. 54-60, 88-89, 155; and the more recent study by Keen, Chivalry, especially pp. 83-101. Keen characterizes the tournament as a powerful tool for both literal and symbolic communication in the Middle Ages, which provides "a crucial link between the literary expression of chivalrous values and the real world" (p. 100). n. Peter Haidu, "Repetition: Modern Reflections on Medieval Aesthetics," Modern Language Notes 92 (1977): 875-87; Michel Zink, "Une mutation de la conscience litteraire: le langage romanesque a travers des exemples fran^ais du XII> siecle," Cahiers de Civilisation Medievale 24 (1981): 3-27 (summarized in Zink, La subjectivite litteraire: autour du siecle de saint Louis [Paris: Presses Universitaires de France, 1985], pp. 27-46). 12. F. Xavier Baron, "Love in Chretien's Charrete\ Reversed Values and Isolation," Modern Language Quarterly 34 (1973): 377-83; Charles Foulon, "Les deux Humiliations de Lancelot," BBSIA 8 (1956): 79-90; Douglas Kelly, "Sens" and "Conjointure" in the "Chevalier de la Charrete" (The Hague: Mouton, 1966), p. 149. 13. The same sort of narrative loop appears at the end of the tournament scene, when Lancelot's return is narrated and then repeated as a signal for Meleagant to begin the tower building (w. 6057-58, 6106-15). The two narrative loops create a symmetrical frame around the tournament episode that reinforces its isolation in the narrative sequence. 14. See, for example, Moshe Lazar, "Lancelot et la 'mulier mediatrix': la quete de soi a travers la femme," L'Esprit Createur 9 (1969): 243-56. 15. Matilda Tomaryn Bruckner, "An Interpreter's Dilemma: Why Are There So Many Interpretations of Chretien's Chevalier de la Charretet" Romance Philology 40 (1986): 159—80. 16. On the other hand, there is already a marked fictionality in that scene, where the rape is organized and directed by the Hospitable Damsel. See Bruckner, "Essential and Gratuitous Inventions: Thomas's Tristan and Chretien's Lancelot'' Actes du i4e Congres International Arthurien (Rennes: Presses Universitaires de Rennes, 1986), pp. 120-41. On the rape scene in romance, see Kathryn Gravdal, Ravishing Maidens: Writing Rape in Medieval French Literature and Law (Philadelphia: University of Pennsylvania Press, 1991). 17. Cf. Kelly, "Sens" pp. 62-63. 18. This is another link between the tournament and the cart episodes, since both furnish, according to the narrator's comments, the origin of a proverb (cf. w. 339-44). See Marie-Louise Oilier, "Proverbe et sentence: le discours d'autorite chez Chretien de Troyes," Revue des Sciences Humaines 41 (1976): 329-57. Paule Le Rider suggests in "Or est venuz qui I'aunera ou la fortune litteraire d'un proverbe," in Melanges Jeanne Lods (Paris: Collection de 1'Ecole Normale Superieure de Jeunes Filles 10, 1978), Vol. I, pp. 393-409, that Chretien is really taking a street cry that was commonly used in the mercantile streets and fairs of Champagne to announce

24O

Notes to Pages 72-84

the arrival of the official who verified the accuracy and standardization of measurements. The displacement from commercial to chivalric context helps explain by its incongruity some of the crowd's surprise and confusion, when thay hear the call and ask what it means (w. 5566-67). 19. Cf. the history of philological dispute about auner described by Le Rider ("Orestvenuz" pp. 394-96). 20. Le Rider, "Or est venuz" p. 39321. Cf. Robert Manning's strong argument for romance as symbolic and parodic, demanding interpretation of puzzles without giving answers: "The Audience as Co-Creator of the First Chivalric Romances," Yearbook of English Studies n (1981): 1-28 (esp. pp. 13-14). 22. See my discussion of delay as the characteristic mode of open-endedness in "Le Chevalier de la Charrete (Lancelot)," in The Romances of Chretien de Troyes: A Symposium, ed. Douglas Kelly. Edward C. Armstrong Monographs on Medieval Literature 3 (Lexington, Ky.: French Forum Pub., 1986), pp. 167-75. 23. This pattern will be elaborately amplified in the tournament episode that crowns Partonopeu's marriage quest of Melior. See the discussion in Chapter 4. 24. This twofold use of the message au noauz recalls the process of recognition in the Folie Tristan d'Oxford, as divided between Iseut's point of view (trying to figure out who the fool is) and Tristan's (testing the Queen's fidelity). Cf. also the role of belief in the interpretation of signs, as explored in the Folly episode. The Charrete seems to go further in problematizing the credibility of signs. 25. Kelly, "Sens" pp. 214-15. 26. Cf. David Hult, "Author/Narrator/Speaker: The Voice of Authority in Chretien's Charrete" in Discourses of Authority in Medieval and Renaissance Literature, ed. Kevin Brownlee and Walter Stephens (Hanover, N.H.: University Press of New England), pp. 76-96. 27. Cf. Donald Maddox, "Roman et manipulation au XII> siecle," Poetique 66 (1986): 179-90. 28. Cf. Juri Lotman, La structure du texte artistique, tr. A. Fournier et al. (Paris: Gallimard, 1973), pp. 332-39, on the "personnage mobile." 29. See my "Interpreter's Dilemma," pp. 159-60. 30. Jean Rychner, "Le sujet et la signification du Chevalier de la Charrete" VoxRomanica 27 (1968): 72-73; Richard L. Michener, "Courtly Love in Chretien: The 'Demande d'amour,'" Studia neophilolqgica 42 (1970): 353-60; Jean Frappier, "Le Prologue du Chevalier de la Charrete et son interpretation," Romania 93 (1972): 337-77; Jerome Mandel, "Proper Behaviour in Chretien's Charrete" French Review 48 (1975)- 683-89; Jean Deroy, "Chretien de Troyes et Godefroi de Leigni, conspirateurs contre la Fin'Amor adultere," Cultura Neolatina 38 (1978): 67—78; Peter S. Noble, Love and Marriage in Chretien de Troyes (Cardiff: University of Wales Press, 1982), pp. 65-82 (see my review in Romance Philology 39 [1986]: 512); Derek Brewer, "The Presentation of the Character of Lancelot: Chretien to Malory," Arthurian Literature 3 (1983): 26—52. 31. The opening lines of the romance as it appears in Wendelin Foerster's edition, "Des que ma dame de Champaigne / Viaut que ..." (Christian von Troyes, Samtliche erhaltene Werke: IV. Der Karrenritter und das Wilhelmsleben [Halle: Nie-

Notes to Pages 84—90

241

meyer, 1899]) are repeated in Lancelot's reply the second day: "Des qu'ele le comande, / li respont, la soe merci" (Le Chevalier de la Charrete, ed. Roques, w. 5856-57). Cf. Brand, Chretien de Troyes^ pp. 139-44, where he comments on the fictional persona of the narrator in each of Chretien's romances, as it responds to the main themes of the romance in question. 32. Our interest in Chretien vis-a-vis Marie and Godefroi may be more particularly the reaction of modern readers than that of a medieval audience whose reception is still largely influenced by an oral reading of the romance. On the other hand, Chretien's self-representation as narrator does inscribe within the romance certain ambiguities in his stance vis-a-vis public and story, at once separated from and enmeshed in the story he tells. On the ramifications of Chretien's irony, see Peter Haidu, "Le sens historique du phenomene stylistique: la semiose dissociative chez Chretien de Troyes," Europe 642 (1982): 36-47, and Rupert T. Pickens, "Historical Consciousness in Old French Narrative," French Forum 4 (1979): 168-84. 33. I. D. O. Arnold and M. M. Pelan, eds., La Partie arthurienne du Roman de Brut (Paris: Klincksieck, 1962), w. 1247-58. 34. Cf. Erich Kohler, Uaventure chevaleresque: ideal et realite dans le roman courtois^ tr. Elaine Kaufholz (Paris, Gallimard, 1956,1974), pp. 7-4335. See Gerard Genette, Figures III (Paris: Seuil, 1972), pp. 265—67. Hult argues directly for the fictionality of Godefroi ("Author," p. 88); Roger Dragonetti speculates in a similar manner in La vie de la lettre au moyen age (Le Conte dugraal) (Paris: Seuil, 1980), pp. 13—18. Cf. Rainer Warning, "Pour une pragmatique du discours fictionnel," Poetique 38 (1979): 221-37. On the historical status of the twelfth-century writer as "Sujet semiotique," see Haidu, "Le sens historique," PP- 45-4736. The lack of any scribal notice of the change in authors, as represented in the Charrete's manuscript tradition (Hult, "Author"), mirrors the effect of an oral reading of the romance. In "Narrative Voices in Chretien de Troyes: A Prolegomenon to Dissection," Romance Philology 32 (1979): 261-73, John L. Grigsby, compares Godefroi to Calogrenant and thus suggests a link between the Charrete and Tvain. 37. Alison Adams, "Godefroi de Leigni's Continuation of Lancelot" Forum for Modern Language Studies 9 (1974): 295—99. 38. Jean Frappier, Etude sur (CYvain" ou "le Chevalier au Lion" de Chretien de Troyes (Paris: SEDES, 1969), pp. 12-16; and Shirt, "Godefroi de Lagny" and "How Much of the Lion Can We Put Before the Cart?" French Studies 31 (1977): 1—17. 39. Cf. Roberta L. Krueger, "Love, Honor, and the Exchange of Women in Tvain: Some Remarks on the Female Reader," Romance Notes 25 (1985): 302-17, esp. pp. 315-17, where she analyzes the way different readers can see the ending as satisfactorily happy or paradoxically unsatisfactory. 40. Kelly, ed., Romances, pp. 162—68. See also Donald Maddox, "Lancelot et le sens de la coutume," Cahiers de Civilisation Medievale 29 (1986): 339—53. 41. The Charrete*s 7,112 verses (in Roques's edition) are not significantly longer than Erects 6,877 or Yvairfs 6,808, but they appear relatively short when compared with other romances that include continuations, like Partonopeu de Blois (10,656 verses plus almost 4,000 more in the Continuation), or even Chretien's

242

Notes to Pages 90-95

Conte du Graal (left unfinished at over 9,000 verses and continued by later romancers for over 65,000 more). Inasmuch as Partonoperfs author used Chretien's romances as the model for so many of his reinventions, one wonders if the notion of a romance with a "planned" continuation may not be traceable to Chretien's voluntary and involuntary (?) experiments in that area. 42. Kevin Brownlee's "Transformations of the Charrete: Godefroi de Leigni Rewrites Chretien de Troyes," Stanford French Review 14 (1990): 161-78, offers a sustained effort to analyze the consequences of reading the Charrete as romance and continuation. Despite the Epilogue's declared duality of authorship, however, Chretien's manuscript tradition and the Prose Lancelot's retelling of the Cart episode continue to regard the Charrete as a single unit, even when that unit is combined with others in Chretien's corpus. See Alexandre Micha, La tradition manuscrite des romans de Chretien de Troyes (Geneva: Droz, 1966), p. 37. 43. Anthime Fourrier, "Encore la chronologic des ceuvres de Chretien de Troyes," BBSIA 2 (1950): 69-88; Frappier, Etude sur "Tvain," pp. 13-15. Barbara Nelson Sargent-Baur argues for associating the Charrete and Tvain as romances commissioned by Marie de Champagne, composed simultaneously, and sharing the Charrete prologue: "The Missing Prologue of Chretien's Chevalier au Lion" French Studies 41 (1987): 385-94. 44. Manuscripts that contain both Tvain and the Charrete place first now one, now the other: e.g., Tvain I Charrete (B.N. 12560, Chantilly 472, B.N. 1450); Charrete I Tvain (Vatican 1725, B.N. 794—with Cliges in between). As Micha points out, the changing order from one manuscript to another suggests that no order was established either by Chretien or the tradition, but determined by availability and the tastes of scribe or collector (La Tradition manuscrite, pp. 277—79). 45. Cf. Karl D. Uitti, "Le Chevalier au Lion (Tvain)" in Romances, ed. Kelly, pp. 219-27. 46. The coordination of these two romances, already apparent in the parrallels suggested by their titles—Le Chevalier de la Charrete, Le Chevalier au Lion— invites exploration at all levels of the echoes between them. For some promising forays, see Roberta L. Krueger's "Reading the TvainlCharrete: Chretien's Inscribed Audiences at Noauz and Pesme Aventure," Forum for Modern Language Studies 19 (1983): 172-87; and "Contracts and Constraints: Courtly Performance in Tvain and the Charrete"Houston German Studies6 (1986): 92-104. 47. We might compare this with Thomas's commentary about other versions and his insistence on the truth of his own: Douce, w. 835-84, and Sneyd2, w. 828—29. See the discussion in the previous chapter. 48. Paul Zumthor, "Intertextualite et mouvance," Litterature 41 (1981): 8-16. 49. Jean-Charles Payen, "Lancelot contre Tristan, ou la conjuration d'un mythe subversif," in Melanges de langue et de litterature medievales offerts a Pierre Le Gentil, ed. J. Dufournet and D. Poirion (Paris: SEDES, 1973), pp. 617-33. 50. Cf. Michelle Freeman's discussion of the links between Cliges and the Tristan story in Romances, ed. Kelly, pp. 115-30. 51. This kind of separation between love and prowess, at least in terms of society's benefits, may be reflected in the first part of Chretien's Cliges, where readers have often commented on the way episodes focused respectively on love

Notes to Pages 95-105 243 and combat alternate rather than integrate Alexandra's love for Soredamor and his knightly service for Arthur. 52. Cf. Donald Maddox, "Specular Stories, Family Romance, and the Fictions of Courtly Culture," Exemplaria 3 (1991): 299—326. 53. Bruckner, "Interpreter's Dilemma." 54. Le Rider, "Or est venuz," pp. 408-9. 55. Gottfried von Strassburg, Tristan, tr. A. T. Hatto (New York, Penguin Books, 1975), pp. 241-42. 56. May we understand in this light why Chretien refers to one of his works, enumerated in the Prologue to Cliges, as something on King Mark and Iseut: "Del roi Marc et d'Ysalt la blonde" (v. 5)? The combination of Mark and Iseut, so enigmatic and atypical, is unexpected; nor is it required by the metrics of the verse, which could easily accommodate the lover's rather than the husband's name in the same spot. 57. Is this a subtle recall of the episode we know from Gottfried (Tristan, tr. Hatto, pp. 205-13), where the white shifts, one dirty, one clean, are used by Brangane as a metaphor for her own and the Queen's virginity? 58. David Shirt, "Chretien de Troyes and the Cart," Studies in Medieval Language and Literature in Memory ofF. Whitehead, ed. W. Rothwell et al. (Manchester: Manchester University Press, 1973), PP- 279-301, and "Chretien de Troyes et une coutume anglaise," Romania 94 (1973): 178—95. 59. Cf. Jerome Mandel, "Elements in the Charrete World: The Father-Son Relationship," Modern Philology 62 (1964): 97-104; Brand, Chretien de Troyes, pp. 68-70. 60. Cf. Donald Maddox, Structure and Sacring: The Systematic Kingdom in Chretien's Erec et Enide (Lexington, Ky.: French Forum Pub., 1978) and The Arthurian Romances of Chretien de Troyes: Once and Future Fictions (Cambridge: Cambridge University Press, 1991). 61. Andre Jolles, Les formes simples, tr. Antoine Marie Buguet (Paris: Seuil, 1972), pp. 137-5762. Though medieval Christianity might claim to offer an ultimate measure for all such competing norms, Chretien does not introduce such a perspective in the Lancelot, as he does explicitly in the Conte dugraal (where it nevertheless poses more problems than it resolves). 63. Cf. Rupert T. Pickens, "Estoire, Lai, and Romance: Chretien's Erec et Enide and Cliges" Romanic Review 66 (1975): 247—62. The relation between lai and romance will be discussed further in the next two chapters. 64. Cf. Nancy Freeman Regalado, "Tristan and Renart: Two Tricksters," L'Esprit createur 16 (1976): 34-37. 65. The Prose Lancelot leaves no doubt that medieval readers expected Lancelot and the Queen's love to continue in both its physical and spiritual dimensions (hence the conflicts in the Queste and the Mort le RoiArtu, when Lancelot tries to abstain from sexual contact with the Queen during the Grail quest and then "relapses" after its completion and his return to court). 66. Cf. the sermon quoted by D. W. Robertson, Jr., in A Preface to Chaucer: Studies in Medieval Perspectives (Princeton, N.J.: Princeton University Press, 1962),

244

Notes to Pages 105—107

pp. 89-90. See also Haidu, "Repetition," pp. 880-83, and E. Jane Burns's discussion of the rivalry between Rhetoric and Scripture from the third to the thirteenth centuries, in Arthurian Fictions: Rereading the Vulgate Cycle (Columbus: Ohio State University Press for Miami University, 1985), pp. 21-27. 67. This short overview follows closely the discussion offered by Peter Haidu in "Repetition," pp. 875-77. 68. On the other hand, we should remember the Chartrians' interest in using fictions to find or express divine truths on the allegorical level. See, for example, Winthrop Wetherbee, Platonism and Poetry in the Twelfth Century: The Literary Influence of the School of Chartres (Princeton, N.J.: Princeton University Press, 1972). Franz Bauml makes a strong argument for aligning the expectations of the romance public, at least in its first stages, with those of a literate public formed by Latin writing (with its typological or tropological reception): "Varieties and Consequences of Medieval Literacy and Illiteracy," Speculum 55 (1980): 237-65. 69. Douglas Kelly, "Romance and the Vanity of Chretien de Troves," in Romance: Generic Transformation from Chretien de Troyes to Cervantes, ed. Kevin Brownlee and Marina Scordilis Brownlee (Hanover, N.H.: University of New England Press, 1985), pp. 74-90. 70. Cf. Nancy Freeman Regalado, "'La Chevalerie CelestieP: Spiritual Transformations of Secular Romance in La Queste del Saint Graal" in Romance, ed. Brownlee and Brownlee, pp. 91—113. 71. In a modern context, cf. Michael Riffaterre's Fictional Truth (Baltimore and London: Johns Hopkins University Press, 1990). 72. On the other hand, both Chretien and Marie bring together the issues of truth and meaning, each in a distinctive way, according to the particularities of romance and lais. 73. This is the only explicit truth claim attached to the story's source in any of Chretien's prologues. As such, it provides an ironic commentary on the conflation of true historical facts and fiction in the most obviously fabulous and rhetorical of Chretien's romances. Cf. the discussion of history and romance in the following chapter, as well as Erich Kohler's analysis of what happens when romance mixes legend (L'aventure, pp. 118—19). 74. See Bruckner, "Interpreter's Dilemma," pp. 173-74, and Kelly, ed., Romances, pp. 132-41. 75. Cf. below the discussion of two kinds of authority, textual and experiential, represented and intertwined in the figure of Marie de France, who as female clerk has collected and written down a group of oral tales, thereby combining vernacular and Latin traditions in a new way. 76. The range of meanings associated with san in twelfth-century usage includes meaning, understanding, good sense, sense, and direction. See Karl D. Uitti, Story, Myth, and Celebration in Old French Narrative Poetry, 1150-1200 (Princeton, N.J.: Princeton University Press, 1973), PP-135-36 n. 6, and Albert Gier, "Das Verwandtschaftsverhaltnis von afr. sens und sen" RomantischesJahrbuch 28 (1977): 54-72. 77. I have repeatedly used Julia Kristeva's term, non-disjunctive opposition, to characterize this mode of thought and action in romance, Semeiotike: recherches

Notes to Pages 107—113

245

pour une semanalyse (extra-its), Collection Points (Paris: Seuil, 1969), pp. 55-62. Cf. Eugene Vance's discussion of the way Chretien uses certain figures of thought, all of which have in common a play on difference and identity according to varied degrees of opposition, "Signs of the City: Medieval Poetry as Detour," New Lit erary History 4 (1973): 557-74, especially pp. 567-70.

Chapter 4 1. Anthime Fourrier, Le coumnt realiste dans le roman courtois en France au moyen age: I. Les debuts (XIIe siecle) (Paris: Nizet, 1960), pp. 315-17, 441-46. 2. Arguments for or against the "authenticity" of the Continuation are inconclusive. Fourrier (Le courant realiste, pp. 316-17) and K. Sneyders de Vogel ("La Suite du Parthenopeu de Blois et la version hollandaise," Revue des Langue Romanes 48 [1905]: 5-29) argue for considering the Continuation as part of the "Vulgate" and probably by the original author. Leon Smith and Joseph Gildea, the latest editors, argue for two different authors (Partonopeu de Blois: A French Romance of the Twelfth Century [Villanova, Pa.: Villanova University Press, 1967], vol. 2, pt. 2, p. 3). Medieval readers of a manuscript with a continuation would be unlikely to worry about such a problem. 3. S. P. Uri, "Some Remarks on Partonopeus de Blois" Neophilologus 37 (i953): 934. The story of Cupid and Psyche offers fertile ground for exploring how a culture defines and distributes roles according to gender. Cf. Erich Neumann, Amor and Psyche: The Psychic Development of the Feminine—a Commentary on the Tale ofApuleius, tr. Ralph Manheim, Bollingen Series 54 (Princeton, N. J.: Princeton University Press, 1971), and the critique offered by Lee R. Edwards, "The Labors of Psyche: Toward a Theory of Female Heroism," Critical Inquiry 6 (Autumn 1979): 33-495. Partonopeu de Blois, w. 63-71. The translations are my own and follow as literally as possible the language of the original. 6. Cf. Roberta L. Krueger, "The Author's Voice: Narrators, Audience, and the Problem of Interpretation," in The Legacy of Chretien de Troyes, ed. Norris J. Lacy, Douglas Kelly and Keith Busby (Amsterdam: Rodopi NV, 1987,1988), vol. i, pp. 115-40; and "Textuality and Performance in Partonopeu de Blois" in Assays: Critical Approaches to Medieval and Renaissance Texts 3 (1985): 57-72. See also Michel Zink's discussion of autobiographical fiction in lyric and romance in La subjectivite litteraire: autour du siecle de saint Louis (Paris: Presses Universitaires de France, 1985), pp. 44-46. 7. Robert Manning, "The Audience as Co-Creator of the First Chivalric Romances," Yearbook of English Studies n (1981): 12-14. Cf. also Rainer Warning, "Pour une pragmatique du discours fictionnel," Poetique 38 (1979): 221-37. 8. "Audience," pp. 17—18. Cf. also Joan Ferrante, "Male Fantasy and Female Reality in Courtly Literature," Women}s Studies n (1984): 88-91. 9. The expression "fables as Sarrasins" not only raises the issue of stories

246

Notes to Pages 113-117

that may be untrue but also involves the question of pagans, problematic figures from a Christian point of view. As Roland puts it succinctly, pagans are wrong, Christians right: "Paien unt tort e chrestiens unt dreit" (Joseph Bedier, ed. La Chanson de Roland [Paris: Piazza, 1964], v. 1015). His sentiments are echoed in the Sornegur episode of Partonopeu, when the pagan kings meet to discuss whether or not to continue the war: "'II ont le droit et nos le tort; / Serons nos dont por ce mains fort?'" (w. 2440-41: "They're in the right and we're in the wrong. Will we be less strong for that?'"). Fursin's question measures the difference between the world view of a chanson de geste and that of romance, as it suggests how the romancer of Partonopeu plays with genres and mixes categories. 10. See the passage quoted in the Charrete chapter. 11. Cf. the Prologue to Chretien's Cliges (w. 18-24), where he identifies the written source of his story as a guarantor of truth: is the book at Beauvais merely a literary topos or should we believe Chretien's claim? In its overt and covert play with the intersection of truth and fiction, contemporary history and fabulous romance, Cliges certainly paves the way for Partonopeu de Blois. 12. The translation from Latin into the vernacular is a topos of romance, from which it derives its name (mettre en romanz). But some romancers elaborate more insistently the difficulties or possibilities of such translatio. Cf. Hue de Rotelande's Prologue to Ipomedon, w. 21-46. See the next chapter's discussion of Marie de France's General Prologue to the Lais, where she intertwines Latin and vernacular traditions in combination with oral and written ones. 13. Cf. Stephen Nichols's discussion of John the Scot, his views of history as mysteria engaging historical and narrative events, and the implications for vernacular inventio: "Remodeling Models: Modernism and the Middle Ages" in Modernite au moyen age: le defi du passe, ed. Brigitte Gazelles and Charles Mela (Geneva: Droz, 1990), pp. 45-67. 14. Fourrier, Le courant realiste, pp. 441—42. 15. On genealogies in the twelfth century, see R. Howard Bloch, Etymologies and Genealogies: A Literary Anthropology of the French Middle Ages (Chicago and London: University of Chicago Press, 1983), pp. 80-81. Cf. Elisabeth Schmid, Familiengeschichten un Heilsmythologie: die Verwandtschafitrukturen in den franzosischen und deutschen Gralromanen des 12. und 13. jahrhunderts, Beihefte zur Zeitschrift fur romanische Philologie 211 (Tubingen: Max Niemeyer, 1986), on genealogy and the contributions of matrilinear and patrilinear lines of descent, particularly in relation to the accessibility of knowledge. On the narrator's treatment of Trojan descent and the Merovingian line, see Fourrier, Le courant realiste, pp. 392-98. 16. Cf. Robert Hanning's discussion of chivalric romance in the twelfth century as "lacking in a developed view of history with which the hero could interact," "Beowulf as Heroic History," Medievalia et Humanistica 5 (1974): 83. 17. Paul Zumthor, Essai depoetique medievale (Paris: Seuil, 1972), pp. 346-51. 18. Douglas Kelly, "Matiere and Genera dicendi in Medieval Romance," Tale French Studies 51 (1974): 147-59. 19. Cf. Carole Berconici-Huard, "Partonopeus de Blois et la couleur byzan-

Notes to Pages 117-121 247 tine," in Images etsignes de POrient dans ^Occident medieval: litterature et civilisation (Aix-en-Provence: CUERMA, Universite de Provence, 1982), p. 192. 20. Fourrier, Le courant realiste, pp. 401-2. 21. Cf. Jacques Derrida, De la grammatologie (Paris: Minuit, 1967), pp. 203-34. 22. Douglas Kelly, "The Art of Description," in Legacy, ed. Lacy, Kelly, and Busby, vol. i, pp. 209-10. Cf. Eugene Vance's discussion of the distinction between formal truth and ontological truth, based on topical argumentation from principles of logic (in From Topic to Tale: Logic and Narrativity in the Middle Ages, Theory and History of Literature 47 [Minneapolis: University of Minnesota Press, 1987], esp. pp. 47-48, 52, 80-81). 23. En0in in the romance world is polyvalent and can be used for good or bad purposes: the narrator's own self-portrait includes the adjective engignous (v. 65). For a study of engin in twelfth-century romance, see Robert Manning's The Individual in Twelfth-Century Romance (New Haven, Conn.: Yale University Press, 1977), pp. 105-38. 24. Fourrier has discussed this aspect of Partonopeu in light of contemporary history as a reflection of the author's conservative political philosophy critical of royal policy regarding non-noble auxilium et consilium. He has also interpreted the linking of the house of Blois with the Merovingian line as a glorification of the poet's patron (Le courant realiste, pp. 411-28 and 397). Later I expect to nuance Fourrier's evaluation of the romancer's political conservatism by contrast with the romance's implicit strategies for opening questions. 25. In the Continuation, the debate on the "fils a vilain" that opens Partonopeu's encounter with the lost Anselot includes a discussion of Xerxes's lineage that recalls Eneas's in the genealogy (w. 355-4o8c). 26. Alice M. Colby, The Portrait in Twelfth-Century French Literature (Geneva: Droz, 1965). 27. Cf. Julia Kristeva's discussion of romance and non-disjunctive opposition in "Le texte clos," Semeiotike: recherches pour une semanalyse (extraits), Collection Points (Paris: Seuil, 1969), pp. 55-62. Although Kristeva locates the semiotic shift from symbol to sign in romance of the fifteenth century, with the example of Le Petit Jean de Saintre, I would suggest that romances of the twelfth century are already engaged in the same problematic. 28. In Marie de France's Bisclavret we may find another mise en question of categories and how they operate. See my analysis, "Of Men and Beasts in Bisclavret" Romanic Review 82 (1991): 251-69. 29. G. D. West, "The Description of Towns in Old French Verse Romances," French Studies 9 (1957): 50-59. 30. The narrator's description shows Partonopeu repeatedly ascribing all he sees to fantosme or songe (e.g., w. 879-80, 895-96, 906-10, 916), or fable, which must be the work of diable (w. 983-84, 1054-56). Even after Melior's explanations, the narrator seems to tease his character and public with the ominous black horses and black dogs provided by the Empress: they give Partonopeu a moment of fear, but have no other adverse affects (w. 1616-17,1831-32, 1995). On

248

Notes to Pages 121-127

the romance's subtle and programmatic use of colors, see Pierre-Marie Joris, "Note sur le noir dans le Partonopeu deBlois" Senefiance 24 (1988): 141-52. 31. See Fourrier, Le coumnt realiste, p. 385; Thomas H. Brown, "The Relationship Between Partonopeus de Blois and the Cupid and Psyche Tradition," Brigham Toung University Studies 5 (1964): 193-202; and Uri, "Some Remarks." 32. Helaine Newstead, "The Traditional Background of Partonopeus de Blois" PMLA 61 (1946): 916-46. On fairies in the context of medieval literature in general, see Laurence Harf-Lancner, Les fees au moyen age (Geneva: Slatkine, 1984)Her brief analysis of Partonopeu describes it as a variation that, because of the reversal in male/female roles, combines features of "les contes melusiniens" and "les contes morganiens," the two major models for stories involving fairies (pp. 317-28). 33. Cf. Catherine Hilton, "Convention and Innovation in Partonopeu de Blois" Ph.D. Diss., University of Massachusetts, 1984, pp. 96-108. 34. Newstead, "Traditional Background," pp. 932—35. For example, we might compare the Lady of the Lake's kidnapping and education of Lancelot in the thirteenth-century Lancelot en prose with Melior's luring of Partonopeu to Chef d'Oire, followed by her advice on knighthood and the generous provisions supplied when she sends him back to France to help the French king fight against Sornegur's invasion. 35. Cf. William D. Paden, Jr., "Utrum copularentur: Of Cors" UEsprit createur\9 (1979): 81. 36. Cf. Peter Haidu, "Narrative Structure in Floire et Blancheflor: A Comparison with Two Romances of Chretien de Troyes," Romance Notes 14 (1972): 383-86; my Narrative Invention in Twelfth-Century French Romance: The Convention of Hospitality (1160-1200) (Lexington, Ky.: French Forum Pub., 1980), pp. 172-75; and Donald Maddox, "Trois sur deux: theories de bipartition et de tripartition des oeuvres de Chretien " CEuvres et critiques 5 (1980-81): 91-102. 37. Partonopeu^ intertextual dialogue with Chretien and his tendency to combine disparate elements from Chretien's romances is a good illustration of his special affinity and talent for fusion. Bruckner, Narrative Invention, pp. 152—54; "Repetition and Variation in Twelfth-Century French Romance," in The Expansion and Transformation of Courtly Literature, ed. N.B. Smith and J. T. Snow (Athens: University of Georgia Press, 1980), pp. 101-4; and "Intertextuality," in Legacy, ed. Lacy, Kelly, and Busby, pp. 226-30. See also Hanning, Individual, pp. 216-18. 38. See, for example, in the trobairitz corpus, the tensos between Alamanda and Giraut de Bornelh, Almucs de Castelnau and Iseut de Capio in Meg Bogin, ed. and tr., The Women Troubadours (New York and London: Norton, 1976) and the salut d'amor by Azalais d'Altier in V. Crescini, "Azalais d'Altier," Zeitschrift fur Romanische Philologie 14 (1890): 128—32. 39. In retrospect, this evocation of amor de lonh introduces a scenario that playfully contrasts with the situation in the romance plot. Multiple possibilities for male/female reversals appear, if we compare Melior's faraway love bringing Partonopeu to Chef d'Oire with Jaufre Rudel's poems, in which the male poet dreams of voyages east to the lady (a voyage actually realized in his vidtfs narrative

Notes to Pages 127-137

249

version). Or we might compare Melior with LanvaFs fairy mistress, who journeys from Avalon to Arthur's kingdom for the sake of her beloved. 40. Verse numbers followed by "c" are taken from the Continuation given in Vol. II, Pt. I, of Gildea's edition. 41. Douglas Kelly, "The Logic of the Imagination in Chretien de Troyes," in The Sower and His Seed: Essays on Chretien de Troyes, ed. Rupert T. Pickens (Lexington, Ky.: French Forum Pub., 1983), pp. 22-23. 42. Cf. Colby, Portrait, pp. 72-88. 43- We may appreciate the extent to which Partonopeu^s author uses crisscrossing gender roles to realize this goal, if we compare his tournament ending with the sort of beauty contest that resolves the trial at the end of Lanval. In Marie de France's lai, there is no integration of the couple: Lanval moves completely to the lady's domain and disappears; he remains outside of Arthur's court, just as he was forgotten in the opening distribution of rewards. His lady exercises power through her magic resources as fairy (although Marie never uses that word to identify her), although once Lanval has betrayed their secret, she appears publicly at Arthur's court to save Lanval from the Queen's accusation not through magic, but by the overt power of her physical charms: her beauty alone decides the juridical question without resort either to LanvaPs prowess or to his merit. He owes everything to the lady who carries him off—though he goes by his own choice. The roles of man and woman, beauty and prowess, will be significantly redistributed in Partonopeu, so that each member of the couple may contribute to their union, so that the couple itself may be publicly constituted and acclaimed. 44. Uri, "Some Remarks," p. 22. 45. Cf. E. Bozoky's analysis of "le Chateau desert" as a narrative model whose steps usually lead to liberation or disenchantment, "Roman medieval et conte populaire: le Chateau desert," Ethnologic Fran$aise 4 (1974): 353. Melior's loss of power is a kind of disenchantment located in the character rather than in the place. Chef d'Oire is, we recall, a kind of emanation of Melior and her love for a beautiful Partonopeu. 46. In the feudal vocabulary of the twelfth century, "onor" (w. 5231, 5239) may mean either abstract honor or tangible fief. Melior clearly represents both for Partonopeu. His mother is, by extension, cast in the role of Eve. Consider in this light the son's complaints against her betrayal (w. 5289-94) and the repeated speculations about who plays the devil in this story—a playful shakeup and recombination of the Bible story's elements. 47. Melior's repeated complaints about her shame feed her anger and continued refusal to pardon Partonopeu, but, in fact, we hear no more about that public shame in the second part of the romance, except in Melior's own private monologues and dialogues with Urraque. Cf. the treatment of Lancelot's shame as the cartrider in Chretien's Chevalier de la Charrete: public outcry in the first part, private fault against love in the second. 48. This aspect is particularly tricky during the "rape" scene, when Partonopeu first meets Melior in her bed and overcomes her initial resistence (w. 1267-1304). Newstead has compared her ambivalent conduct to that of the fairy

250

Notes to Pages 137-150

mistress in Gmelent ("Traditional Background," pp. 926-27). See also Hilton, "Convention and Innovation," p. 39 and n. 15. Keith Busby gives a useful analysis of the intersection of fabliau and romance in the representation of Partonopeu's sexual advances: "Cristal et Clarie: A Novel Romance?" in Convention and Innovation in Literature, ed. Theo D'haen, Rainer Griibel and Helmut Lethen (Amsterdam and Philadelphia: John Benjamins, 1989), pp. 93-9549. On the possibilities for play with the linguistic pun Eva/Ave, see, for example, Gail McMurray Gibson, The Theater of Devotion: East Anglian Drama and Society in the Late Middle Ages (Chicago and London: University of Chicago Press, 1989), pp. 137-39,159-66. 50. Cf. Eugene Vance, "Signs of the City: Medieval Poetry as Detour," New Literary History 4 (1973): 557-74, especially where he discusses Henry the Liberal's choice of eighteen guardians of the fair, ten from the bourgeoisie, eight from the nobility. 51. We might compare this with contemporary models of philosophical and theological inquiry which invite the exploration of authorities with a questioning mind, rather than an automatic acceptance of them on the basis of auctoritas alone. Consider this quotation from Abelard's preamble to the Sic et non: . . . we have been pleased . . . to gather the differing remarks of the Holy Fathers, when they occur to us, formulating questions which may provoke young readers to the greatest exercise of inquiry after Truth, assiduous and frequent questioning, the desirability of which that most perspicacious of all philosophers, Aristotle, urges upon students who are in doubt over anything. It is difficult to speak confidently about such things until they have been perused frequently. It is useful to have doubts about many things. Quoted by Urban T. Holmes, Jr., "Transition in European Education," in TwelfthCentury Europe and the Foundations of Modern Society, ed. Marshall Clagett, Gaines Post, and Robert Reynolds (Madison: University of Wisconsin Press, 1961), p. 24. 52. The double stance of the Partonopeu narrator recalls Thomas's narrator as he is simultaneously engaged and disengaged, while at the same time specifically denying any personal experience that corresponds to the lovers'. 53. Cf. the discussion of substitutability in Marie de France's Lais in Chapter 5. 54. Even in the lais, the lady does not have to be a fairy to play such a role, if we consider the case of Guildeluec in Eliduc, for example. Cf. also the active persona of a female troubadour like the Comtessa de Dia. 55. Even within the main part of the romance, we can observe a crescendo in the narrator's presence: his digressions tend to increase in frequency and length as the romance draws to a close with a growing contrast between the characters' successful union and the narrator's continued hope (Krueger, "Author's Voice," p. 128). 56. Cf. Krueger, "Textuality," pp. 66-69. 57. Anfors's portrait of the Sultan as one of the finalists in the tournament emphasizes the combination of dergie and chevalerie (w. 10005—7). Margaris's

Notes to Pages 150-153

251

learning in "all the arts" recalls Melior's education in the romance proper. Whereas her knowledge required no apology, we may note here Anfors's insistent assurance that Margaris is no less good as a knight for all his higher education. Gender substitutions inevitably introduce variations. 58. Cf. the discussion of repetition in Chapter 3 as "serious play" in the pursuit of different possibilities. 59. Dafydd Evans, "Wishfulfilment: The Social Function and Classification of Old French Romances," Court and Poet: Selected Proceedings of the Third Congress of the International Courtly Literature Society (Liverpool 1980), ed. Glyn S. Burgess, ARCA 5 (Liverpool: Francis Cairns, 1981), pp. 129-34. Cf. Northrop Frye's identification of the whole category of romance with the wishfulfilment of dream (Anatomy of Criticism: Four Essays [New York: Atheneum, 1969], p. 186). 60. Evans, "Wishfulfilment," p. 132. 61. Cf. Hilton's discussion of analyses offered by J. P. Collas, Peter Haidu, and Rupert T. Pickens, on the differences between lai and romance, which she relates to the specific variations in Partonopeu ("Convention and Innovation," pp. 46-55). While Hilton focuses on issues related to the integration of the hero into a real society with political responsibilities, in order to distinguish between lai and romance, her remarks also include some analysis of narrative structure. In that respect, she contrasts the single action of the lai with the characteristic bipartition of romance, in which one may have the impression that two lais have been combined, as Pickens suggests in Estoire, Lai, and Romance: Chretien's Erec et Enide and Cliges" Romanic Review 66 (1975): 247-62. Although some lais do indeed reflect the characteristics Hilton uses to generalize (e.g., singleness of action, contrast between the Other World and the real, focus on the individual, etc.), we can find many exceptions in Marie de France's Lais, for example, which seem to experiment with romance features as much as romance borrows and plays on characteristics more typically associated with the lai. These two genres elude our efforts to pin them down with neat definitions: we may be convinced they are different types, but we are not always able to delineate precisely their respective features, especially as they appear to enjoy renewing their own identity as frequently as they interact with each other. 62. Consider also Godefroi de Leigni's epilogue to the Charrete (discussed in the previous chapter) or the epilogue to Ille et Galeron, w. 5803-6: Ne en 1'estorie plus n'en ot, Ne plus n'en a, ne plus n'i mist Gaiters d'Arras qui s'entremist d'Eracle, ains qu'il fesist ceste uevre. (Nor was there more in the story, nor is there more, nor has Gautier d'Arras, who wrote Eracle before this work, put any more there.) Even in the varied manuscript tradition of Partonopeu, we can find an example of this topic used to authorize one version's ending after only the Anselot episode of the Continuation. Ms. L begins its epilogue after v. io42c:

252

Notes to Pages 153-157 L'istorie ici finerai Qar ge plus nen trovai E suns autres vos dit avant Ne len cruz ne tant ne quant Qar gauter mape plus nen dist. (Vol. 2, pt. i, p. 38) (I shall finish the story here, because I have found no more of it, and if another tells more, don't believe him at all, for Gautier Map said no more about it.)

The epilogue to Ipomedon acknowledges there is more to the hero's story not told in Hue de Rotelande's romance, but this is a clear allusion to the Roman de Thebes, where the eponymous model for Ipomedon is to be found. I would speculate that Hue is alluding playfully to Partonopeu as well, if we accept Fourrier's chronology and order of imitation (pp. 447-48). 63. Cf. John L. Grigsby, "The Narrator in Partonopeu deElois, le Eel Inconnu, andjoufroi de Poitiers" Romance Philology 21 (May 1968): 536-45. 64. Only ms. T offers a complete Continuation with this ending, as it appears in Gildea's edition. Within the ten extant manuscripts in French, we have three different forms of the story, with or without the promise of Continuation, with or without the Continuation itself (Gildea, ed., vol. 2, pt. 2, pp. 7-13). In Gildea's inventory of manuscripts (pp. 5-11), what is striking is the number, by far the majority, that are incomplete, whether because of lost folios or incomplete models or scribal selection: the manuscript tradition itself dramatizes the game of ending and continuing. 65. In L'ethique du changement dans le roman frangais du XIP siecle (Geneva: Droz, 1990), Denyse Delcourt analyses negative and positive paradigms of change available within the general context of the twelfth century and, more specifically, in twelfth-century romance.

Chapters 1. The incipit and explicit of B.N. nouv. aq. fr. 1104 (ms. S) identify it as a collection of "lays de Breteigne" (Richard Baum, Recherches sur les oeuwes attributes a Marie de France [Heidelberg: Carl Winter Universitatsverlag, 1968], p. 49). In his edition (Marie de France, Les Lais de Marie de France, Classiques Frangais du Moyen Age 93 [Paris: Champion, 1969], pp. xix-xx), Jean Rychner briefly identifies each of the five manuscripts containing one or more of Marie's lais and gives the customary sigla (H, S, P, C, and Q). 2. The twelfth-century Roman des Sept Sages and its popular tradition offers a very different use of a story collection within a narrative frame. Cf. Marie-Louise Oilier, "Les Lais de Marie de France ou le recueil comme forme," in La Nouvelle: genese, codification et rayonnement d'un genre medieval, ed. Michaelangelo Piccone,

Notes to Pages 157-161 253 Guiseppe Di Stefano, and Pamela Stewart (Montreal: Plato, 1983), pp. 64-79 (especially pp. 71, 78). 3. All quotations from the Lais are taken from Rychner's edition. Page numbers following translations generally refer to that of Glyn S. Burgess and Keith Busby, The Lays of Marie de France (London: Penguin Books, 1986); occasionally I have translated myself or used Robert Manning and Joan Ferrante's translation, The Lais of Mane de France (New York: E. P. Button, 1978), as indicated, in order to supply an English equivalent as literal as possible to illuminate the original language. 4. Given our inability to identify with any certainty the unique historical referent of her proper name, Marie's identity remains purely textual, defined by and defining her authority as text. Some parts of this chapter, which deal with Marie's use of proper names, have already appeared in "Strategies of Naming in Marie de France's Lais: At the Crossroads of Gender and Genre," Neophilologus 75 (1991): 31-40. 5. For an analysis of how she rewrites Ovid, see, for example, Kristine Brightenback, "The Metamorphoses and Narrative Conjointure in 'Deux Amanz,' Tonec," and cLe Laustic,'" Romanic Review 72 (1981): 1-12. Ovid's Metamorphoses furnish, moreover, a classical model for collecting together separate stories—an example not lost on Marie, as she reworks individual tales. Rychner includes in his notes the rapprochements with Brut and the Eneas made by E. Levi, E. Hoepffner, and M. Pelan. See also Earl Jeffrey Richards, "Les rapports entre le Lai de Guigemar et le Roman d'Eneas: Considerations generiques," in Le recit bref au moyen age: actes du colloque des 27, 28 et 29 awil 1979, Universite de Picardie, Centre d'Etudes Medievales, publics par les soins de Danielle Buschinger (Paris: Champion, 1980), pp. 45-56. 6. Consult Emanuel J. Mickel, Jr., Marie de France (New York: Twayne, 1974), chap. 6 on plots and sources and the Bibliography, pp. 176—78, and Glyn S. Burgess, Marie de France: An Analytic Bibliography (London: Grant & Cutler, 1977 and the Supplement i (1985), for the many articles on Marie's sources, both classical and vernacular. It is interesting to note that in Guigemar the character and place names all refer us to the Celtic sources, while other miscellaneous names introduced in comparisons, descriptions, and so on, establish a classical frame of reference (e.g., Venus, Ovid, Nature, Fortune). 7. See Douglas Kelly, "Theory of Composition in Medieval Narrative Poetry and Geoffrey of Vinsauf's PoetriaNova" Mediaeval Studies 31 (1969)'- 127-30. 8. Rupert T. Pickens, "La poetique de Marie de France d'apres les Prologues des Lais," Lettres Romanes 32 (1978): 378. 9. Florence McCullough, "Length, Recitation, and Meaning of the Lais of Marie de France," Kentucky Romance Quarterly 25 (1978): 257-68. The elasticity of narrative reflected in the changing length of lais is also common to romance writing: compare the length of a Chretien romance to that ofPartonopeu or the Perceval Continuations. 10. Jean Frappier, Etude sur La Mort le RoiArtu: roman duXIIP siecle (Paris: Minard/Geneva: Droz, 1936, rev. ed. 1961), pp. 24-146,440-55. In Frappier's view, the best way to account for both the unity and the diversity of the Lancelot-Grail

254

Notes to Pages 161-173

cycle is to hypothesize that an "architecte unique" (p. 440) plans the ensemble constituted by the Prose Lancelot^ the Queste du saint Graal^ and the Mort le roi Artu, even if he writes only the Prose Lancelot. n. E. Jane Burns, Arthurian Fictions: Rereading the Vulgate Cycle (Columbus: Ohio State University Press [for Miami University], 1985), esp. the Introduction and Chapter One. 12. On narrative blocks, see my Narrative Invention in Twelfth-Century French Romance: The Convention of Hospitality (1160-1200) (Lexington, Ky.: French Forum Pub., 1980). I borrow the term dispositio associative from Eugene Vance, "Le combat erotique chez Chretien de Troyes," Poetique 12 (1972): 544-71. On episodic structure, see, for example, Peter Haidu, "The Episode as Semiotic Module in Twelfth-Century Romance," Poetics Today 4,4 (1983): 655-81. 13. The manuscript tradition sometimes makes collections of romances that may invite similar kinds of effects—e.g., B.N. f.fr. 1450 interpolates Chretien's five romances in the middle of Wace's Brut; Chantilly, Musee Conde, 472 collects together a large number of Arthurian romances. See Lori Walters, "Le role du scribe dans ^organisation des manuscrits des romans de Chretien de Troyes," Romania 106(1985): 303-25. 14. It is probably no accident that these two operations recall the fundamental linguistic processes, as described by Roman Jakobson. See, for example, "Deux aspects du langage et deux types d'aphasie," in Essais de linguistique generale, tr. Nicolas Ruwet (Paris: Minuit, 1963), pp. 43-67. The paradigmatic and syntagmatic axes, associated with the traditional rhetorical tropes of metaphor and metonymy, continue to play a powerful role in modern critical discourse. 15. Cf. doubling in Thomas's Tristan discussed in Chapter 2. 16. Rupert T. Pickens, "Thematic Structure in Marie de France's Guigemar" Romanians (1979): 335~36. 17. Cf. Lanval, where bipartition and tripartition appear with the same plot structures, but with certain key elements reversed. See Michele Koubichkine, "A propos du lai de Lanval," MoyenAge 78 (1972): 467—88. 18. See, for example, J. A. Frey, "Linguistic and Psychological Coupling in the Lays of Marie de France," Studies in Philology 61 (1964): 12; Edgar Sienaert, Les Lais de Marie de France: du conte merveilleux a la nouvelle psychologique (Paris: Champion, 1978), p. 188; and Mickel, Marie de France, p. 136. Cf. Rupert Pickens, "Equitan: Anti-Guigemar," Romance Notes 15 (1973-74): 361-67. 19. Robert Sturges, "Texts and Readers in Marie de France's Lais" Romanic Review 71 (1980): 252-53, 263-64. 20. Cf. the verbal character of memory images in the medieval models for memory discussed by Mary Carruthers, The Book of Memory: A Study of Memory in Medieval Culture (Cambridge: Cambridge University Press, 1990), pp. 16—32. 21. W. Ann Trindade, "The Man with Two Wives—Marie de France and an Important Irish Analogue," Romance Philology 27 (1974): 466-78. 22. Cf. Howard S. Robertson, "Love and the Other World in Marie de France's Eliduc" in Essays in Honor of Louis Frances Solano, ed. Raymond J. Cormier and Urban T. Holmes (Chapel Hill: North Carolina University Studies in the Romance Languages and Literatures 92,1970), pp. 167-76. 23. The same strategy of optimism also appears in Marie's Chievrefoil^ which

Notes to Pages 173—181 255 alludes to the lovers' death in its opening frame (v. 10) and in the symbol of the honeysuckle, but emphasizes in its single episode the lovers' successful (re)union. This emphasis on successful union also characterizes the Folk Tristan d* Oxford, whereas the Charrete's "happy ending" is more problematic with respect to the future of the lovers (see the analyses above). 24. Cf. Brewster Fitz, "The Storm Episode and the Weasel Episode: Sacrificial Casuistry in Marie de France's Eliduc" Modern Language Notes 89 (1974): 542-49. 25. E.g., Robertson, "Love and the Other World." 26. Jacques de Caluwe, "La conception de Pamour dans le lai d'Eliduc de Marie de France," Moyen Age 77 (1971): 53-77, and "L'element chretien dans les Lais de Marie de France," Melanges Jeanne Lods (Paris: Collection de PEcole Normale Superieure de Jeunes Filles, 10,1978), vol. i, pp. 95—114. 27. Cf. Donald Maddox, "Triadic Structure in the Lais of Marie de France," Assays: Critical Approaches to Medieval and Renaissance Texts 3 (1985): 19-40, and "L'intersubjectivite et 1'analyse du recit medieval," in Mittelalterbilder aus neuer Perspektive: Wiirzburger Kolloquium 1985, ed. Ruhe and Behrens (Munich: Wilhelm Fink, 1986), pp. 303-13. 28. Cf. Jean-Charles Huchet's analyses of the interplay between duality and unity, as explored in individual lais, in "Nom de femme et ecriture feminine au moyen age: les Lais de Marie de France," Poetique 48 (1981): 407—30. 29. Denis Piramus, Vie seint Edmund le ra, w. 35—48, quoted in Rychner's ed., p. x n. 6. 30. Paul Zumthor, Essai depoetique medievale (Paris, Seuil, 1972), pp. 73-74, 159-60; Jean Batany, Approches du "Roman de la Rose" (Paris: Bordas, 1973), pp. 51-52; Roger Dragonetti, "Le lai narratif de Marie de France, 'pur quei fu fez, coment e dunt,'" in Litterature, histoire, linguistique: recueil d'etudes offert a Bernard Gagnebin (Lausanne: L'Age de 1'Homme, 1974), pp. 31—53; Michelle Freeman, "Marie de France's Poetics of Silence: The Implications for a Feminine Translatio" PMLA 99 (1984): 860-83. 31. According to Rychner's description (p. 194), there are no titles given as rubrics within the copy of Harley 978, but a later hand has added the title of each lai on the top of the page where it begins. On the manuscript tradition, consult Rychner's introduction, pp. xix-xxviii, and his notes on individual lais, as well as Baum, Recherches, pp. 42—58. 32. See the argument for including Guigemar's Prologue with the General Prologue, as the "commencement" announced in v. 56: Maurice Delbouille, "'El chief de cest comencement' . . . (Marie de France, Prologue de Guigemar)" in Etudes de civilisation medievale (IXe-XIIe siecles): melanges offerts a Edmond-Rene Labande (Poitiers: Centre d'Etudes Superieures de Civilisation Medievale, 1974), pp. 185-96. Rychner's notes on G, w. 1-18 (pp. 238-39), also support such a view. 33. Within the context of romance, the anonymity of Partonopeu de Bloiis author strikes us, on the contrary, as problematic, an atypical absence of name that invites analysis and interpretation as much as the unusual presence of Marie's name in the context of the lais. Cf. the oral character of the narrator's persona in Partonopeu, as suggested by the statistics on narrative interventions given below. 34. Cf. Michel Zink, "Remarques sur les conditions de 1'anonymite dans la

256

Notes to Pages 181-188

poesie lyrique franchise du Moyen Age," in Melanges de langue et litterature frangaises du moyen age et de la Renaissance offerts a Monsieur Charles Foulon (Rennes: Institut de France, Universite de Haute-Bretagne, 1980), pp. 421-27. 35. See, for example, Recherches, p. 170: Baum contrasts this lack of authorial naming with the Epilogues of the Fables and the Espurgatoire Saint Patrice', where Marie does name herself. Although Baum uses this as part of his argument to question the identity of "Marie" as author, it does not strike me as proof in that regard. We can find similar variations in a given author's habits, if we compare it to Chretien's varied use of prologues and epilogues to name himself: out of the four romances completed, Chretien's name appears once only in a prologue (Erec), once only in an epilogue (Yvain), and twice in both (Climes and Lancelot). 36. Verse 25 may either connect with the preceding verse and end the sentence (subject of est avenu), as Burgess and Busby translate it, or it may continue into the following verse as object of the verb cunterai ("For the adventure from which the lay was composed, just as it happened, I shall tell it to you"). 37. Huchet, "Nom de femme," p. 430, and Nancy Vine Durling, "The Knot, the Belt, and the Making of Guigemar" Assays: Critical Approaches to Medieval and Renaissance Texts 6 (1991): 29-53, both explore the anagrammatical linking of names. 38. In Unspeakable Sentences: Narrative and Representation in the Language of fiction (Boston: Routledge & Kegan Paul, 1982), pp. 228-29, 240-41, Ann Banfield considers and dismisses medieval examples of style indirect libre suggested by other linguists. She does not analyze any passages from Marie de France. Bernard Cerquiglini has demonstrated with numerous examples, including some from Marie de France, that free indirect discourse is not limited to modern literary usage: "Le style indirect libre et la modernite," Langages 73 (1984): 7—16. 39. See, for example, Graelent, w. 373-76, 487-98; Guingamw, w. 494-96, in Prudence M. O. Tobin, Les Lais anonymes desXIP etXIIIe siecles (Geneva: Droz, 1976). Chretien seems to make occasional use of style indirect libre: Cliges , w. 1072-76 (Alexandre Micha, ed. Classiques Francois du Moyen Age, 84 [Paris: Champion, 1968]); Tvain^ w. 4272-75 (Mario Roques, ed. Classiques Francois du Moyen Age, 89 [Paris: Champion, 1960]); cf. Charrete, w. 6713—17, with the use of mar (Mario Roques, ed., Classiques Francois du Moyen Age 86 [Paris: Champion, 1967]). 40. That is what we saw in Guigemar, w. 825-36, when the Lady retells her aventure to Guigemar. Cf. the famous passage in Chievrefoil, w. 62-78, where scholars have long debated where Tristan's words are to be located. 41. Jean Rychner, "La presence et le point de vue du narrateur dans deux recits courts: le Lai deLanval et la Chatelaine de Vergi" VoxRomanica 39 (1980): 95. 42. There is another curious example, in which we see a mix of voices, though indirect discourse does not quite give way to style indirect libre. During the storm at sea, the sailors call on Saint Nicholas and Saint Clement and "ma dame seinte Marie" (v. 823: "my lady saint Mary"). The introduction of the first person possessive pronoun adds the narrator's call to that of the participants—and acts as a discrete signature, given the identity of names between author and saint. 43. Cf. Bruckner, "Of Men and Beasts in Bisclavret" Romanic Review 82

Notes to Pages 188-191 257 (1991): 251-69, and Michelle Freeman, "Dual Natures and Subverted Glosses: Marie de France's 'Bisclavret,'" Romance Notes 25 (1985): 288-301. 44. Cf. Beate Schmolke-Hasselmann, "Tristan als Dichter—ein Beitrag zur Erforschung des lai lyrique breton" Romanische forchungen 98 (1986): 258—76. 45. Evelyn Birge Vitz has reopened the question of romance and orality in "Rethinking Old French Literature: The Orality of the Octosyllabic Couplet," Romanic Review 77 (1986): 307-21. 46. The verb mustrer (to show) frequently appears in the Marie's own lais to introduce the indirect discourse of her characters. See, for example, Eliduc, w. 57-75, where the verb mustra introduces a passage that moves from indirect to free indirect discourse. Rychner's edition mistakenly puts w. 59-74 in quotation marks: though we do not have direct discourse here, we clearly hear Eliduc speaking in those verses. 47. In order to make valid comparisons that minimize the ambiguities involved in identifying where narrative commentary begins and ends, I have counted in Tables i and 2 only those verses where the narrator's first person made his or her presence indisputable and characteristic of oral storytelling. My corpus of anonymous lais includes Guingamor, Graelent, Tyolet, Doon, Tydorel, POiselet, POmbre, I'Epervier, Amours, Lecheor. The corpus of romances includes the five romances of Chretien, Beroul's and Thomas's Tristan, and Partonopeu de Blois—that is, those principally concerned in this book. I also looked at the two Folies Tristan, by way of comparison, but did not include them in the summarizing tables. I made two sets of calculations to consider (i) the percentage of verses in Prologues and Epilogues in relation to the work's total number of verses (the unfinished Perceval and the Tristan fragments are not included here among the romance calculations), and (2) the percentage of verses with first person interventions in relation to the number of verses in the story itself (or, in the case of the eight romances examined, the first 1,184 verses—the length of Eliduc). The results can be seen in Tables i and 2 respectively. While in the first table (Prologue and Epilogue / Total), Marie and the lais have more or less the same distribution, in the second (First Person / Story), Marie stands out in the i% range, the other lais in the 2-5% range. While the romances generally occupy the low end of both categories, Partonopeu de Bloiis narrator looks a lot more like an oral storyteller in each category (Thomas and Beroul join him in the second category). On the other hand, individual lais like Eliduc, Graelent, and Guingamor take on romance characteristics in their low percentage of first person interventions within the story. Cf. the statistics on first person interventions offered in John L. Grigsby, "Narrative Voices in Chretien de Troyes: A Prolegomenon to Dissection," Romance Philology 32 (1979): 268-72; and Pierre Jonin, "LejV de Marie de France dans les Lais" Romania 103 (1982): 170-96. Cf. Rupert T. Pickens, "Historical Consciousness in Old French Narrative," French Forum 4 (1979): 168—84; and Suzanne Fleischman, "Evaluation in Narrative: The Present Tense in Medieval 'Performed Stories,'" Tale French Studies 70 (1986): 202 n. 9. 48. Carruthers (Book of Memory, pp. 8—13) stresses the continuity between written and oral contexts in relation to the memorial operations of medieval culture.

TABLE i.

Percentage of Verses in the First Person in Prologues and Epilogues Compared to Total Number of Verses in Romances and Lais

0%

.4%

.6%

.7%

1%

2%

3%

4%

5%

6%

/ 7%

8%

20%

35%

Fb Fo

Erec TV

Charr

1'Ois

Guing Tyd Fr Lanv Glides

Gmel Mil

Doon Yon Eli

2 Am G

PdeE

Bis Laiis Eq POmbre

Tyo Amours I'Eper

Chait

Chiewe

Leckeor

Summary

Percent ofw.

Fraction of total works in category

Percent of works in category

0-1%

4/5

romances (Chretien's four romances)

80%

0-3%

5/10 5/12

lais MdeF's Lais

50% 42%

4-8%

1/5 4/10 6/12

romances (only PdeE) lais MdeF's Lais

20% 40%

1/10 1/12

lais MdeF's Lais

10% 8%

above 8%

50%

TABLE 2. Percentage of First Person Inventions

0%

.1%

.3%

.4%

.5%

.6%

.7%

.8%

1%

2%

3%

5%

Eq

Fo Eli Gmel

Cliges T v Per Guing

Fr

Fb

Charr

Erec Laiis

Tyo

Beroul PdeB Chievre Chait Ton Mil Lan G Doon Lecheor Tyd Amours

Bis 2 Am I'Ombre

UOis

Thomas lJEper

Summary Fraction of total works in category

Percent of works in category

below 1%

5/8 3/10 4/12

romances IMS MdeF's Lais

63% 30% 33%

1%

2/8 3/10 6/12

romances lais MdeF's Lais

25% 30% 50%

2-5%

1/8 4/10 2/12

romances lais MdeF's Lais

12% 40% 16%

Percent ofw.

26o

Notes to Pages 191-195

49. Cf. the problem of signs and multiple points of view in romance, discussed in previous chapters. See also Renate Blumenfeld-Kosinski, "Transgressing Boundaries: The Oral and the Written in Robert de Boron," Stanford French Review 14 (1990): 143—60. 50. See G, w. 285, 313, 527, 817; Eq, w. 126, 225; Fr, w. 47, 75, 133, 174, 47i; Bis, v. 313; Lan, w. 63, 376, 445, 453,523; Deux Am, v. 42; Ton, w. 282-541; Laii, w. 65,130; Mil, w. 416, 437; Eli, w. 28, 516, 686, 858,1017,1080,1088,1092. Jeanette M. A. Beer includes a chapter on Marie de France in her Narrative Conventions of Truth in the Middle Ages (Geneva: Droz, 1981), pp. 63-72. 51. Cf. Isidore of Seville's definition of argumentum, commonly available in the Middle Ages and quoted by Beer, Narrative Conventions, p. 51: "argumenta sunt quae etsi facta non sunt, fieri tamen possunt." Cf. also the role of experience as analyzed above in Thomas's Tristan and Partonopeu de Blois. 52. For a useful bibliography, see Alfred Foulet and Karl D. Uitti, "The Prologue to the Lais of Marie de France: A Reconsideration," Romance Philology 35 (1981): 243 n. 3. 53. Cf. Robert Manning and Joan Ferrante's translation for w. 15-16: "might gloss the letter / and supply its significance from their own wisdom" (p. 28). They also offer alternate translations for w. 19-22. Rychner's note on the General Prologue (pp. 236-37) describes some of the translation problems, which will become more apparent in the discussion below. 54. Cf. Marie-Dominique Chenu, O.P., "Auctor, Actor, Autor," Bulletin du Cange: ArchivumLatinitasMediiAevi 3 (1927): 81-86. 55. Julia M. H. Smith, "Oral and Written: Saints, Miracles, and Relics in Brittany, c. 850-1250," Speculum 65 (1990), p. 308. When he attacks the cults of Breton saints, Guibert de Nogent "equates the oral, the popular, the inauthentic and the disreputable" (Brian Stock, The Implications of Literacy: Written Language and Models of Interpretation in the Eleventh and Twelfth Centuries [Princeton, N. J.: Princeton University Press, 1983], p. 250, quoted by Smith, "Oral and Written," p. 309). 56. Herbert Grundmann, "Litteratus-illitteratus: der Wandel einer Bildungsroman von Altertum zum Mittelalter," Archivfiir Kulturgeschichte 40 (1958): 1-65; Franz Bauml; "Varieties and Consequences of Medieval Literacy and Illiteracy," Speculum 55 (1980): 237-65, esp. pp. 237-49; and D. H. Green, "Orality and Reading: The State of Research in Medieval Studies," Speculum 65 (1990): 267-80, esp. pp. 274-75. See also the special issue of New Literary History 16, no. i (Autumn 1984), which brings together a collection of articles on "Oral and Written Traditions in the Middle Ages"—a useful starting point for further work in this area. 57. Smith, "Oral and Written," p. 311. She refers (p. 310) to Jacques Le Goff's "Clerical Culture and Folklore Traditions in Merovingian Civilization," in Time, Work and Culture in the Middle Ages, tr. Arthur Goldhammer (Chicago: University of Chicago Press, 1980), pp. 153—58. 58. Translations of v. 27 generally do not retain the association with the hard labor of birth implied in the choice of the verb delivrer, combined with grant dolur. Burgess and Busby give "rid oneself of suffering" (p. 41); Manning and Ferrante translate v. 27 as "free of great sorrow" (p. 28). Sturges ("Texts and Readers,"

Notes to Pages 195—200

261

p. 255) associates Marie's sleepless nights with those of the lovers she describes later. But we can also be reminded of a mother's sleepless nights. Cf. the attention for birth and the care of children that appears in Fresne and Milun, for example: these are issues not generally introduced in courtly literature. Another woman writer in the later Middle Ages, Christine de Pizan, uses birth as a metaphor for textual production in her Avision^ pp. 163-64. I would like to thank Douglas Kelly for bringing this passage to my attention. 59. See, for example, two articles by Freeman, "Dual Nature" and "Marie de France's Poetics of Silence." 60. Tony Hunt, "Glossing Marie de France," Romanische Forschungen 86 (i974): 396-418. 61. Ibid., pp. 408-10. 62. Cf. the debate on w. 9—22 of the General Prologue and Brewster Fitz's comments on the controversy that surrounds the letters Tristan carved on the baton in Chievrefoil: "Desire and Interpretation: Marie de France's Chiewefoil" Tale French Studies 58 (1979): 182-89. For further analysis of the Lais*s self-reflexive character, see Sturges, "Texts and Readers." 63. Cf. Joseph J. Duggan, "Ambiguity in Twelfth-Century French and Provenc,al Literature: A Problem or a Value?" in Jean Mismhi Memorial Volume: Studies in Medieval Literature, ed. Hans R. Runte, Henri Niedzielski, and William L. Hendrickson (Columbia, S.C.: French Literature Publications, 1977), pp. 136—49; Eugene Vinaver, "Landmarks in Arthurian Romance," in The Expansion and Transformations of Courtly Literature (Athens: University of Georgia Press, 1980), pp. 20-22. 64. Many scholars have commented on the way Marie's collection seems to grow by a series of experiments designed to differentiate the same set of materials. See, for example, S. Foster Damon, "Marie de France: Psychologist of Courtly Love," PMLA 44 (1929): 968-96; Frey "Linguistic and Psychological Coupling," pp. 3-18; Frederick Hodgson, "Alienation and the Otherworld in Lanval, Yonec, and Guigemar" Comitatus 5 (1974): 19—31, Emanuel Mickel, Jr., "A Reconsideration of the Lais of Marie de France," Speculum 46 (1971): 39-65; Pickens, "Equitan," pp. 361-67. 65. Cf the twelfth-century interest in seeing man as a microcosm of the macrocosm, described by Marie-Dominique Chenu, O.P., Nature, Man, and Society in the Twelfth Century: Essays on New Theological Perspectives in the Latin West, ed. and tr. Jerome Taylor and Lester K. Little (Chicago: University of Chicago Press, 1968), pp. 24-37. On the other hand, the twelfth century and its various renaissances showed renewed interest in the literal or historical level of texts and human experience. See Beryl Smalley on Biblical exegesis of the Victorines in The Study of the Bible in the Middle Ages (Oxford: Blackwell, 1952), and Chenu (Nature, Man, and Society, passim). 66. Cf. Roland Barthes, "From Work to Text," in Textual Strategies: Perspectives in Post-Structuralist Criticism^ ed. Josue V. Harari (Ithaca, N.Y.: Cornell University Press, 1979), pp. 73-81. See also Peter Haidu, "The Semiotization of Death: Open Text or Closed?" Style 20 (1986): 220-51, especially pp. 220-21. 67. Evelyn Birge Vitz argues that Marie's stylistic treatment can create a sat-

262

Notes to Pages 200-208

isfactory ending, even where there is no definitive point of equilibrium: "The Lais of Marie de France: Narrative Grammar and the Literary Text," Romanic Review 74 (1983): 383—404; rpt. in Medieval Narrative and Modern Narratology: Subjects and Objects of Desire (New York and London: New York University Press, 1989), pp. 149-7568. Menard, Les Lais de Marie de France: contes d'amour et d'aventures du moyen age (Paris: Presses Universitaires de France, 1979), pp. 92-93; Zumthor, Essai^ p. 391. Jean-Charles Payen's definition of the lai insists on the importance of its plot, its closure and self-sufficiency, which he contrasts with comparable short works that are not identified as lais because they remain open by inviting debate ("Le lai narratif," Tyfologie des sources du moyen age occidental Fasc. 13 [Turnhout: Brepols, 1975], PP- 46-47). 69. Maddox, "Triadic Structure" and."L'intersubjectivite." 70. Cf. the nexus of truth and meaning in Chretien's Charrete, articulated around the figure of Marie de Champagne. 71. Philippe Hamon, "Clausules," Poetique 24 (1975): 502. See also his discussion of the link between "clausules" and genres, pp. 500—501. Cf. Barbara Herrnstein Smith, Poetic Closure: A Study of How Poems End (Chicago: University of Chicago Press, 1968), pp. 63-67,182-88, on the use of stylistic devices for establishing closure. 72. Frank Kermode, The Sense of an Ending: Studies in the Theory of Fiction (London and New York: Oxford University Press, 1966, 1967; rpt. 1970). See especially the general reflections, pp. 3-8. 73. Cf. Pickens, "La poetique de Marie de France," pp. 367-84. 74. Ernst Robert Curtius, European Literature and the Latin Middle Ages, tr. Willard R. Trask (New York: Harper Torchbooks, 1953), PP- 89-91. Curtius states that the most natural reason given for ending a poem in the Middle Ages is weariness (p. 90). We might compare this with Marie's description in the Prologue: the sleepless nights needed to accomplish hcrgrevose owe (w. 24-42). 75. Cf. H. Wayne Storey's review of The Concepts of Closure, ed. David Hult, in Romance Philology 44 (1990): 91-99, where he cites Bakhtin's The Dialogic Imagination (pp. 68-69) on the medieval handling of quotation and parody: "the boundary lines between someone else's speech and one's own speech were flexible, ambiguous, often deliberately distorted and confused" (quoted p. 97). 76. Cf. Frank Kermode, The Genesis of Secrecy: On the Interpretation of Narrative (Cambridge, Mass.: Harvard University Press, 1979). Some authors have felt compelled to rewrite Marie's Eliduc: possibly Gautier d'Arras (unless his Ille et Galeron is based on a version common to both authors), certainly John Fowles in The Ebony Tower.

Conclusion i. Hans Robert Jauss, "Theory of Genres and Medieval Literature," in Toward an Aesthetic of Reception, tr. Timothy Bahti, Theory and Literature 2 (Min-

Notes to Pages 208-209 263 neapolis: University of Minnesota Press, 1982), pp. 82-87. See also "The Alterity and Modernity of Medieval Literature," New Literary History 10 (1979): 208—22, 228—29, where Jauss discusses "The Little Genres of the Exemplary as a Literary Communications System" and gives an appendix with an overview of the constituent elements that characterize nine "little genres" in the Middle Ages. Of these nine, the tale is of particular interest here as a "substratum to the lai and the Arthurian romance" (p. 229). For a mise en question of the notion of dominant, see Kathryn Gravdal, "En Fabliaus Doit Fables Avoir: Medieval Genre Theory and the 'Problem' Fabliau," given at a symposium on "Intergenres: Intergeneric Perspectives on Medieval French Literature," held on Nov. 9-10, 1990, at the University of Massachusetts, Amherst. See also Nico van den Boogaard, "Vocabulaire de lais, vocabulaire de fabliaux," Rapports-Het Fmnse Boek 49 (1979): 97-106; Peter Haidu, "Fragments in Search of Totalization: Roland and the Historical Text" in Modernite au nwyen age: le defi du passe, ed. Brigitte Gazelles and Charles Mela (Geneva: Droz, 1990), pp. 80—81. 2. On the jongleur's persona, see, for example, Erich Auerbach, "Roland Against Ganelon," in Mimesis: The Representation of Reality in Western Literature, tr. Willard R. Trask (Princeton, N.J.: Princeton University Press, 1953, 1968), pp. 96-122; Eugene Vance, "Roland and the Poetics of Memory," in Textual Strategies: Perspectives in Post-Structuralist Criticism, ed. Josue V. Harari (Ithaca, N.Y.: Cornell University Press, 1979), pp. 374-403; and Franz Bauml, "Varieties and Consequences of Medieval Literacy and Illiteracy," Speculum 55 (1980): 249—50, esp. n. 39. On the romancer's narrative persona, I hardly need emphasize how seminal Peter Haidu's formulation of "aesthetic distance" has been for subsequent work on Chretien in particular and romance in general. In Aesthetic Distance in Chretien de Troyes: Irony and Comedy in "Cliges" and "Perceval" (Geneva: Droz, 1968), Haidu describes Chretien's irony as a way "to range the reader with the author as against the characters . . . It [irony] defined one's relationship to the text, a relationship both sympathetic and reserved; this was not a story to become 'involved' in, these were not characters with whom a reader might identify himself; one was to keep one's aesthetic distance" (pp. 9-10). 3. Pierre Gallais, "Recherches sur les mentalites des romanciers fran^ais du moyen age," Cahiers de civilisation medievale 7 (1964): 479-93, and 13 (1970): 333-47; Tony Hunt, "The Rhetorical Background to the Arthurian Prologue: Tradition and the Old French Vernacular Prologues," Forum for Modern Language Studies 6 (1970): 1-28; and Pierre-Yves Badel, "Rhetorique et polemique dans les prologues de romans au moyen age," Litterature 20 (1975): 81—94. 4. Cf. Zumthor's concept of the "registre": Essai depoetique medievale (Paris: Seuil, 1972), pp. 231-32. 5. "Discours du recit," Figures III (Paris: Seuil, 1972), pp. 238-39. 6. Cf. Douglas Kelly, The Art of Medieval French Romance (Madison: University of Wisconsin Press, 1992), pp. 4-5. In "The Performing Self in TwelfthCentury Culture," Viator 9 (1978): 194-212, Martin Stevens discusses his preference for the term "perfoming self" to that of persona, since the latter may suggest a false image of total independence between author and narrator that is as misleading as the opposite tendency to see them as one and the same (pp. 193—94). He

264

Notes to Pages 209-212

gives numerous examples in which twelfth-century artists represent themselves in manuscripts and carvings, and thus set up an interplay between art and reality such that "the boundaries between the worlds of myth and reality (on several planes) are obscured" (p. 206). See a similar discussion of the interpenetration of "realities" inside and outside narrative and pictorial works of this period in Bauml, "Varieties and Consequences," pp. 258-62. 7. Cf. how chevalerie and dergie are associated in the well-known elaboration of the translatio studii et imperil topos in the Climes Prologue (w. 28—42). 8. Cf. Peter Haidu, "The Hermit's Pottage: Deconstruction and History in Yvain" in The Sower and His Seed: Essays on Chretien de Troyes, ed. Rupert T. Pickens [Lexington, Ky.: French Forum Pub., 1983], pp. 127-45; Haidu, "Text and History: The Semiosos of Twelfth-Century Lyric as Sociohistorical Phenomenon [Chretien de Troyes: 'D'Amors qui m'a tolu']," Semiotica 33 [1981]: 1—62); Eugene Vance, "Love's Concordances: The Poetics of Desire and the Joy of the Text," Diacritics (Spring 1974): 40-52; "Signs of the City: Medieval Poetry as Detour," New Literary History 4 (1973): 557-74; Marie-Luce Chenerie, Le Chevalier errant dans les romans arthuriens en vers desXIP etXIIP siecles (Geneva: Droz, 1986). 9. Rainer Warning, "Pour une pragmatique du discours fictionnel," Poetique 38 (1979): 221-37; and Bauml, "Varieties and Consequences," p. 252. 10. Gautier d'Arras, Ille et Galeron (v. 5809). On literary patrons, see Maurice Keen, Chivalry (New Haven, Conn, and London: Yale University Press, 1984), pp. 84-85; John F. Benton, "The Court of Champagne as a Literary Center," Speculum 36 (1961): 551-91; and Robert Hanning, "The Audience as Co-Creator of the First Chivalric Romances," Yearbook of English Studies n (1981): 1-28. n. Keen, Chivalry, pp. 29—30. Many studies have emphasized the changes still taking place in the character and definitions of nobility and knighthood in the twelfth century. See Marc Bloch, La societefeodale (Paris: Albin Michel, 1939,1968), pp. 271-94, 395-478; Joseph Strayer, "The Development of Feudal Institutions," in Twelfth-Century Europe and the Foundations of Modern Societyp, ed. Marshall Clagett, Gaines Post, and Robert Reynolds (Madison: University of Wisconsin Press, 1961), pp. 77—88; Georges Duby, "Dans la France du nordouest au XIP siecle: les 'jeunes' dans la societe aristocratique," Annales: Economies, Societes, Civilisations 19 (1964): 835-46; and Keen, Chivalry', pp. 18-23. Erich Kohler's sociological interpretation sees romance as a reflection of contemporary efforts to integrate the higher and lower nobility: UAventure chevaleresque: ideal et realite dans le roman courtois, tr. Elaine Kaufholz (Paris, Gallimard, 1956,1974). 12. Peter Haidu, "Le sens historique du phenomene stylistique: la semiose dissociative chez Chretien de Troyes," Europe 642 (1982): 36-47. For further comments on the link between dergie and chivalerie, see Kohler, UAventure, pp. 44-76, and Keen, Chivalry, pp. 102-41. 13. Jean Frappier has insisted on "courtly love" in lyric and romance as an art involving considerable savoir vivre and savoir faire: "Vues sur les conceptions courtoises dans les litteratures d'oc et d'o'il au XIP siecle," Cahiers de Civilisation Medievale 2 (1959): 135-56. Robert Guiette has similarly pictured the courtly public as requiring an apprenticeship, an initiation into the conventions of its literary forms: "Questions de litterature," Romania Gandensia 8 (1960): 17.

Notes to Pages 212-222 265 14. See the discussion in Robert Guiette's "'Li conte de Bretaigne sont si vain et plaisant,'" Romania 88 (1967): 1-12. Bodel associates the tales of Brittany with vain pleasure, those of Rome with wisdom, and those of France with truth. 15. Cf. Kohler's chapter on Arthur's role in romance (UAventure, pp. 7-43). See Robert Hanning, "Arthurian Evangelists: The Language of Truth in Thirteenth-Century French Prose Romances," Philological Quarterly 64 (1985): 347-65. 16. Cf. Roland Barthes, "From Work to Text," in Textual Strategies, ed. Harari, pp. 73-81 . 17. See, for example, Marcia Colish, The Mirror of Language: A Study in the Medieval Theory of Knowledge (New Haven, Conn.: Yale University Press, 1968); Eugene Vance, "Mervelous Signals: Poetics, Sign Theory, and Politics in Chaucer's Troilus" New Literary History 10 (1979): 293—337; Jean Jolivet, Arts du langage et theologie chezAbelard (Paris: Vrin, 1969). 18. Paul Zumthor, "Le Texte fragment," Langue Frangaise 40 (1978): 75-82. 19. Peter Haidu, "Fragments," pp. 89-99. 20. Douglas Kelly, "Matiere and Genera dicendi in Medieval Romance," Tale French Studies 51(1974): 147-59 and "Topical Invention in Medieval French Literature," in Medieval Eloquence: Studies in the Theory and Practice of Medieval Rhetoric, ed. James J. Murphy (Berkeley: University of California Press, 1978), pp. 231-51. 21. Jean-Charles Payen, "Lancelot contre Tristan, ou la conjuration d'un mythe subversif," in Melanges de langue et de litterature medievales ojferts a Pierre Le Gentil, eds. Jean Dufournet and Daniel Poirion (Paris: SEDES, 1973), Pp- 617-33. Cf. Haidu, "Fragments," p. 77. 22. Emmanuele Baumgartner, Tristan et Iseut: de la legende aux recits en vers (Paris: Presses Universitaires de France, 1987), p. 76. 23. On "thematic analogues," see Norris Lacy, "Thematic Analogues in Erec" Esprit Createur 9 (1969): 267-74, and The Craft of Chretien de Troyes: An Essay on Narrative Art (Leiden: Brill, 1980). On "narrative blocks" in romance, see Bruckner, Narrative Invention in Twelfth-Century French Romance: The Convention of Hospitality (1160-1200) (Lexington, Ky.: French Forum Pub., 1980). 24. It is no doubt in such a way that we can understand Derrida's boutade that there is no "hors-texte." De lagrammatologie (Paris: Minuit, 1967), p. 227. 25. Jolivet, Arts du langage (Paris: Vrin, 1969), p. 76 (italics in the original). 26. It may recall as well modern theories of the sign that stem from Peirce's triangular linking of representamen (sign), object, and interpretant, where the latter may stand for the meaning of the sign in the broadest sense or, more strictly, the paradigmatic relationship between a sign and another sign: "1'interpretant est done toujours aussi signe, qui aura son interpretant, etc." (Oswald Ducrot and Tzvetan Todorov, Dictionnaire encyclopedique des sciences du langage [Paris: Seuil, 1972], p. 114, under "Semiotique"). 27. Cf. the links Winthrop Wetherbee explores between the writings of the Chartrians, like Bernardus Sylvestris and Alain de Lille, and vernacular romance: Platonism and Poetry in the Twelfth Century: The Literary Influence of the School of Chartres (Princeton, N.J.: Princeton University Press, 1972). 28. Cf. the romance process of recognition analyzed here and the rather different process analyzed by Augustine (who bases his own system on Plotinus), as

266

Notes to Pages 222-223

the three stages that lead to the recognition of God. These systems are compared by David Knowles in The Evolution of Medieval Thought (New York: Random House, 1962), pp. 40-45. 29. Hugh of Saint Victor develops that idea when he writes: "This entire perceptible world is a book written by the finger of God, that is created by divine power, and individual creatures are as figures within i t . . . instituted by divine authority to make manifest the invisible things of God" (quoted by Stevens, "Performing Self," p. 209). 30. Cf. Colish, Mirror of Language, especially the chapter on Augustine, pp. 8-81. 31. The narrator's allusion to Adam and Eve reminds us that Genesis is notoriously filled with such models as well, as God works through human frailties to realize salvation history. 32. Vv. 3041-59, 3316-17, 4222-23, 6941-4433. See Erec et Enide^ w. 2016 and 5208; Charrete^ w. 4674-81. In La Chastelaine de Vergi^ w. 435-47, the problem of describing such joy is linked to the lived experience of the reader: without direct experience of love the reader cannot understand what he hears about it. 34. Cf. Hanning, "Arthurian Evangelists": the changes from twelfth- to thirteenth-century romance engage precisely whether or not fictionality itself can be foregrounded. Bauml analyzes fiction in romance as a third category beside truth and falsehood ("Varieties and Consequences," pp. 255-62).

Bibliography of Works Cited

PRIMARY TEXTS Aebischer, Paul. Le Mystere d'Adam (Ordo representacionis Ade). Textes Litteraires Francois 99. Paris and Geneva: Droz, 1964. Andreas Capellanus. The Art of Courtly Love. Tr. John Jay Parry. New York: Norton, 1941,1969. Arnold, I. D. O., and M. M. Pelan, eds. Lapartie arthurienne du Roman de Brut. Paris: Klincksieck, 1962. Aucassin etNicolette. Ed. and tr. Jean Dufournet. Paris: Garnier-Flammarion, 1973. Augustine. De Doctrina Christiana. Tr. D. W. Robertson, Jr. New York: Library of Liberal Arts, 1958. Benoit de Sainte-Maure. Roman de Troie. Ed. L. Constans. 6 vols. Societe des Anciens Textes Francois. Paris: Champion, 1904-12. Beroul. Le Roman de Tristan: poeme duXII* siede. Ed. Ernest Muret (4th ed. L. M. Defourques). Classiques Francois du Moyen Age 12. Paris: Champion, 1970. Boccaccio, Giovanni. The Decameron. Harmondsworth: Penguin, 1972,1976. Bogin, Meg, ed. and tr. The Women Troubadours. New York and London: Norton, 1976. La Chanson de Roland. Ed. and tr. Joseph Bedier. Paris: Piazza, 1964. La Chastelaine de Vergi. Ed. Gaston Raynaud (4th ed. Lucien Foulet). Classiques Frangais du Moyen Age, i. Paris: Champion, 1967. Le Charroi de Nimes: chanson de geste du XIIe siecle. Ed. J.-L. Perrier. Classiques Francois du Moyen Age 66. Paris: Champion, 1968. Chretien de Troyes. Lancelot or, The Knight of the Cart (Le Chevalier de la Charrete). Ed. and tr. William W. Kibler. Garland Library of Medieval Literature, series A, I. New York: Garland, 1981. —————. Le Roman de Perceval ou le Conte du Graal. Ed. William Roach. Textes Litteraires Francois 71. Geneva: Droz; Paris: Minard, 1959. —————. Les Romans de Chretien de Troyes: I. Erec et Enide. Ed. Mario Roques. Classiques Francois du Moyen Age 80. Paris: Champion, 1970. —————. Les Romans de Chretien de Troyes: II. Cliges. Ed. Alexandre Micha. Classiques Frangais du Moyen Age 84. Paris, Champion, 1968,1969. —————. Les Romans de Chretien de Troyes: III. Le Chevalier de la Charrete. Ed. Mario Roques. Classiques Frangais du Moyen Age 86. Paris: Champion, 1967. —————. Les Romans de Chretien de Troyes: IV. Le Chevalier au Lion (Tvain). Ed. Mario Roques. Classiques Francois du Moyen Age 89. Paris: Champion, 1960.

268

Bibliography

Christian von Troyes. Samtliche erhaltene Werke: IV. Der Karrenritter und das Wilhelmsleben. Ed. Wendelin Foerster. Halle: Niemeyer, 1899. Christine de Pizan. Pizan's Lavision-Christine: Introduction and Text. Ed. Sister Mary Louis Towner. Washington, D.C.: Catholic University of America, 1932; rpt. New York: AMS Press, 1969. Crescini, V. "Azalais d'Altier." Zeitschrift fur Romanische Philologie 14 (1890): 128-32. Denis Piramus. Vie seint Edmund le rei: poeme an0lo-normand du XIIe siecle. Ed. Hilding Kjellman. Goteborg: Elanders Boktryckeri, 1935. Les deux poemes de la Folie Tristan. Ed. Joseph Bedier. Paris: Firmin-Didot, 1907. Floire et Blancheflor. Ed. Margaret M. Pelan. Publications de la Faculte des Lettres de 1'Universite de Strasbourg, Textes d'Etudes 7. Paris: Belles Lettres, 1956. Jean Bodel. Chanson des Saisnes: Edition critique. Ed. Annette Brasseur. Textes Litteraires Fran^ais, 369. Geneva: Droz, 1989. La Folie Tristan de Berne. Ed. Ernest Hoepffner. Paris: Belles Lettres, 1949. La Folie Tristan d'Oxford. Ed. Ernest Hoepffner. Paris: Belles Lettres, 1938. Fowles, John. The Ebony Tower. New York: New American Library, 1974,1975. Gautier d'Arras. Ille et Galeron. Ed. Frederick A. G. Cowper. Paris: Picard, 1956. Geoffroi de Vinsauf. Poetria Nova of Geoffrey of Vinsauf. Tr. Margaret F. Nims. Toronto: Pontifical Institute of Mediaeval Studies, 1967. Gottfried von Strassburg. Tristan. Tr. A. T. Hatto. New York, Penguin Books, 1975Guillaume de Lorris and Jean de Meun. Roman de la Rose. Ed. Felix Lecoy. 3 vols. Classiques Francois du Moyen Age 92, 95, 98. Paris: Champion, 1968, 1966, 1970. Hue de Rotelande. Ipomedon: poeme de Hue de Rotelande (fin du XIIe siecle). Ed. A. J. Holden. Paris: Klincksieck, 1979Marguerite de Navarre. L'Heptameron. In Conteurs frangais du XVP siecle, pp. 7011131. Ed. Pierre Jourda. Bibliotheque de la Pleiade. Paris: Gallimard, 1965. Marie de France. Les Lais de Marie de France. Ed. Jean Rychner. Classiques Francois du Moyen Age 93. Paris: Champion, 1969. —————. The Lais of Marie de France. Tr. Robert Hanning and Joan Ferrante. New York: E. P. Button, 1978. —————. The Lays of Marie de France. Tr. Glyn S. Burgess and Keith Busby. London: Penguin Books, 1986. —————. Marie de France: Fables. Ed. and tr. Harriet Spiegel. Toronto: University of Toronto Press, 1987. Lancelot: roman en prose duXIIP siecle. Ed. Alexandre Micha. 9 vols. Textes Litteraires Frangais, 288, 307, 247, 249, 262, 278, 283, 286, 315. Paris and Geneva: Droz, 1980, 1982, 1978, 1978, 1979, 1979, 1980, 1980, 1983. La Mort le roi Artu: roman du XIIe siecle. Ed. Jean Frappier. Textes Litteraires Francois 58. Geneva: Droz; Paris: Minard, 1964. Ovid. The An of Love, and Other Poems. Ed. and tr. J. H. Mozley. Loeb Classical Library. London: William Heinemann/Cambridge, Mass.: Harvard University Press, 1929,1969. —————. Metamorphoses. Ed. and tr. Frank Justus Miller. 2 vols. Loeb Classical Li-

Bibliography

269

brary. London: William Heinemann/New York: G. P. Putnam's Sons, 1916. Partonopeu de Blois: A French Romance of the Twelfth Century. Ed. Joseph Gildea, O.S.A. 2 vols. Villanova, Pa.: Villanova University Press, 1967. La Queste del saint graal: roman du XIIP siecle. Ed. Albert Pauphilet. Classiques Francois du Moyen Age 33. Paris: Champion, 1967. Renaut de Beaujeu. Le Bel Inconnu: roman d'aventures. Classiques Franqais du Moyen Age 38. Paris: Champion, 1967. Le Roman deRenart, ed. Mario Roques. 6 vols. Classiques Francois du Moyen Age 78, 79, 81, 85, 88, 90. Paris: Champion, 1971,1969,1955,1958,1960,1963. Roman de Thebes. Ed. Guy Raynaud de Lage. 2 vols. Classiques Francois du Moyen Age 94 & 96. Paris: Champion, 1969. Roman de Tristan. Ed. Joseph Bedier. 2 vols. Paris: Firmin-Didot, 1902,1905. Roman des Sept Sages de Rome: A Critical Edition of the Two Verse Redactions of a Twelfth-Century Romance. Ed. Mary B. Speer. Edward C. Armstrong Monographs on Medieval Literature 4. Lexington, Ky.: French Forum, Pub., 1989. Tristan etlseut. Ed. & tr. Jean-Charles Payen. Paris: Gamier, 1974. Thomas. Les fragments du Roman de Tristan: poeme du XIP siecle. Ed. Bartina H. Wind. Textes Litteraires Francois 92. Geneva: Droz, 1960. Tobin, Prudence M. O., ed. LesLais anonymes desXIP etXIIP siecles. Geneva: Droz, 1976. Wace. Le Roman de Brut. Ed. I. Arnold. 2 vols. Societe des Anciens Textes Francois. Paris: Champion, 1938—40. SECONDARY TEXTS Adams, Alison. "Godefroi de Leigni's Continuation of Lancelot" Forum for Modern Language Studies 9 (1974): 295—99. Auerbach, Erich. "Roland Against Ganelon." In Mimesis: The Representation of Reality in Western Literature. Tr. Willard R. Trask. Princeton, N.J.: Princeton University Press, 1953,1968. 96-122. Badel, Pierre-Yves. "Rhetorique et polemique dans les prologues de romans au moyen age." Litterature 29 (1975): 81—94. Banfield, Ann. Unspeakable Sentences: Narrative and Representation in the Language of Fiction. Boston: Routledge & Kegan Paul, 1982. Baron, F. Xavier. "Love in Chretien's Charrete: Reversed Values and Isolation." Modern Language Quarterly 34 (1973): 377-83. Barteau, Francpise. Les romans de Tristan et Tseut: introduction a une lecture plurielle. Paris: Larousse, 1972. Barthes, Roland. "From Work to Text." In Textual Strategies: Perspectives in PostStructuralist Criticism, ed. Josue V. Harari. Ithaca, N.Y.: Cornell University Press, 1979. 73-81. —————. LePlaisir du texte. Paris: Seuil, 1973. Batany, Jean. Approches du "Roman de la Rose." Paris: Bordas, 1973. Baum, Richard. Recherches sur les ceuwes attributes a Marie de France. Heidelberg: Carl Winter Universitatsverlag, 1968.

270

Bibliography

Baumgartner, Emmanuele. Tristan et Iseut: de la legende au recit en vers. Paris: Presses Universitaires de France, 1987. Baumgartner, Emmanuele and R. L. Wagner. "'As enveisiez e as purvers': commentaire sur les vers 3125-3129 du Roman de Tristan de Thomas." Romania 88 (1967): 527-37. Bauml, Franz. "Varieties and Consequences of Medieval Literacy and Illiteracy." Speculum 55 (1980): 237-65. Bedier, Joseph. "Les Lais de Marie de France." Revue des Deux Mondes 107 (1891): 835-63. Beer, Jeanette M. A. Narrative Conventions of Truth in the Middle Ages. Geneva: Droz, 1981. Benson, Robert L. and Giles Constable with Carol D. Lanham, eds. Renaissance and Renewal in the Twelfth Century. Cambridge, Mass.: Harvard University Press, 1982. Benton, John F. "The Court of Champagne as a Literary Center." Speculum 36 (1961): 551-91. Berconici-Huard, Carole. "Partonopeus de Blois et la couleur byzantine." In Images et signes de POrient dans ^Occident medieval: litterature et civilisation. Aix-enProvence: CUERMA, Universite de Provence, 1982.179-96. Blakeslee, Merritt. Love's Masks: Identity, Intertextuality and Meaning in the Old French Tristan Poems. Cambridge: D.S. Brewer, 1989. Bloch, Marc. Lasocietefeodale. Paris: Albin Michel, 1939,1968. Bloch, R. Howard. Etymologies and Genealogies: A Literary Anthropology of the French Middle Ages. Chicago and London: University of Chicago Press, 1983. —————. "Signs and the Times." University Publishing 9 (1980): 20—22. —————. "Tristan, the Myth of the State and the Language of the Self." Tale French Studies 51 (1974): 61—81. Blumenfeld-Kosinski, Renate. "Remarques sur songelmensonge." Romania 102 (1980): 385-90. —————. "Transgressing Boundaries: The Oral and the Written in Robert de Boron." Stanford French Review 14 (1990): 143—60. Bozoky, E. "Roman medieval et conte populaire: le Chateau desert." Ethnologic franfoise 4 (1974): 349-56. Brand, Wolfgang. Chretien de Troyes: zur Dichtungstechnik seiner Romane. Freiburger Schriften zur romanischen Philologie, 19. Munich: Wilhelm Fink, 1972. Brewer, Derek. "The Presentation of the Character of Lancelot: Chretien to Malory." Arthurian Literature 3 (1983): 26-52. Brightenback, Kristine. "The Metamorphoses and Narrative Conjointure in 'Deux Amanz,' 'Yonec," and 'Le Laiistic.'" Romanic Review 72 (1981): 1—12. Brooks, Peter. "Freud's Masterplot: Questions of Narrative." Yale French Studies 557 56 (1977): 280-300. Brown, Thomas H. "The Relationship Between Partonopeus de Blois and the Cupid and Psyche Tradition." Erigham Young University Studies 5 (1964): 193-202. Brownlee, Kevin. "Transformations of the Charrete: Godefroi de Leigni Rewrites Chretien de Troyes." Stanford French Review 14 (1990): 161—78. Brownlee, Kevin and Marina Scordilis Brownlee, eds. Romance: Generic Transfer-

Bibliography

271

motion from Chretien de Troyes to Cervantes. Hanover, N.H.: University Press of New England, 1985. Bruckner, Matilda Tomaryn. "Le Chevalier de la Charrete (Lancelot)." In The Romances of Chretien de Troyes: A Symposium, ed. Douglas Kelly. Edward C. Armstrong Monographs on Medieval Literature 3. Lexington, KY: French Forum Pub., 1986. 132-81. —————. "Essential and Gratuitous Inventions: Thomas's Tristan and Chretien's Lancelot" Actes du i4e Congres International Arthurien. Rennes: Presses Universitaires de Rennes, 1986.120—41. —————. "The Folie Tristan d'Oxford: Speaking Voice, Written Text." Tristania 7 (Autumn i98i-Spring 1982): 47-59. —————. "An Interpreter's Dilemma: Why Are There So Many Interpretations of Chretien's Chevalier de la Charrete^ Romance Philology 40 (1986): 159-80. —————. "Intertextuality." In The Legacy of Chretien de Troyes, ed. Norris J. Lacy, Douglas Kelly, and Keith Busby. 2 vols. Amsterdam: Rodopi NV, 1987,1988. Vol. I, 224-65. —————. Narrative Invention in Twelfth-Century French Romance: The Convention of Hospitality (1160-1200). Lexington, Ky: French Forum Pub., 1980. —————. "Of Men and Beasts in Bisclavret." Romanic Review 82 (1991): 251-69. —————. "Repetition and Variation in Twelfth-Century French Romance." In The Expansion and Transformation of Courtly Literature, ed. N. B. Smith and J. T Snow, pp. 95-114. Athens: University of Georgia Press, 1980. —————. "The Representation of the Lovers' Death: Thomas' Tristan as Open Text." Tristania 9 (Autumn 1983-Spring 1984): 49-61. —————. Rev. of Love and Marriage in Chretien de Troyes, by Peter S. Noble. Cardiff: University of Wales Press, 1982. Romance Philology 39 (1986): 512. —————. "Strategies of Naming in Marie de France's Lais'. At the Crossroads of Gender and Genre." Neophilologus 75 (1991): 31-40. Burgess, Glyn S. Marie de France: An Analytic Bibliography. London: Grant & Cutler, 1977. —————. Marie de France: An Analytic Bibliography. Supplement i. London: Grant & Cutler, 1985. Burns, E. Jane. Arthurian Fictions: Rereading the Vulgate Cycle. Columbus: Ohio State University Press for Miami University, 1985. Busby, Keith. "Cristal et Clarie\ A Novel Romance?" In Convention and Innovation in Literature, eds. Theo D'haen, Rainer Grubel, and Helmut Lethen. Amsterdam and Philadelphia: John Benjamins, 1989. 77-103. Caluwe, Jacques de. "La conception de Pamour dans le lai d'Eliduc de Marie de France." MoyenAge 77 (1971): 53-77—————. "L'element chretien dans les Lais de Marie de France." Melanges Jeanne Lods. Paris: Collection de 1'Ecole Normale Superieure de Jeunes Filles, 10, 1978. Vol. I, 95-114. Carruthers, Mary. The Book of Memory: A Study of Memory in Medieval Culture. Cambridge: Cambridge University Press, 1990. Gazelles, Brigitte. "Alexis et Tristan: les effets de 1'enlaidissement." Stanford French Reviews (1981): 85-95.

272

Bibliography

Gazelles, Brigitte and Charles Mela, eds. Modernite au moyen age: le defi du passe. Geneva: Droz, 1990. Cerquiglini, Bernard. Eloge de la variante: histoire critique de la philohgie. Paris: Seuil, 1989. —————. "Le style indirect libre et la modernite." Langages 73 (1984): 7-16. Chenerie, Marie-Luce. Le Chevalier errant dans les romans arthuriens en vers desXIP etXIIP siedes. Geneva: Droz, 1986. Chenu, Marie-Dominique, O.P. "Auctor, Actor, Autor." Bulletin du Cange:Archivum Latinitas Medii Aevi 3 (1927): 81-86. —————. Nature, Man, and Society in the Twelfth Century: Essays on New Theological Perspectives in the Latin West. Ed. and tr. Jerome Taylor and Lester K. Little. Chicago: University of Chicago Press, 1968. —————. La theologie au douzieme siecle. Etudes de Philosophic Medievale 45. Paris: Vrin, 1957. Cline, Ruth Huff. "The Influence of Romances on Tournaments of the Middle Ages." Speculum 20 (1945): 204-11. Colby, Alice M. The Portrait in Twelfth-Century French Literature. Geneva: Droz, 1965. Colish, Marcia. The Mirror of Language: A Study in the Medieval Theory of Knowledge. New Haven, Conn,: Yale University Press, 1968. Curtis, Renee L. "The Humble and Cruel Tristan: A New Look at the Two Poems of the Folie Tristan." Tristania 2 (1976): 3—11. Curtius, Ernst Robert. European Literature and the Latin Middle Ages. Tr. Willard R. Trask. New York: Harper Torchbooks, 1953. Dallenbach, Lucien. Le recit speculaire: essai sur la mise en abyme. Paris: Seuil, 1977. Damon, S. Foster. "Marie de France: Psychologist of Courtly Love." PMLA 44 (1929): 968-96. Dannenbaum, Susan. "Doubling and Fine Amor in Thomas' Tristan." Tristania 5 (i979): 3-14Dean, Ruth and Elspeth Kennedy. "Un fragment anglo-normand de la Folie Tristan de Berne." Moyen Age 79 (1973) • 57-72. Delbouille, Maurice. "CE1 chief de cest comencement' . . . (Marie de France, Prologue de Guigemar)" In Etudes de civilisation medievale (IXe-XIIe siedes): melanges offerts a Edmond-Rene Labande. Poitiers: Centre d'Etudes Superieures de Civilisation Medievale, 1974.185-96. —————. "Le fragment de Cambridge et la genese des 'Folies Tristan.'" Travaux de linguistique et de litterature 16 (1978): 119-22. Delcourt, Denyse. L}ethique du changement dans le roman du XIIe siecle. Geneva: Droz, 1990. Deroy, Jean. "Chretien de Troyes et Godefroi de Leigni, conspirateurs contre la Fin'Amor adultere." Cultura Neolatina 38 (1978): 67-78. Derrida, Jacques. De lagrammatologie. Paris: Minuit, 1967. Dornbush, Jean. "Ovid's Pyramus and Thisbe and Chretien's Le Chevalier de la Charrete." Romance Philology 36 (1982): 34—43. Dragonetti, Roger. "Le lai narratif de Marie de France, 'pur quei fu fez, coment e dunt.'" In Litterature, histoire, linguistique: recueil d'etudes offert a Bernard Gagnebin. Lausanne: L'Age de 1'Homme, 1974. 31-53.

Bibliography

273

—————. La vie de la lettre au moyen age (Le Conte dugraal). Paris: Seuil, 1980. Duby, Georges. "Dans la France du nordouest au XIP siecle: les 'jeunes' dans la societe aristocratique." Annales: Economies, Societes, Civilisations 19 (1964): 835-46. —————. Guerriers et paysans: VIP-XIP siecle, premier essor de Veconomie europeenne. Paris: Gallimard, 1973. —————. The Knight, the Lady, and the Priest: The Making of Modern Marriage in Medieval France. Tr. Barbara Bray. New York: Pantheon Books, 1983. Ducrot, Oswald and Tzvetan Todorov. Dictionnaire encyclopedique des sciences du langage. Paris: Seuil, 1972. Duggan, Joseph J. "Ambiguity in Twelfth-Century French and Provencal Literature: A Problem or a Value?" In Jean Misrahi Memorial Volume: Studies in Medieval Literature, ed. Hans R. Runte, Henri Niedzielski, and William L. Hendrickson. Columbia, S.C.: French Literature Publications, 1977. 136-49. —————. The Song of Roland: Formulaic Style and Poetic Craft. Berkeley: University of California Press, 1973. Durling, Nancy Vine. "The Knot, the Belt, and the Making of Guigemar" Assays: Critical Approaches to Medieval and Renaissance Texts 6 (1991): 29—53. Eckard, Gilles. "A propos d'un passage de la Folie Tristan de Berne." Travawc de Linguistique et de Litterature 16 (1978): 161-68. Edwards, Lee R. "The Labors of Psyche: Toward a Theory of Female Heroism." Critical Inquiry 6 (Autumn 1979): 33—49. Eisner, Sigmund. The Tristan Legend: A Study in Sources. Evanston, 111.: Northwestern University Press, 1969. Evans, Dafydd. "Wishfulfllment: The Social Function and Classification of Old French Romances." In Court and Poet: Selected Proceedings of the Third Congress of the International Courtly Literature Society (Liverpool 1980), ed. Glyn S. Burgess. ARCA, 5. Liverpool: Francis Cairns, 1981. 129—34. Faral, Edmond. Les arts poetiques duXIP et duXIIP siecle. Paris: Champion, 1971. —————. Mimes frangais du XIIP siecle: contribution a I'histoire du theatre comique au moyen age. Paris: Champion, 1910. Farmer, Sharon A. "Softening the Hearts of Men: Women, Embodiment, and Persuasion in the Thirteenth Century." In Embodied Love: Sensuality and Relationship as Feminist Values, ed. Paula M. Cooey, Sharon A. Farmer, and Mary Ellen Ross. San Francisco: Harper & Row, 1983. 115-33. Ferrante, Joan. "Artist Figures in the Tristan Stories." Tristania 4 (1979): 25-35. —————. "Male Fantasy and Female Reality in Courtly Literature." Women's Studies ii (1984): 88-91. Fitz, Brewster. "Desire and Interpretation: Marie de France's Chievrefoil" Tale French Studies 58 (1979): 182-89. —————. "The Prologue to the Lais of Marie de France and the Parable of the Talents: Gloss and Monetary Metaphor." Modern Language Notes 90 (1975): 558-64. —————. "The Storm Episode and the Weasel Episode: Sacrificial Casuistry in Marie de France's Eliduc." Modern Language Notes 89 (1974): 542-49. Fleischman, Suzanne. "Evaluation in Narrative: The Present Tense in Medieval 'Performed Stories.'" Tale French Studies 70 (1986): 199-251.

274

Bibliography

Foulet, Alfred and Karl D. Uitti. "The Prologue to the Lais of Marie de France: A Reconsideration." Romance Philology 35 (1981): 242-49. Foulon, Charles. "Les deux humiliations de Lancelot." BBSIA 8 (1956): 79-90. Fourrier, Anthime. Le Courant realiste dans le roman courtois en France au moyen age: I. Les debuts (XIIf siede). Paris: Nizet, 1960. —————. "Encore la chronologic des oeuvres de Chretien de Troyes." BBSIA 2 (1950): 69-88. Fox, John C. "Marie de France." English Historical Review 25 (1910): 303-6. Frappier, Jean. Etude surLaMort leRoiArtu: roman duXIIP siede. Paris: Minard; Geneva: Droz, 1936; rev. ed. 1961. —————. Etude sur (CYvain" ou "le Chevalier au Lion" de Chretien de Troyes. Paris: SEDES, 1969. —————. "Le Prologue du Chevalier de la Charrete et son interpretation." Romania 93 (1972): 337-77—————. "Structure et sens du Tristan: version commune, version courtoise." Cahiers de Civilisation Medievale 6 (1963): 255-80, 441-54. —————. "Sur le mot 'raison' dans le Tristan de Thomas d'Angleterre." In Linguistic and Literary Studies in Honor of Helmut A. Hatzfeld^ ed. A. S. Crisafulli. Washington, D.C.: Catholic University of America Press, 1964.163-76. —————. "Vues sur les conceptions courtoises dans les litteratures d'oc et d'oil au XII' siecle." Cahiers de Civilisation Medievale 2 (1959): 135-56. Freeman, Michelle. "C%&." In The Romances of Chretien de Troyes. Ed. D. Kelly. 115-30. —————. "Dual Natures and Subverted Glosses: Marie de France's cBisclavret.'" Romance Notes 25 (1985): 288—301. —————. "Marie de France's Poetics of Silence: The Implications for a Feminine Translation PMLA 99 (1984): 860-83. Freud, Sigmund. Beyond the Pleasure Principle. Tr. & ed. James Strachey. New York: Norton, 1961. Frey, J. A. "Linguistic and Psychological Coupling in the Lays of Marie de France." Studies in Philology 61(1964): 3-18. Frye, Northrup. Anatomy of Criticism: Four Essays. New York: Atheneum, 1969. Gallais, Pierre. "Bleheri, la cour de Poitiers et la diffusion des recits arthuriens sur le continent." Actes du VIP congres national de litterature comparee: Poitiers 27-29 mai 1965: moyen age et litterature comparee. Paris: Didier, 1967. 47-79. —————. "Recherches sur les mentalites des romanciers francos du moyen age." Cahiers de Civilisation Medievale 7 (1964): 479-93 and 13 (1970): 333-47. Genette, Gerard. Figures III. Paris: Seuil, 1972. Gibson, Gail McMurray. The Theater of Devotion: East Anglian Drama and Society in the Late Middle Ages. Chicago and London: University of Chicago Press, 1989. Gier, Albert. "Das Verwandtschaftsverhaltnis von afr. sens und sen" Romantisches Jahrbuch 28 (1977): 54-72. Gossman, Lionel. Medievalism and the World of the Enlightenment: The World and Work ofLacurne de Sainte-Palaye. Baltimore: Johns Hopkins University Press, 1968.

Bibliography 275 Gravdal, Kathryn. Ravishing Maidens: Writing Rape in Medieval French Literature and Law. Philadelphia: University of Pennsylvania Press, 1991. Green, D. H. "Orality and Reading: The State of Research in Medieval Studies." Speculum 65 (1990): 267—80. Grigsby, John L. "L'empire des signes chez Beroul et Thomas: cle sigle est tut neir.'" Melanges FoulonlMarche Romane 30 (1979): 115-25. —————. "Narrative Voices in Chretien de Troyes: A Prolegomenon to Dissection." Romance Philology 32 (1979): 268-72. —————. "The Narrator in PartonopeudeBlois, leEelInconnu, zndjoufroide Poitiers." Romance Philology 21 (May 1968): 536-45. Grundmann, Herbert. "Litteratus-illitteratus: der Wandel einer Bildungsroman von Altertum zum Mittelalter." Archiv fur Kulturgeschichte 40 (1958): 1-65. Guiette, Robert. "'Li conte de Bretaigne sont si vain et plaisant.'" Romania 88 (1967): 1-12. —————. "Questions de litterature." Romania Gandensia 8 (1960): 9-158. —————. Review of Hoepffner's Folie editions. Revue Beige de Philologie et d'Histoire 15 (1936): 526-27 and 18 (1939): 996-97Haidu, Peter. Aesthetic Distance in Chretien de Troyes: Irony and Comedy in "Cliges" and "Perceval." Geneva: Droz, 1968. —————. "The Episode as Semiotic Module in Twelfth-Century Romance." Poetics Today 4.4 (1983): 655-81. —————. "Fragments in Search of Totalization: Roland and the Historical Text." In Modernite au moyen age: le defi du passe, ed. Brigitte Gazelles and Charles Mela. Geneva: Droz, 1990. 73-99. —————. "The Hermit's Pottage: Deconstruction and History in Yvain" In The Sower and His Seed: Essays on Chretien de Troyes, ed. Rupert T. Pickens. Lexington, Ky: French Forum Pub., 1983. 127-45. —————. "Making It (New) in the Middle Ages: Towards a Problematics of Alterity" Diacritics 4 (1974): 1-12. —————. "Narrative Structure in Floire et Blancheflor. A Comparison with Two Romances of Chretien de Troyes." Romance Notes 14 (1972): 383—86. —————. "Narrativity and Language in Some Twelfth-Century Romances." Tale French Studies 51 (1974): 133—46. —————. "Repetition: Modern Reflections on Medieval Aesthetics." Modern Language Notes 92 (1977): 875-87. —————. "The Semiotization of Death: Open Text or Closed?" Style 20 (1986): 220-51. —————. "Le sens historique du phenomene stylistique: la semiose dissociative chez Chretien de Troyes." Europe 642 (1982): 36-47. —————. "Text and History: The Semiosos of Twelfth-Century Lyric as Sociohistorical Phenomenon (Chretien de Troyes: 'D'Amors qui m'a tolu')." Semiotica 33 (1981): 1-62. —————. "Text, Pretextuality, and Myth in the Folie Tristan dVxford" Modern Language Notes 88 (1973): 712-17. Hamon, Philippe. "Clausules." Poetique 24 (1975): 495-526. Harming, Robert. "Arthurian Evangelists: The Language of Truth in Thirteenth-

276

Bibliography

Century French Prose Romances." Philological Quarterly 64 (1985): 347-65. —————. "The Audience as Co-Creator of the First Chivalric Romances." Yearbook of English Studies u (1981): 1—28.

—————. "Beowulf as Heroic History." Medievalia et Humanistica 5 (1974): 77-102. —————. The Individual in Twelfth-Century Romance. New Haven, Conn.: Yale University Press, 1977. Harf-Lancner, Laurence. Lesfees au moyen age. Geneva: Slatkine, 1984. Haskins, Charles H. The Renaissance of the Twelfth Century. Cambridge, Mass.: Harvard University Press, 1927,1955. Hilton, Catherine. "Convention and Innovation in Partonopeu de Blois." Ph.D. diss., University of Massachusetts, 1984. Hirsch, E. D. The Validity in Interpretation. New Haven, Conn.: Yale University Press, 1967. Hodgson, Frederick. "Alienation and the Otherworld in Lanval, TConec, and Guigemar" Comitatus 5 (1974): 19-31. Holmes, Urban Tigner, Jr. A History of Old French Literature from the Origins to 1300. New York: Crofts, 1937; rev. ed. New York: Russell & Russell, 1962. —————. "Transition in European Education." In Twelfth-Century Europe and the Foundations of Modern Society, ed. Marshall Clagett, Gaines Post, and Robert Reynolds. Madison: University of Wisconsin Press, 1961. 15-38. —————. "A Welsh Motif in Marie's Guigemar." Studies in Philology 39 (1942): 11-14. Horrent, M. J. "La composition de la Folie Tristan de Berne" Revue Beige de Philologie et d'Histoire 25 (1946—47): 21—38. Huchet, Jean-Charles. "Nom de femme et ecriture feminine au moyen age: les Lais de Marie de France." Poetique 48 (1981): 407-30. Hult, David. "Author/Narrator/Speaker: The Voice of Authority in Chretien's Charrete" In Discourses of Authority in Medieval and Renaissance Literature, ed. Kevin Brownlee and Walter Stephens. Hanover, N.H.: University Press of New England, 1989. 76-96. —————. Self-Fulfilling Prophecies: Readership and Authority in the First Roman de la Rose. Cambridge: Cambridge University Press, 1986. Hunt,Tony. "AbelardianEthics andBeroul's Tristan"Romania98 (1977): 501—40. —————. "Aristotle, Dialectic, and Courtly Literature." Viator 10 (1979): 95-129. —————. "Glossing Marie de France." Romanische Forschungen 86 (1974): 396-418. —————. "Of Interpretation and Emendation." Tristania i (1976): 29-59. —————. "The Rhetorical Background to the Arthurian Prologue: Tradition and the Old French Vernacular Prologues." Forum for Modern Language Studies 6 (1970): 1-28. —————. "The Significance of Thomas' Tristan." Reading Medieval Studies 7 (1981): 41-61. Illingworth, R. N. "Celtic Tradition and the Lai of Guigemar." Medium Aevum 31 (1962): 176-87. Jakobson, Roman. "Deux aspects du langage et deux types d'aphasie." In Essais de linguistique generate, tr. Nicolas Ruwet. Paris: Minuit, 1963. 43-67. Jauss, Hans Robert. "The Alterity and Modernity of Medieval Literature." New Literary History 10 (1979): 181-230.

Bibliography

277

—————. "Literary History as a Challenge to Literary Theory" and "Theory of Genres." In Toward an Aesthetic of Reception, tr. Timothy Bahti. Theory and Literature, vol. 2. Minneapolis: University of Minnesota Press, 1982. 3-45. —————. "Theory of Genres and Medieval Literature." In Toward an Aesthetic of Reception, tr. Timothy Bahti. Theory and Literature, vol. 2. Minneapolis: University of Minnesota Press, 1982. 76-109. —————. Toward an Aesthetic of Reception. Tr. Timothy Bahti. Theory and Literature, vol. 2. Minneapolis: University of Minnesota Press, 1982. Jodogne, Omer. "Comment Thomas d'Angleterre a compris 1'amour de Tristan et d'Iseut." Lettres Romanes 19 (1965): 103—19. Johnson, Barbara. "The Critical Difference." Diacritics 8 (1978): 2-9. Jolivet, Jean. Arts du langage et theologie chezAbelard. Paris: Vrin, 1969. Jolles, Andre. Les formes simples. Tr. Antoine Marie Buguet. Paris: Seuil, 1972. Jonin, Pierre. "Le je de Marie de France dans les Lais." Romania 103 (1982): 170-96. —————. "Merveilleux celtique et symbolisme universel dans Guigemar de Marie de France." In Melanges Wathelet-Willem. Liege: Cahiers de PARULg, 1978. 239-55—————. Les personnages feminins dans les romans fran$ais de Tristan au XIIe siecle. Aix-en-Provence: Ophrys, 1958. Joris, Pierre-Marie. "Note sur le noir dans le Partonopeu de Blois" Senefiance 24 (1988): 141-52. Kasprzyk, Krystyna. "Fonction et technique du souvenir dans la Folie Tristan (Berne 354)." In Etudes Felix Lecoy. Paris: Champion, 1973. 261-270. Keen, Maurice. Chivalry. New Haven, Conn, and London: Yale University Press, 1984. Kelly, Douglas. "The Art of Description." In Lacy, Kelly, and Busby, ed., The Legacy of Chretien de Troyes, vol. 1,191—221. —————. The An of Medieval French Romance. Madison: University of Wisconsin Press, 1992. —————. "Courtly Love in Perspective: The Hierarchy of Love in Andreas Capellanus." Traditio 24 (1960): 119-47. —————. "'En uni dire' (Tristan, Douce, 839) and the Composition of Thomas' Tristan" Modern Philology 67 (1969-70): 9-17. —————. "The Logic of the Imagination in Chretien de Troyes." In The Sower and His Seed: Essays on Chretien de Troyes, ed. Rupert T. Pickens. Lexington, Ky.: French Forum Pub., 1983. 9—30. —————. "Matiere and Genera dicendi in Medieval Romance." Tale French Studies 51 (1974): 147-59—————. "Romance and the Vanity of Chretien de Troyes." In Brownlee and Brownlee, ed., Romance: Generic Transformation from Chretien de Troyes to Cervantes. 74-90. —————. "Sens" and "Conjointure" in the "Chevalier de la Charrete." The Hague: Mouton, 1966. —————. "Theory of Composition in Medieval Narrative Poetry and Geoffrey of Vinsauf's PoetriaNova" Mediaeval Studies 31 (1969): 117-48.

278 Bibliography —————. "Topical Invention in Medieval French Literature." In Medieval Eloquence: Studies in the Theory and Practice of Medieval Rhetoric, ed. James J. Murphy. Berkeley: University of California Press, 1978. 231-51. Kelly, Douglas, ed. The Romances of Chretien de Troyes: A Symposium. Edward C. Armstrong Monographs on Medieval Literature, 3. Lexington, Ky: French Forum Pub., 1986. Kermode, Frank. The Genesis of Secrecy: On the Interpretation of Narrative. Cambridge, Mass.: Harvard University Press, 1979. —————. The Sense of an Ending: Studies in the Theory of Fiction. London and New York: Oxford University Press, 1966,1967; rpt. 1970. Knapton, Antoinette. Mythe et psychologic chez Marie de France dans "Guigemar." Chapel Hill: North Carolina Studies in Romance Languages and Literatures, 1975Knowles, David. The Evolution of Medieval Thought. New York: Random House, 1962. Kohler, Erich. UAventure chevaleresque: ideal et realite dans le roman courtois. Tr. Elaine Kaufholz. Paris: Gallimard, 1956,1974. Koubichkine, Michele. "A propos du lai de Lanval." MoyenAge 78 (1972): 467-88. Kristeva, Julia. Semeiotike: recherches pour une semanalyse (extraits). Collection Points. Paris: Seuil, 1969. Krueger, Roberta L. "The Author's Voice: Narrators, Audience and the Problem of Interpretation." In Lacy, Kelly, and Busby, ed., The Legacy of Chretien de Troyes, vol. 1,115-40. —————. "Contracts and Constraints: Courtly Performance in Tvain and the Charrete" Houston German Studies 6 (1986): 92-104. —————. "Love, Honor, and the Exchange of Women in Tvain: Some Remarks on the Female Reader." Romance Notes 25 (1985): 302—17. —————. "Reading the TvainlCharrete: Chretien's Inscribed Audiences at Noauz and Pesme Aventure." Forum for Modern Language Studies 19 (1983): 172-87. —. "Textuality and Performance in Partonopeu deBlois." Assays: Critical Approaches to Medieval and Renaissance Texts 3 (1985): 57—72. Lacy, Norris. The Craft of Chretien de Troyes: An Essay on Narrative Art. Leiden: Brill, 1980. —————. "Thematic Analogues in Erec." L'Esprit Createur 9 (1969): 267—74. Lacy, Norris J., Douglas Kelly, and Keith Busby, eds. The Legacy of Chretien de Troyes. 2 vols. Amsterdam: Rodopi NV, 1987,1988. Laurie, Helen C. "A Note on the Composition of Marie's Guigemar." Medium Aevum^4 (1975): 242-48. Lazar, Moshe. "Lancelot et la 'mulier mediatrix': la quete de soi a travers la femme." L'Esprit Createur 9 (1969): 243-56. Le Gentil, Pierre. "Sur Pepilogue de Tristan de Thomas." In Melanges Jeanne Lods. Paris: Collection de 1'Ecole Normale Superieure de Jeunes Filles, 10, 1978. Vol. I, 365-70. Le Goff, Jacques. "Clerical Culture and Folklore Traditions in Merovingian Civilization." In Time, Work and Culture in the Middle Ages. Tr. Arthur Goldhammer. Chicago: University of Chicago Press, 1980. 153-58.

Bibliography

279

Le Rider, Paule. "Or est venuz qui I'aunera ou la fortune litteraire d'un proverbe." In Melanges Jeanne Lock. Paris: Collection de PEcole Normale Superieure de Jeunes Filles, 10,1978. Vol. 1,393-409. Looze, Laurence de. "Generic Clash, Reader Response, and the Poetics of the Non-Ending in Le Eel Inconnu" In Courtly Literature: Culture and Context, Proceedings of the sth Triennial Congress of the International Courtly Literature Society, Dalften, The Netherlands, 9-16 Aug. 1986, ed. Keith Busby and Erik Kooper. Amsterdam and Philadelphia: John Benjamins, 1990. 113-23. Lotman, Juri. La structure du texte artistique. Tr. A. Fournier et al. Paris: Gallimard, 1973. Lutoslawski, W. "Les Folies de Tristan." Romania 15 (1886): 511-33. Maddox, Donald. The Arthurian Romances of Chretien de Troyes: Once and Future Fictions. Cambridge: Cambridge University Press, 1991. —————. "L'intersubjectivite et Panalyse du recit medieval." In Mittelalterbilder aus neuer Perspektive: Wurzburger Kolloquium, ip8s, ed. Ruhe and Behrens. Munich: Wilhelm Fink, 1986. 303-13. —————. "Lancelot et le sens de la coutume." Cahiers de Civilisation Medievale 29 (1986): 339-53—————. "Roman et manipulation au XI? siecle." Poetique 66 (1986): 179-90. —————. "Specular Stories, Family Romance, and the Fictions of Courtly Culture." Exemflaria 3 (1991): 299-326. —————. Structure and Sacring: The Systematic Kingdom in Chretien's Erec et Enide. Lexington, Ky: French Forum Pub., 1978. —————. "Triadic Structure in the Lais of Marie de France." Assays: Critical Approaches to Medieval and Renaissance Texts 3 (1985): 19-40. —————. "Trois sur deux: theories de bipartition et de tripartition des oeuvres de Chretien de Troyes." CEuwes et Critiques 5 (1980/81): 91-102. Mandel, Jerome. "Elements in the Charrete World: The Father-Son Relationship." Modern Philology 62 (1964): 97-104. —————. "Proper Behaviour in Chretien's Charrete" French Review 48 (1975): 683-89. McCullough, Florence. "Length, Recitation and Meaning of the Lais of Marie de France." Kentucky Romance Quarterly 25 (1978): 257—68. "Medieval Literature and Contemporary Theory." New Literary History 10 (1979). Menard, Philippe. Les Lais de Marie de France: contes d3amour et d'aventures du moyen age. Paris: Presses Universitaires de France, 1979. Micha, Alexandre. La tradition manuscrite des romans de Chretien de Troyes. Geneva: Droz, 1966. Michener, Richard L. "Courtly Love in Chretien: the 'Demande d'amour.'" Studia neophilologica 42 (1970): 353-360. Mickel, Emanuel J., Jr. Marie de France. New York: Twayne, 1974. —————. "A Reconsideration of the Lais of Marie de France." Speculum 46 (1971): 39-65. Mitsch, Ruthmarie H. "The Monologues of Tristan in the Tristan of Thomas." Tristania 2 (1977): 29-39. Neumann, Erich. Amor and Psyche: The Psychic Development of the Feminine—A

280

Bibliography

Commentary on the Tale ofApuleius. Tr. Ralph Manheim. Bollingen Series, vol. 54. Princeton, N.J.: Princeton University Press, 1971. Newstead, Helaine. "The Traditional Background of Partonopeus de Blois" PMLA 61 (1946): 916-46. Nichols, Stephen G., Jr. "Realistic Perception in Twelfth-Century French Literature." UEsprit Createur 5 (1965): 208-18. —————. "Remodeling Models: Modernism and the Middle Ages." In Gazelles and Mela, ed., Modernite au moyen age: le defi du passe. 45-72. —————. Romanesque Signs: Early Medieval Narrative and Iconography. New Haven, Conn.: Yale University Press, 1983. Noble, Peter S. Love and Marriage in Chretien de Troyes. Cardiff: University of Wales Press, 1982. Oilier, Marie-Louise. "Le discours en abyme ou la narration equivoque." Medioevo Romanzo i (1974): 351-64. —————. "Les Lais de Marie de France ou le recueil comme forme." In LaNouvelle: genese, codification et rayonnement d}un genre medieval, ed. Michaelangelo Piccone, Guiseppe Di Stefano, and Pamela Stewart. Montreal: Plato Academic Press, 1983. 64-79. —————. "Proverbe et sentence: le discours d'autorite chez Chretien de Troyes." Revue des Sciences Humaines 41 (1976): 329—57. Ong, Walter. Interfaces of the Word: Studies in the Evolution of Consciousness and Culture. Ithaca: Cornell University Press, 1977. —————. "Literacy and Orality in Our Times." Profession 79 (selected articles from the Bulletins of the ADE and the ADFL): 1-7. —————. The Presence of the Word: Some Prolegomena for Cultural and Religious History. New Haven, Conn.: Yale University Press, 1967. —————. Rhetoric, Romance and Technology: Studies in the Interaction of Expression and Culture. Ithaca, N.Y.: Cornell University Press, 1971. "Oral and Written Traditions in the Middle Ages." New Literary History 16,1 (1984). Paden, William D., Jr. "Utrum copularentur: Of Cors." L'Esprit Createur 19 (1979): 70-83. Painter, Sidney. French Chivalry: Chivalric Ideas and Practices in Medieval France. Baltimore: Johns Hopkins University Press, 1940. Paris, Gaston. "Etudes sur les romans de la Table Ronde. Lancelot du Lac. II. Le Conte de la Charrete." Romania 12 (1883): 459-534. Patterson, Lee. "On the Margin: Postmodernism, Ironic History, and Medieval Studies." Speculum 65 (1990): 87-108. Payen, Jean-Charles. "Le lai narratif." Typologie des sources du moyen age occidental, Fasc. 13. Turnhout: Brepols, 1975. 33-63. —————. "Lancelot contre Tristan, ou la conjuration d'un mythe subversif." In Melanges de langue et de litterature medievales offerts a Pierre Le Gentil, ed. Jean Dufournet and Daniel Poirion. Paris: SEDES, 1973. 617—33. Pickens, Rupert T. "Equitan: anti-Guigcmar" Romance Notes 15 (1973-74): 361-67. —————. "Estoire, Lai and Romance: Chretien's Erec et Enide and Cliges" Romanic Review 66 (1975): 247-62.

Bibliography

281

—————. "Historical Consciousness in Old French Narrative." French Forum 4 (1979): 168-84. —————. "La poetique de Marie de France d'apres les Prologues des Lais." Lettres romanes 32 (1978): 367—83. —————. "Thematic Structure in Marie de France's Guigemar" Romania 95 (1979): 328-41. —————, ed. The Sower and His Seed: Essays on Chretien de Troyes. Lexington, Ky.: French Forum Pub., 1983. Regalado, Nancy. "'La Chevalerie CelestieP: Spiritual Transformations of Secular Romance in La Queste del Saint Graal" In Brownlee and Brownlee, eds., Romance. 91—113. —————. "Tristan and Renart: Two Tricksters." UEsprit Createur 16 (1976): 30-38. Reiss, Timothy J. The Discourse of Modernism. Ithaca, N.Y.: Cornell University Press, 1982. Richards, Earl Jeffrey. "Les rapports entre le Lai de Guigemar et le Roman #Eneas: considerations generiques." In Le recit brefau moyen aae: Actes du Colloque des 27, 28 et 29 avril 1079, Universite de Picardie, Centre d'Etudes Medievales, publics par les soins de Danielle Buschin0er. Paris: Champion, 1980. 45-56. Riffaterre, Michael. Fictional Truth. Baltimore and London: Johns Hopkins University Press, 1990. Robertson, D. W., Jr. A Preface to Chaucer: Studies in Medieval Perspectives. Princeton, N.J.: Princeton University Press, 1962. Robertson, Howard S. "Love and the Other World in Marie de France's Eliduc." In Essays in Honor of Louis Frances Solano, ed. Raymond J. Cormier and Urban T. Holmes. Chapel Hill: North Carolina University Studies in the Romance Languages and Literatures, 92,1970.167-76. Robertson, J. Duncan. "Literary Tradition and Poetic Realization in the Folies Tristan" Ph.D. diss., Princeton University, 1972. Rychner, Jean. "La presence et le point de vue du narrateur dans deux recks courts: le Lai de Lanval et la Chatelaine de Vergi" Vox Romanica 39 (1980): 86-103. —————. "Le sujet et la signification du Chevalier de la Charrete." Vox Romanica 27 (1968): 50-76. Sargent-Baur, Barbara Nelson. "The Missing Prologue of Chretien's Chevalier au Lion" French Studies 41 (1987): 385—94. Schaefer, Jacqueline T. "Tristan's Folly: Feigned or Real?" Tristania 3 (Nov. 1977): 3-15Schmid, Elisabeth. Familiengeschichten un Heilsmytholo0ie: die Verwandtschafstrukturen in den franzosischen und deutschen Gralromanen des 12. und 13. jahrhunderts. Beihefte zur Zeitschrift fur romanische Philologie 211. Tubingen: Max Niemeyer, 1986. Schmolke-Hasselmann, Beate. Der arthurische Versroman von Chrestien bis Froissart: Zur Geschichte einer Gattun0. Beihefte zur Zeitschrift fur romanische Philologie 177. Tubingen: Max Niemeyer, 1980. —————. "Tristan als Dichter—ein Beitrag zur Erforschung des lai lyrique breton." Romanische Forchunpjen 98 (1986): 258-76.

282

Bibliography

Scholes, Robert and Robert Kellogg. The Nature of Narrative. Oxford and New York: Oxford University Press, 1966. Shirt, David J. "Chretien de Troyes and the Cart." In Studies in Medieval Language and Literature in Memory ofF. Whitehead^ ed. W. Rothwell et al. Manchester: Manchester University Press, 1973. 279-301. —————. "Chretien de Troyes et une coutume anglaise." Romania 94 (1973): 178-95. —————. "Godefroi de Lagny et la composition de la 'Charrete.'" Romania 96 (1975): 27-52. —————. "How Much of the Lion Can We Put Before the Cart>" French Studies 31 (1977): 1-17Sienaert, Edgar. Les Lais de Marie de France: du conte merveilleux a la nouvelle psychologique. Paris: Champion, 1978. Sklute, Larry M. "The Ambiguity of Ethical Norms in Courtly Romance." Genre ii (1978): 315-32. Smalley, Beryl. The Study of the Bible in the Middle Ages. Oxford: Basil Blackwell, 1952; Notre Dame, Ind.: Notre Dame University Press, 1964. Smith, Barbara Herrenstein. Poetic Closure: A Study of How Poems End. Chicago: University of Chicago Press, 1968. Smith, Julia M. H. "Oral and Written: Saints, Miracles, and Relics in Brittany, c. 850-1250." Speculum 65 (1990): 309-43. Southern, R. W. Medieval Humanism and Other Studies. New York: Harper & Row, 1970. Speer, Mary Blakely Review of T. B. W. Reid, The Tristan ofBeroul: A Textual Commentary. In Romance Philology 30 (1977): 639. Stefanini, Jean. La voix pronominale en ancien et en moyenfran$ais. Publication des Annales de la Faculte des Lettres, N.S. 31. Ak-en-Provence: Ophyrs, 1962. Stevens, Martin. "The Performing Self in Twelfth-Century Culture." Viator 9 (1978): 194-212. Stock, Brian. Implications of Literacy: Written Language and Models of Interpretation in the Eleventh and Twelfth Centuries. Princeton, N.J.: Princeton University Press, 1983. Storey, H. Wayne. Review of The Concepts of Closure, ed. David Hult. In Romance Philology 44 (1990): 91-99. Strayer, Joseph. "The Development of Feudal Institutions." In Twelfth-Century Europe and the Foundations of Modern Society, ed. Marshall Clagett, Gaines Post, and Robert Reynolds. Madison: University of Wisconsin Press, 1961. 77-88. Sturges, Robert. "Texts and Readers in Marie de France's Lais" Romanic Review 71 (1980): 244-64. Trindade, W. Ann. "The Enemies of Tristan." Medium Aevum 43 (1974): 6-21. —————. "The Man with Two Wives—Marie de France and an Important Irish Analogue." Romance Philology 27 (1974): 466-78. Uitti, Karl D. "Le Chevalier au Lion (Tvain)." In Kelly, ed., The Romances of Chretien de Troyes. 182-231. —————. Story and Celebration in Old French Narrative Poetry,. 1150-1200. } Myth, Princeton, N.J.: Princeton University Press, 1973.

Bibliography

283

Uri, S. P. "Some Remarks on PartonopeusdeBlois" Neofkilologus 37 (1953): 83—98. Vance, Eugene. "Le combat erotique chez Chretien de Troyes." Poetique 12 (1972): 544-71—————. From Topic to Tale: Logic and Narrativity in the Middle Ages. Theory and History of Literature, 47. Minneapolis: University of Minnesota Press, 1987. —————. "Love's Concordances: The Poetics of Desire and the Joy of the Text." Diacritics 4 (1974): 40-52. —————. "Mervelous Signals: Poetics, Sign Theory, and Politics in Chaucer's Troilus? New Literary History 10 (1979): 293-337. —————. "Roland and the Poetics of Memory." In Textual Strategies: Perspectives in Post-Structuralist Criticism, ed. J. V. Harari. Ithaca, N.Y.: Cornell University Press, 1979- 374-403—————. "Signs of the City: Medieval Poetry as Detour." New Literary History 4 (1973): 557-74. Van den Boogaard, Nico. "Vocabulaire de lais, vocabulaire de fabliaux." RapportsHet Franse Boek 49 (1979): 97-106. Verchere, Chantal. "Peripherie et croisement: aspects du nains dans la litterature medievale." Senefiance 5 (1978): 251-79. Vinaver, Eugene. "Landmarks in Arthurian Romance." In The Expansion and Transformations of Courtly Literature, ed. N. B. Smith and J. T. Snow. Athens: University of Georgia Press, 1980. 17-31. Vitz, Evelyn Birge. "Desire and Causality in Medieval Narrative." Romanic Review 71 (1980): 213-43. Reprinted in Medieval Narrative and Modern Narratology: Subjects and Objects of Desire. New York and London: New York University Press, 1989.176-212. —————. "The Lais of Marie de France: Narrative Grammar and the Literary Text," Romanic Review 74 (1983): 383-404. Reprinted in Medieval Narrative and Modern Narratology: Subjects and Objects of Desire. New York and London: New York University Press, 1989. 149-175. —————. "Rethinking Old French Literature: The Orality of the Octosyllabic Couplet. " Romanic Review 77 (1986): 307-21. Vogel, K. Sneyders de. "La suite du Parthenopeu de Blois et la version hollandaise." Revue des Langues Romanes 48 (1905): 5-29. Walters, Lori. "Le role du scribe dans Forganisation des manuscrits des romans de Chretien de Troyes." Romania 106 (1985): 303-25. Warning, Rainer. "Pour une pragmatique du discours fictionnel." Poetique 38 (1979): 221-37. West, G. D. "The Description of Towns in Old French Verse Romances." French Studies 9 (i957): 5O-59. Wetherbee, Winthrop. Platonism and Poetry in the Twelfth Century: The Literary Influence of the School ofChartres. Princeton, N.J.: Princeton University Press, 1972. Woods, William S. "Femininity in the Lais of Marie de France." Studies in Philology 67 (1950): 1-19. York, Ernest C. "Isolt's Trial in Beroul and La Folie Tristan d'Oxford" Medievalia et Humanistica 6 (1975): 157-61.

284

Bibliography

Zink, Michel. "The Allegorical Poem as Interior Memoir." Tale French Studies 70 (1986): 100-126. —————. "Chretien et ses contemporains." In Lacy, Kelly, and Busby, ed., The Legacy of Chretien de Troyes. Vol. 1,5-32. —————. "Une mutation de la conscience litteraire: le langage romanesque a travers des exemples franqais du XIP siecle." Cahiers de Civilisation Medievale 24 (1981): 3-27. —————. "Remarques sur les conditions de Panonymite dans la poesie lyrique franc,aise du Moyen Age." In Melanges de langue et litterature frangaises du moyen age et de la Renaissance offerts a Monsieur Charles Foulon. Rennes: Institut de France, Universite de Haute-Bretagne, 1980. 421-27. —————. La subjectivite litteraire: autour du siecle de saint Louis. Paris: Presses Universitaires de France, 1985. Zumthor, Paul. Essai depoetique medievale. Paris: Seuil, 1972. —————. "Intertextualite et mouvance." Litterature 41 (1981): 8-16. —————. Parler du moyen age. Paris: Minuit, 1980. —————. La poesie et la voix dans la civilisation medievale. Paris: Presses Universitaires de France, 1984. —————. "Le texte fragment." Langue frangaise 40 (1978): 75-82.

Index

Adam and Eve. See Partonopeu de Blois Anchises, 117,118 Andreas Capellanus, 84, 86 Anonymity: of authors, 28,112,160-61, 179-81, 210, 255 n.33; of characters, 70,181, 183. See also Identity; Disguises, incognito Anselot, 144,147,149, 155 Arthur, 98,104; in Chevalier de la Charrete, 91, 95-97,101; granting of Rash Boons, 65-66, 68, 81-82; in Folie Tristan d'Oxford, 3; as historical and fictive figure, 86, 113, 213; in Roman de Brut, 85,105 Auerbach, Erich, 263 n.2 Author, 9,55-56,112,150,157-58,161,194; and characters, 57,177,180-84,188, 208-10; and patron, 7, 61, 84-88, 94, 106, in, 158,178,181,189, 210, 24on.3i; and public, 7,10,105, m, 197, 208-12, 216, 263 n.2. See also Narrator, authority of Bademagu: in Chevalier de la Charrete, 67, 77, 78, 80, 98-102, 99, 107; chastising Guenevere, 79; fatherly love of, 100,101; and Lancelot, 99-100; and Meleagant, 101 Banfield, Ann, 256 n. 38 Barteau, Franchise, 44 Barthes, Roland, 23in.22 Baum, Richard, 203, 256 n. 35 Baumgartner, Emmanuele, 215—16, 234 n.i Bauml, Franz, 228n.9, 233^40, 244n.68, 26on.56, 263 n.2 Beauty, no, 117,121,128,145,150,151,155, 222; and birth, 8,117-21,130-31,143-44; contest of, no, 120,130-32,141-42, 24911.43; and judgment, 120,131-132,139-43; male and female, 119-20,126-33,137,142,146, 249n.43; in opposition to fits a vilain, 8, 118,130,143,144,153; and power, 137 Bedier, Joseph, 29, 229nn.i, 20, 232nn.29, 38, 233nn.44, 48, 237H.33 Benoit de Sainte More (Troie), 203

Berkeley, Gail, 234^55 Berne Folly. See Folie Tristan de Berne Beroul, 28-29, 30, 89, 97-100, 23on.9, 232 n.34, 234n.5i; compared with Thomas, 54; narrator of, 45 Bisclawet, 171,192, 200 Blakeslee, Merritt, 236n.2i Bloch, Marc, 264n.ii Bloch, R. Howard, 14, 245^15 Blumenfeld-Kosinski, Renate, 229n.i7 Bodel, Jean, 212 Bogin, Meg, 248 n. 38 Bozoky, E., 249^45 Brand, Wolfgang, 238n.5, 24in.3i Brengain: in Folie Tristan d'Oxford, 13,14, 19, 20, 25, 27, 32; recognition of fool, 17; as witness, 34; in Tristan, 54 Breri, 56 Brewer, Derek, 240 n. 30 Brightenback, Kristine, 253 n.5 Brooks, Peter, 235n.n Brownlee, Kevin, 228n.8, 242n.42 Bruckner, Matilda Tomaryn, 228n.8, 238n.5, 239n.i6, 248n.37, 265n.23 Burgess, Glyn S., 253nn.3, 6, 256^36 Burns, E. Jane, 244n.66 Busby, Keith, 25on.48, 253^3, 256^36 Calogrenant, in Cliges, 91; as narrator, 9-10 Carruthers, Mary, 229n.i8, 23on.i3, 25411.20, 257n.48 Cerquiglini, Bernard, 256 n. 38 Chaitivel, 170,183,188,192 Chanson de Roland, 89 Charrete. See Chevalier de la Charrete Chef d'Oire, 116-117,121,122 Chenu, Marie-Dominique, 261 n.65 Chevalier de la Charrete, 6-7, 60-108. See also Author; Narrator; Prologues and epilogues; Tournaments; Tristan legend Chievrefoil, 28, 32,159,188. See also Closure; Gender; Tristan legend; Truth Chretien de Troyes, 7,10-11, 60-108,153-

286

Index

Chretien de Troyes (continued] 54. See also Author; Continuation; Fictionality in romance; Godefroi de Leigni; Intertextuality; Marie de Champagne; Marie de France; Narrator; Signs; Tristan legend; Truth Christine de Pizan, 26in.58 Cliges, 61, 69-70, 75, 91,136. See also Disguises, incognito Cliges, 91, 94,106,158, 203 Cline, Ruth Huff, 238 n.9 Closure, 2, 5, 7, u, 36, 89-90, 93, 96, 149, 158,169,176,184,188, 26in.67, 262n.68; and open-endedness, 5-6, u, 87,102-4, 108,151-56,170,199-206, 207, 213-19, 225, 240 n.22, 241 n. 39; operating at different levels, 9, 90,178,199-201, 214, 219. See also Intertextuality Colish, Marcia, 234^54 Collas, J. P., 25in.6i Conte du Graal. See Perceval Continuation, 7, 8, 90, 93,104, 218, 219, 228n.8, 24in.4i, 242^42; and Partonopeu de Blots, no, 111-12,146,147,148-50,15156, 210, 245 n.2, 247n.25, 252^64 Crescini, V., 248 n.38 Curtis, Renee L., 232 n.29, 233 n.44 Curtius, Ernst Robert, 203, 262n.74 Dallenbach, Lucien, 65, 238 n.8 Damon, S. Foster, 26in.64 Dannenbaum, Susan, 234n.i, 235 n.2 De Amove. See Andreas Capellanus Dean, Ruth, 233^44 Delbouille, Maurice, 33, 234^50, 255 n. 32 Delcourt, Denyse, 23on.io, 252^65 Derrida, Jacques, 227n.4, 23in.2o, 265^24 DeusAmanz, 167,182, 200 Discourse, 167, 220—23, 228n.n; from action to, 2,169-70,187-88; style indirect libre, 169,185-86,197, 256nn.38, 42; use of direct and indirect, 28,168,185,186, 257n.46, 262n.75 Disguises, 12-13, 20, 21, 25-26, 27, 32, 23in.23; incognito, 68-69, 71, 75, 82, 99, 126,135-36; invisibility, 8, no, 122, 12829,133—38,151,154. See also Anonymity; Identity; Voice Doubling, 5, 61—62, 90—92, 95, 98, 218; in Marie de France, 164-66,168,172; in Thomas, 5,37-46,50-52, 54, 57-58, 210,

216, 235n.n, 237n.28. See also Selection and substitution Dragonetti, Roger, 241 n. 35 Duby, Georges, 228n.n, 264n.n Duggan, Joseph, 233^40 Durling, Nancy Vine, 256 n. 37 Eilhart, 50 Eliduc, 173-76,185-86,187 Eliduc, 162-63,171,178,179,185-86; Guildeluec ha Guilliadun, 183. See also Closure; Prologues and epilogues; Selection and substitution Eneas, 118 Epilogues, 50-52, 87-90, 92-93, no, in, 153-54,155,178,179, 181, 202, 203-5, 214, 215, 218, 219, 236n.23, 242n.42, 25in.62,

262n.74- See also Prologues Equitan, 171, 200 Erec etEnide, 62, 84, 93, 94, 97,158; and Guenevere and White Stag Custom, 81. See also Medieval textuality; Renarration, bipartition and tripartition Evans, Dafydd, 151 Fables. See Marie de France Fairies. See Lais and fairies Faral, Edmond, 232 n.31 Farmer, Sharon A., 234n.52 Ferrante, Joan, 253 n.3, 26onn.53,58 Fictionality in romance, i, 60, 64-65, 66, 76, 84, 87, 94—95, IO2, 112, 138, 155, 211, 213,

223-25, 226n.34, 239n.i6, 24.in.iy, fable, 10, 85,105,109,113-15,122,150, 247n.3o; history and romance, 2, 4, 7, 85—86,104, 106,109,112,113-17,150-51, 208, 212-13, 228n.i3, 244n.73, 246nn.i3,16, 247^24; truth and fiction, 2, 97,100,104—6,107, 192, 213, 218, 220, 223, 246n.ii Fitz, Brewster, 237n. 32, 261 n.62 Floire et Blancheflor, 151,160, 203 Florimont, 154 Foerster, Wendelin, 209, 240 n.31 Folie Tristan de Berne: compared with Folie Tristan d'Oxford, 31, 32-33; compared with Marie de France's Lais, 159. See also Narration. Folie Tristan d'Oxford, 4, 9, n, 12-36, 33,103, 104; compared with Folie Tristan de Berne, 31, 32-33; compared with Marie de France's Lais, 159; compared with Tho-

Index mas's Tristan, 5-6, 31. See also Author; Closure; Intertextuality; Narration; Orality; Renarration; Signs; Voice. Foulet, Alfred, 260 n.52 Fourrier, Anthime, 109,115,117,150, 245n.2, 246n.i5, 247n.24, 252n.62 Fowles, John, 262^76 Fox, John C., 229n.20 Frappier, Jean, 54,57,161, 234 n.i, 236 n. 14, 237n.27, 253n.io, 26411.13 Freeman, Michelle, 242n.so Fresne, 171,183, 200 Freud, Sigmund, 235n.n Frye, Northrop, 251^59 Gallais, Pierre, 233^40, 237n.29 Gauvain, 69-70; in Chevalier de la Charrete, 63, 69, 71, 72, 74, 82, 91, 92,102; contest with Cliges, 69; in Le Conte du Graal, 92; in Yvain, 92 Gender, 42, 44,164-65,170,173,177,183, 187, 245n.4, 250^57, 26on.58; and genre, 2, 8, 138, 144, 147-48, 150, 194-95, 224; interplay of roles, 113,120-27,130,13334,138-43,146,181, 249H.43; reversals, 8, no, 118,123,124, 127,131—32,134—38, 248 n.39. See also Genre; Naming; Truth, and point of view Genette, Gerard, 208-9, 228 n.8 Genre: fusion of different genres, 7,109, 116-17,125,149-50,155-56, 210—n, 213, 249n.48; romance, 2-4, 6, 9, 60, 62, 11213,125-26,135,151,152,154, 207-25, 227n.3, 245n.9, 262n.i; romance and genealogy, 109,115—16,120,144,152,153, 222, 246n.15, 247n.25; romance and lais, i, 8, 102-4, no, 123,126-27,149,152, 15763, 188,189-91, 205, 211, 217-18, 219, 227n.i, 25in.6i; romance and lyric, 7, 5152, no-12,124-25,127-28,138,143,149, 150,153,154, 210, 211. See also Lais and fairies. Geoffrey of Monmouth, 104,115 Geoffroi de Vinsauf, 232 n.31 Gibson, Gail McMurray, 229 n.19, 25on.49 Gildea, Joseph, 245 n.2 Godefroi de Leigni, 61; and Chretien de Troyes, 84—90, 93 Gottfried von Strassburg, 47, 48, 97-98,103 Graelent, 123,159-60 Gravdal, Kathryn, 239n.i6, 263n.i

287

Green, D. H., 26on.56 Grigsby, John L., 235 n.2, 241 n.36 Guenevere, 99,151; in Chevalier de la Charrete, 66, 72, 74, 96-97,101; and Arthur, 80; characterization of, 78-84; and Keu, 81, 98-99; and Lancelot, 73-77, 82-83, 87, 95—96; in Cliges, 65, in Yvain, 92. See also Bademagu; Lancelot; Signs; Tournaments Guiette, Robert, 232^29, 264n.13, 265n.i4 Guigemar: characterization of, 164; and Marie de France, 161,181-82,183-84,195; parallels with his lady, 164-66. See also Naming; Signs, and process of recognition Guigemar, 162—63,179,181,182, 204; compared with Eliduc, 171-77,180; compared with Lanval, 159-60; compared with Partonopeu de Blois, 152; selection and substitution in, 163-70. See also Closure; Doubling; Signs. Guildeluec, 171,189; and Eliduc, 174-76; and Guilliadun, 173,175,176,187. See also Gender; Naming Guildeluec ha Guilliadun. See Eliduc Guilliadun, 171; and Eliduc, 71,173—76; and Guildeluec, 173,175,176,187. See also Gender; Naming Haidu, Peter, 62, 215, 227 n.6, 233^43, 236n.i5, 238n.5, 24inn.32, 35, 244n.67, 25in.6i, 254n.i2, 263n.2 Hamon, Philippe, 262n.7i Hanning, Robert, 111-12, 24on.2i, 246n.i6, 247n.23, 253n.3, 26onn.53, 58, 266n.34 Harley manuscript, 157,162,177,179,180, 206. See also Prologues Hilton, Catherine, 25on.48, 25in.6i Hirsch, E. D., 233n.39 Hodgson, Frederick, 261 n.64 Hoepffner, Ernest, 12, 31, 229n.2, 23onn.8, 15, 232^38, 233^48, 253n.5 Holmes, Urban Tigner Jr., 229n.2o, 25on.5i Horrent, M. J., 233^48 Hospitable Damsel, in Chevalier de la Charrete, 67, 68, 75, 209 Huchet, Jean-Charles, 255 n.28 Hue de Rotelande, 246n.i2 Hult, David, 241 n. 35 Hunt, Tony, 196, 228 n.14, 237^25 Husdent, in Folie Tristan d'Oxford, 13,14,15, 19, 23, 26

288

Index

Identity: of authors, 112,162-63,178-84, 205-6, 217, 22911.20, 25311.4; of characters, 5, 12, 18, 22-23, 70, 73, 76, 77, 82-83, 96, 136-38,176. See also Anonymity; Disguises; Knowledge; Truth; Voice Ille et Galeron, 160, 203 Interpretation, 10,58, 78,158, 201, 212, 221, 24on.2i; evaluation and judgment, 37, 41, 52, 59, 60, 83-84, 94,101-2,107-8, 210; judgment and choice, 109,113-14,126, 131-32,134,138-51,156, 25on.5i. See also Beauty; Signs; Truth Intertextuality, 2,5, 7, 9, 28-29, 31-60, 9094,161-62,163,171,177-78,198, 206, 210, 212, 217, 219, 242n.44, 46, 248n.27. See also Medieval textuality; Tristan legend Ipomedon, 203 Iseut, 91, 93,100,103; in Eliduc, 173; in Folie Tristan d'Oxford, 23-25, 32-34, 35-36; and Guenevere, 99; in Thomas's Tristan, 39-40, 46, 49, 52; death of, 42-43, 5o; and relation to narrator, 45-47; saving powers of, 50; separating from Tristan, 40. See also characters under Identity; Doubling; Signs; Voice Iseut aux Blanches Mains: in Folie Tristan d'Oxford, 31-32; in Thomas's Tristan, 5, 41, 46, 47-48, 49 Isidore of Seville, 26on.5i Jakobson, Roman, 228n.i3, 254n.i4 Jauss, Hans Robert, 3, 207-8, 227n.i, 262 n.i Jeu d'Adam, 133 Jodogne, Omer, 234n.i, 236nn.i4, 22 Jolivet, Jean, 221 Jolles, Andre, 102 Jonin, Pierre, 234 n.i Joris, Pierre-Marie, 248n.3o Kasprzyk, Krystyna, 233^48, 234^52 Keen, Maurice, 4, 212, 228n.9, 239nn.9,10, 264nn.io-i2 Kelly, Douglas, 54, 78,105,128, 22711.3, 23411.53, 237n.25, 26in.58; and literary truth, 117 Kennedy, Elspeth, 233^44 Kermode, Frank, 202, 262n.76 Keu, in Chevalier de la Charrete, 70, 72, 76, 81-82, 91,101; accused of treason, 101; and Guenevere, 98-99; and Rash Boons, 65, 81-82

Kibler, William W., 238n.i Knowledge, 14,17, 21, 26-27, 38-40, 7071, 74-75, 77, ni, 140,167, 169, 216, 22021; experience and identity, 5, 47-49,52, 56,59,139, 209-10, 226n.33; and power, 8,114-15,123-24,133-38,146,154 Knowles, David, 265n.28 Kohler, Erich, 244^73, 264nn.n, 12, 265^15 Koubichkine, Michele, 254 n. 17 Kristeva, Julia, 227n.4, 244n.77, 247n.27 Krueger, Roberta L., 24in.39, 242n.46; 245 n.6 La Coudre, 171,183 Lacy, Norris J., 265^23 Lais and fairies, no, 123-24,133,134,137, 248nn.32, 39, 249nn.43, 48, 250^54. See also Genre, romance and lais Lais (Marie de France), 8-9,58; triadic structure of, 201. See also Closure; Gender; Naming; Prologues; Selection and substitution; Signs; Tristan legend; Truth; Voice Lancelot, 92,107; and author/narrator, 209; in Chevalier de la Charrete, 6, 65-66, 91, 101,107; as Cart Knight, 72, 99; as erotic magnet for women, 67—68, 69; and Guenevere, 65, 66, 67, 68, 73-77, 78-80, 82, 87; herald's recognition of, 71-72; imprisonment of, 88—89; and issue of marriage, 67, 68-69, 70; measurement of, 60, 65, 82-83,101-2; and Meleagant, 61, 62, 76-77, 78, 83, 87, 96, 99-100; places of imprisonment, 61; rumors about, 72-73; singularity of, 69—70, 75; size of heart, 82-83; as touchstone of his society, 7; compared with Partonopeu, 126,151; incognito, 136; love of Guenevere, 95-96; love triangle, 95; as related to Tristan, 94-104. See also Anonymity; Disguises, incognito; Tournament Lancelot. See Chevalier de la Charrete Lancelot-Grail, 157 Lanval, 123,159,185, 200; and parallel characterization, 167 Latin and vernacular traditions, i, 7, 8, 29, 113,115,155,158-59, 194-95, 204, 246n.i2. See also Medieval textuality; Orality Laiistic, 152,167,171, 200 Le Bel Inconnu, 47,153,154 Le Gentil, Pierre, 50, 234n.i Le Rider, Paul, 240 n. 19

Index Le surplus, 57, 58-89, 194, 199, 223, 496-97. See also Love, and desire; Narration Levi, E., 253n.5 Lotman, Juri, 240 n.28 Love: and desire, 40,121-22,128,138,15152; and prowess, 63-64, 68-69, 77, 9597,120,130-31, 242n.5i; and reason, 56-57, 80,140-41,168. See also Substitution and substitutability, errance and novelerie Lutoslawski, W., 233^44 Maddox, Donald, 201, 238 n.5 Magic, 18-19,121,123; powers, no, 117,124, 127,129,132-33,134,137,139, 249n.45- See also Lais and fairies; Knowledge, and Power; Power, and sexuality Mai Pas episode (Beroul), 98 Margaris. See Sultan Marie de Champagne, 87, 94,104,105,106; and Chretien de Troyes, 84—87 Marie de France, 8-9, n, 28,32, 93,102103,106, 123, 157-206, 232n.34, 233n.46; Fables, 204. See also Author; Closure; Gender; Naming; Narrator; Orality; Prologues; Tristan legend; Truth; Voice. Mark: and Arthur, 98; in Folie Tristan d}Oxford, 13—19, 23, 25,32, 34; in Thomas's Tristan, 5,55, 97 McCullough, Florence, 228n.i5, 253 n.9 Medieval textuality, 2, 7, n, 30—31, 60, 90, 212, 215-16, 233n.44, 253n.9. See also Intertextuality Meleagant, in Chevalier de la Charrete, 72, 74, 87, 91, 96, 98-102. See also Lancelot Meleagant's sister, 88 Melior, 8. betrayal by Partonopeu, 122,124— 25; choice of husband, 119,120-21,13132,141—43,146—47; compared with Guenevere, 151; and Cupid and Psyche, 122—23; in role of lyric lady, 126—27; and the Sultan, 155. See also Beauty; Chef d'Oire; Gender; Fictionality in romance, history and romance; Disguises, invisibility; Knowledge; Lais and fairies; Power; Tournaments. Memory, 15,18, 22, 28, 29-30, 33-34,158, 169, 221, 229n.i8, 254n.2O, 257^48; and remembering, 14,179,180,191,197,199, 204, 206. See also Signs. Menard, Philippe, 200 Meriaduc, 165-66,167

289

Micha, Alexandre, 135-36, 242nn.42, 44 Mickel, Emanuel J. Jr., 253n.6, 261 n.64 Milun, 171,173-74,177, 200 Mise en abyme: as representation of feudal society in romance, 4, 6-7, 94-102, 22425; technique in narrative, 6-7,10 28, 6162, 65-70, 83, 87,103,105, 232n.34, 238n.8 Muret, Ernest, 230n.9 Naming, 63, 69,112,158,161,171,173,174, 253 n.6, 256 n. 35; and titles, 9,170,177-84, 204, 255 n. 31 Narration, bipartition and tripartition, 62, 125,152,161,165,172, 216, 238nn.5, 6; complexity of, 1-2, n, 103,152, 210-n; continuity and discontinuity, 9, n, 88,158, 160-62,188, 215, 217-18, 219. See also Intertextuality; Medieval textuality Narrator: and author, 2, 8, 9, 85—86, 90—91, 112,180, 207-11, 218, 24in.32, 263nn.2, 6; interventions of, 30, 31, 45, 48-49,59, 66, 72, no, 121-22, 128, 139-41, 14446,147,150,159,190-91, 200, 210, 218, 25on.545; and public, 5, 7, 9,10,15, 49-52, 72,109, iii-i2,114, 207-8, 24in.32; and story, 5, 7-8, 9, 28, 30, 45-49, 52, 56, 94, 109, in-12,128,144-45,147,149-50, 15355, 177, 180-81,188, 207-13, 25on.52; in the story, 9-10, 63, 88, 91. See also Author Neumann, Erich, 245 n.4 Newstead, Helaine, 248nn.32, 34, 249n.48 Nichols, Stephen G., Jr., 246n. 13 Noble, Peter S., 240n.30 Oilier, Marie-Louise, 229n.17, 238n.8, 239n.i8, 252n.2 Ong, Walter, 24, 231 nn. 18—20, 22, 233 n.40 Orality, 23, 88,191; oral storytelling, 9,15758,163, 179-80, 182, 184,1912,194,199, 201, 205, 26on.55; and written narrative, 2, 4, 8, 24-25, 27-30, 55-56, 93, 157-62, 177-78,180-81,188, 189-99, 201-2, 2046, 219, 24in.36, 257nn.45, 47, 48. See also Medieval textuality; Voice Oxford Folly. See Folie Tristan d'Oxford Painter, Sidney, 239n.io Partonopeu: compared with Adam, 133; compared with Enide and Guenevere, 136; compared with Lancelot, 126,151; compared with Perceval (Conte du Graal), 124; as counterpart to Sultan, 131,

29O

Index

Partonopeu (continued} 146; debate with Anselot, 144; and Hector, 118-19; and Melior, 137-38; and Persewis, 130-31. See also Beauty; Gender; Identity; Disguises, incognito; Knowledge; Power; Tournaments Partonopeu de Blois, 11, 109-56; betrayal compared with Tvain, 125-26; compared with Chevalier de la Charrete, 7,136,15051; compared with Chievrefoil, 152; compared with Guigemar, 152; compared with Laustic, 152; compared with the Adam and Eve story, 138-39; and Garden of Eden, 114; genealogy in, 115-16; narrative structure of, 152; summary of romance, 109-10; as variation of Cupid and Psyche story, 8, no, 132. See also Closure; Continuation; Gender; Truth Passe-Rose, 111-12,147,155 Payen, Jean-Charles, 215, 232^29, 262n.68 Pelan, M., 253n.5 Perceval, 3, 92 Perceval Continuations, 8 Perceval (Le Conte du Graal), 86, 92,106, 124,153-54 Persewis, and Partonopeu, 130-31,136,152 Pesme-Aventure (Tvain), tournament episode, 62 Pickens, Rupert T, 24in.32, 25in.6i Piramus, Denis, 178,184,194,196; and truth in lais, 192 Point of view, 30, 37, 42-45,59, 63, 69, 8384, 87, 96-97,121,128-30,135,147,167, 170,174,182-83, 220, 222-23, 240n.24. See also Narrator; Signs Power, 126-33,147-50, 222; and sexuality, 112,134,136-37,148. See also Knowledge Prologues, 7, 84, no—n, 113,115,120,127, 148,158,159,182,184,191-92, 202, 208, 209, 210, 242n.43, 24411.73, 246n.n, 255 n.32; and epilogues, 5, 7, 9, 28, 85,112, 160,162,179,180,190, 201—4, 218, 256 n. 35; Marie de France's General Prologue, 8, 157,159,162-63,177,179,180-81,187,188, 189-90,193-97, 202, 204, 26in.62. See also Epilogues Prose Lancelot, 82

Queen. See under individual character names

Reader reception and response, 20, 214. See also Author, and public; Narrator, and public Regalado, Nancy Freeman, 23in.24 Reid, T. B. W., 23on.i4 Reiss, Timothy J., 228n.i2 Renarration, 5, 8, 28,152,169-70,186-88, 189, 221, 232 n. 34, 235n.u; in the Folie Tristan, 13-16,18, 22, 23, 32-33, 36, 217 Renart stories, 25-26 Renaut de Beaujeu (LeBellnconnu), 47,153 Repetition and variation, i, 65-68, 75-77, 84-85, 103, 105, 107-8, 153, 166, 169, 198, 208, 215, 216. See also Medieval textuality Robertson, D. W., Jr., 228n.9, 230n. 12, 243n.66 Robertson, J. Duncan, 230 n.6, 233nn.44, 46 Romance public, 8,33,34,114-15,143,144, 211-12, 223-24, 264n.i3. See also Author, and public; Narrator, and public; Reader reception and response Roman de Brut. See Wace Roman de la Rose, 154 Roman de Thebes, in Romans antiques, i, 7,153,159, 203, 212 Roques, Mario, 23in.25, 237n.i Rychner, Jean, 185, 2520.1, 253 nn.3,5, 255 nn.31-32, 257n.46, 26on.53 Sargent-Baur, Barbara Nelson, 2420.43 Schmid, Elisabeth, 246 n. 15 Schmolke-Hasselmann, Beate, 238 n.2 Selection and substitution, 163—78,193, 197-98, 217, 254 n. 14. See also Doubling Signs, 15, 220-22, 247n.27, 2650.26; ambiguity and multiplicity of, 13,15,16-17, 20-21, 23, 34, 74, 88-90, 97-98,108, 214—15, 220, 222—23; ^d belief, 14—15, 18-19, 22, 24, 27, 76-77, 98, 106, 115, 221, 2400.24; disjunctions between words and meaning, 7,13-14, 64, 74, 76,106,136; disjunctions between words and referent, 7, 16, 72, 74, 99-100, 220, 230n.9; interpretation of, 2, 4, 6, 8, 13, 15,17-19, 24, 3334, 98,100,102, 209, 215-16, 219-20; manipulation of, 15-16, 20-21, 82-83, 86, 89, 222; and process of recognition, 2, 8, 12, 18, 19—22, 27, 35, 70—77, 167—70, 171, 220-21, 265n.28; as proof, 14-15, 21-23, 97-98, 99-100, 230n.12; types of, 13,

Index 18,19, 21, 24—25, 220. See also Truth; Voice Smalley, Beryl, 2610.65 Smith, Barbara Herrnstein, 2620.71 Smith, Julia M. H., 194, 26000.55,57 Smith, Leoo, 245 n.2 Soeyders de Vogel, K., 2450.2 Soredamor (Cliges), 65, 91 Speer, Mary Blakely, 2300.14 Stefaoioi, Jeao, 2340.52 Steveos, Martio, 2270.2, 2630.6 Stock, Briao, 2330.40 Storey, H. Wayoe, 2620.75 Strayer, Joseph, 2640.11 Sturges, Robert, 2600.58 Substitutioo aod substitutability, 98,134, 150,177; ermnce aod novelerie, 5, 31-32, 41-42, 45, 46, 57-58,147; mobilizatioo of categories, 8,146-47 Sultao, 147; as author, 149-50; as couoterpart to Partooopeu, 141,142,146; aod Melior, 149,155 Taotris, io Folie Tristan d'Oxford, 14, 23, 26 Thomas d'Aogleterre, 4,5, n, 29, 93, 97—98, 188, 215-16; compared with Folie Tristan d'Oxford, 31; knowledge of Breri versioo, 56; oarratioo compared with Gottfried, 48; regardiog truth of versioos, 30. See also Doubliog; lotertextuality; Le surplus; Tristao legeod; Truth Touroaroeot of Noauz: io Chevalier de la Charrete, 61-77, 87; compared with Cliges, 69; compared with Erec aod Tvain, 62; as preteod war, 64; as reprise of Cart sceoe, 65. See also Mise en abyme; Sigos; Touroameots Touroameots, 170, 222, 2380.9, 23900.10, n, 18; io Chevalier de la Charrete, 15, 60—70, 107,130,151; io Partonopeu de Blois, no, 120,126-27,130-32,134-37,143,146,147, 151,152, 24911.43 Triodade, W. Aoo, 2350.2 Tristao, 3, 91, 95, 98, 99,100,103; io Folie Tristan d'Oxford, 26-29, 35-36; dialogue with marioers, 31; io Folie Tristan d'Oxford, 26-29, 35-36; as fool, 13-14; aod Goveroal, 55; aod Husdeot, 33; aod Iseut aux Blaoches Maios, 39, 40, 46; Iseut's ioterpretatioo of, 18-19; as leper, 54; as

291

Saiot Alexis, 50,54; as Taotris, 14, 23; testiog Iseut's love, 17, 20-23; in Thomas's Tristan, 40,57; as "Tristao PAmerus," 38—39; aod uoity with Iseut, 41—45. See also Disguises; Ideotity; Narrator; Reoarratioo; Sigos; Substitutioos, errance aod novelerie; Tristao legeod Tristan (Beroul), 14, 97 Tristao (Chiewefoil), 188 "Tristao 1'Amerus," 12,19, 22—23 Tristao legeod, 2, 4, 6, 8, 28, 30-33, 34, 36, 37, 53-58,151,155, 210, 216-17, 218, 219, 22800.8,17, 22900.1, 3, 4, 2310.24, 23200.34, 38, 23300.46, 48, 2340.51; aod Chretieo de Troyes, 60, 91-92, 93, 94104, 2430.56, 2540.23; aod Marie de Fraoce, 159,171,173 Tristan (Thomas), 4-5, n, 32, 37-59; compared with Beroul's versioo, 54; compared with Folie Tristan d'Oxford, 5-6 Truth, 5, 30, 45, 53, 55~56, 58, 106,174, 2420.47, 2440.68, 2470.22, 2650.14; aod adveoture, 188-89,191-93,196,197-99, 199—200, 201; appearaoce aod reality, 20, 22, 71, 78-82,106,107,122,154,168, 220; corporeality of, 30, 34-36, 2340.52; aod lies or deceptioo, engin, 8,13, 20, 26—27, 72-73, 76, 85, 93, 99-iQi, 113, H7-I9,135, 138-39,140-41,155,187-88, 220, 223; aod love, 5, 15, 17,19-20, 22, 30, 34-36,13943, 222; aod meaoiog, 104—7,191,193— 94,196-97, 201, 224-25, 24400.72, 76; aod poiot of view, 107,141-43. See also Beauty; Fictiooality io romaoce; Sigos; Voice Uitti, Karl D., 2440.76, 2600.52 Uoity aod cohereoce or diversity, 2, 33,5354, 57-58,133,158,161-62,171, 198, 210, 215-16, 218, 2530.10. See also Truth Urraque, no, 123; descriptioo of, 129; maoipulatioos of, 113; aod Melior, 154; aod Partooopeu, 126,135,136,152 Vaoce, Eugeoe, 29, 2280.13, 2450.77, 2470.22, 2500.50, 2630.2 Vitz, Evelyo Birge, 2340.1, 2570.45, 2610.67 Voice, 4-5, 9,12-14,19, 22-34, 36, 52, 2310.20, 2320.31, aod Marie de Fraoce's

292

Index

Voice (continued) Lais, 158-59,163,180,184-89,190,192, 198—99, 205—6. See also Orality; Truth Wace (Roman de Brut), 85,104,105,113,115, 159; stories of Arthur as mixture of truth and lies, 85,115,192 Wagner, R. L., 234 n.i Walters, Lori, 254 n. 13 Warning, Rainer, 237 n.34, 241 n. 35 Wetherbee, Winthrop, 244n.68, 265^27 Wind, Bartina H., 229n.2o, 235^4, 236n.i8 Tonec, 117,171,182, 200, 202; compared with Eliduc, 173-74,177; and parallel characterization, 167

York, Ernest C, 230 n.9 Yvain (Chevalier au Lion), 10 Tvain, 84, 89, 90—94, 92,153; betrayal compared with Partonopeu deBlois, 125-26; and Chevalier de la Charrete, 62, 88-89; marriage in, 97; mention of Guenevere, 81 Zink, Michel, 236n.2o, 245 n.6 Zumthor, Paul, 93, 200, 215, 216, 227n.6, 228n.io, 263 n.4; on romance and history, 116

This page intentionally left blank

This book has been set in Linotron Galliard. Galliard was designed for Mergenthaler in 1978 by Matthew Carter. Galliard retains many of the features of a sixteenth-century typeface cut by Robert Granjon but has some modifications that give it a more contemporary look. Printed on acid-free paper.