From Song to Book

Front matter

`a rom Corg to V to Book

THE POETICS OF WRITING IN OLD FRENCH LYRIC AND LYRICAL I NARRATIVE do: tail

SYLVIA HUOT

Part One: On The Nature Of The Book In The Thirteenth And Fourteenth. Centuries

1. Scribal Practice and Poetic Process in Didactic and Narrative Anthologies 11

Examples of Thematic Unity: MSS Bibl. Nat. fr. 24428 and 12786 12 Examples of Narrative Organization: MSS Bibl. Nat. fr. 1447 and 375 19 Compilatio and Conjointure: The Poet-Compilers of MSS Bibl.

Nat. fr. 1450 and 1446 27 The Voice of the Scribe: Rubrication in MS Bibl. Nat. fr. 25545 36 The Figure of the Author in a Thirteenth-Century Codex: MS Arsenal 3142 39

2. Scribal Practice in Lyric Anthologies: Structure, Format, and

Iconography of Trouvére Chansonniers 46 A Survey of Chansonnier Format 47 Knights, Poets, and Performers: The Iconography of the Trouvères 53 The Arrangement of an Author Corpus: Adam de la Halle and MS Bibl. Nat. fr. 25566 64 The Iconography of Lyricism in the Alphabetical Chansonnier O 74

Contents

Part Two: Lyricism And The Book In The Thirteenth Century

3. Singing, Reading, Writing: Guillaume de Lorris, Jean de Meun, and the Manuscript Tradition of Le Roman de la rose 83

"Lyric" and "Narrative": An Overview 84

Lyrical Writing in Guillaume de Lorris: The Poetics of

Narcissus 86

L'Aucteur and l'Amant: Lyrico-narrative Discourse in Le Roman

de la rose go

Lyrical Writing in Jean de Meun: The Poetics of Pygmalion 96

The Rose as Book and the Question of Vernacular Poetic

Authority 99 4. Text as Performance, Text as Artifact: Contrasting Models

for the Romance with Lyric Insertions 106

An Anthology of Performance: Jean Renart’s Roman de la rose 108

Biography of a Lyric Persona: Le Roman du castelain de Couci 117

The Material Text and the Reliquary of the Heart 125

Lyrical Narrative and Lyric Compilation: From Performance to

Book 132

5. The Audiovisual Poetics of Lyrical Prose: Li Bestiaire d'amours and Its Reception 135 Lyricization of the Bestiary: Guillaume le Clerc and Richard de Fournival 136 Parole and Painture: Richard’s Audiovisual Poetics 138

Song, the Written Word, and the Word Made Flesh: Procreation and Incarnation in Li Bestiaire d’amours and Le

Roman de la rose 141 Amplification of Li Bestiaire d’amours: Responses and

Continuations 148 Lyricization of Li Bestiaire d'amours: Le Bestiaire d'amour rimé 158 The Iconography of Parole and Painture 164

6. Lyrical Writing and Compilation in Le Roman de la poire and

Le Dit de la panthére d'amours 174

Tibaut's Lyrical Romance: From Book to Nightingale and Back

Again 174 Romance as Book: Codicological Analogues for the Poire 184 Le Roman de la poire and the Evolution of Lyrical Writing 189 Le Dit de la panthére d'amours as a Literary Compilation 194 Performance and Book: From Carol to Lyric Compendium 201 Lyrical Writing in Le Dit de la panthére d’amours 204 vi

Contents

Part Three: Lyricism And The Book In The Fourteenth Century

7. The Vernacular Poet as Compiler: The Rise of the Single-

Author Codex in the Fourteenth Century 211

"Rustebuef qui rudement oevre": A Thirteenth-Century Author

Corpus 213 Transformations of Le Roman de la rose: Baudouin and Jean de

Condé 219 Watriquet’s diz en escrit: The Minstrel as Compiler 224 Machaut’s Books: The Conjoining of Song and Dit 232 The Compilations of Jean Froissart 238

8. From Song to Book in an Early Redaction of the Oeuvre of

Guillaume de Machaut: The Codex Bibl. Nat. fr. 1586 242 The MS Bibl. Nat. fr. 1586: General Description and Background 243 Le Remede de Fortune: A Poetic and Iconographic Analysis 249 The Iconography of the Lays in MS 1586: Carol or Compilation? 260 Conclusion 272 g. A Late Redaction of Machaut’s Oeuvre: The Codex Bibl. Nat.

fr. 1584 274 Le Remede de Fortune in MS 1584 275 Le Livre du voir dit: A Mise-en-Scéne of Lyrical Writing and Compilation 280 Performance and Book in Le Dit de la harpe 286 Writer, Lover, and Performer in La Fonteinne amoureuse 293

10. The Poetics of Lyrical Writing in the Works of Jean

Froissart 302 Froissart's Prologue: Le Paradis d'amour 303 Dream, Image, and Poem: Froissart Responds to La Fonteinne amoureuse 305 Midpoint of Froissart’s Rook: La Prison amoureuse 311 Lyricism and Escapism in Le Joli Buisson de jonece 316 Froissart’s Epilogue 323

Conclusion 328

Appendix A: The Rubrication of Guillaume de Lorris in MS Bibl. Nat. fr. 378 339

Appendix B: Table of Miniatures in Selected Texts by Machaut, MSS Bibl. Nat. fr. 1584 and 1586 343 vii

Contents

Appendix C: Excerpt from an Unedited Volume of Le Roman de Perceforest, MS Bibl. Nat. fr. 346 347

Bibliography of Works Cited 351 Index 865 vill

ACKNOWLEDGMENTS

One of the greatest pleasures of extended research is the opportunity that it affords for exchange with other scholars, and I feel particularly fortunate in the number of people who have offered advice and moral support along the way. My first thanks go to Karl D. Uitti and Michael J. Curschmann of Princeton University, who as directors of my dissertation presided over my initial investigations of medieval lyric and lyrical narrative poetry and its manuscript tradition. My debts to both of them are considerable. Among the many other people who have contributed to my thinking during the years I have spent with this work, I thank Emmanuèle Baumgartner, Kevin Brownlee, Peter Dembowski, Nancy Freeman Regalado, Lori Walters, and Winthrop Wetherbee; the book is significantly richer for the questions, comments, and corrections they have offered. Robert F. Cook made many very helpful suggestions as reader for Cornell University Press. And to Leo Krumpholz I owe a very special thanks for support and encouragement during all phases of this project.

Research for this book was conducted in many American and European libraries and could not have proceeded without the helpful cooperation of librarians and other staff members of the following institutions: the Biblioteca Apostolica Vaticana; the Biblioteca Laurenziana and the Biblioteca Riccardiana, Florence; the Bibliothéque de l'Arsenal and the Bibliothéque Nationale, Paris; the Bibliothéque Municipale, Arras; the Bibliothéque Municipale, Dijon; the Bodleian Library, Oxford; the British Library, London; Firestone Library, Princeton University; the Musée Condé, Chantilly; the Newberry Library, Chicago; the Pierpont Morgan Library, New York; Regenstein Library, the University of Chicago; and the Walters Art Gallery, Baltimore. I have also made use of the invaluable research facilities of the Institut de Recherche et d'Histoire des Textes, Paris, a branch of the Centre National de Recherche Scientifique. I am grateful to the Hill Monastic Manuscript Library, St. John's Abbey and ix

Acknowledgments

University, Collegeville, Minnesota, for providing a microfilm of MS 2609 of the Nationalbibliothek, Vienna.

This study was supported in part by a Fellowship for Independent Study and Research from the National Endowment for the Humanities, which enabled me to devote full-time efforts to the final preparation of the typescript. I am especially grateful for this generous support.

SyLviA HUOT DeKalb, Illinois

FROM SONG TO BOOK

I began this book with a seeming paradox: medieval literature is at once more oral and more visual than a modern printed book. The medieval manuscript, with its colorful initials, miniatures, and decorative margins, appeals to the eye as an object of visual delight. Yet the texts that it contains are designed to appeal to the ear—lyric poems set to music, narratives that address a listening audience. In a literary tradition conceived for oral delivery, what is the role of the book?! And as the practice of silent reading grew during the later Middle Ages, what was its effect on poetic practices??

These are large questions and have been treated from various perspectives by a number of scholars.? I have limited myself to French literature of the thirteenth and fourteenth centuries and have focused on courtly lyric and lyrical narrative poetry. I have chosen this material because the relation of lyricism and writing is a particularly interesting case in point, and an examination of the lyrico-narrative tradition reveals a well-defined movement from a more performative toward a more writerly poetics. My approach combines poetic, iconographic, and codicological analysis, aimed at describing the relationship between poetics and manuscript format. Poetic analysis will focus on the thematization of writing and performance, on the figure of the poet as singer or writer, and on the transformations and modifications of lyric poetics that each author in turn effects. My readings are based on a study of the texts in manuscript—in each case I have examined as many manuscripts of the text as were accessible to me—and aim to provide a reading not merely of the poem but of the poem as manifested in book form.

I employ the term “lyrico-narrative” not as a specific generic designa-

1. See Crosby; Zumthor, Poésie et la voix.

2. See the study by Saenger.

3. For some works that have provided broad historical and conceptual bases for the present study, see Chaytor, Clanchy, Ong, Stock, and Zumthor in the Bibliography.

tion but rather as a loosely defined but useful category, comprising texts on the topic of love, with first-person narrative discourse, lyric insertions, or both. For the purposes of this study, the term "lyric" refers to the courtly love lyric, or chanson courtoise. 1 freely acknowledge that this represents an arbitrary exclusion of such important lyric genres as the political sirventes, the Crusade song, religious lyric, or the frequently ironic pastourelle. I have limited myself in this way so that the remaining body of literature, which is still quite large, may be treated in some degree of depth. I have, moreover, adapted the term “lyrical” to refer not only to nonnarrative poetry in stanzaic form but also to narrative poetry in octosyllabic couplets (romance or dit) and even prose works that share certain thematic and grammatical features of the chanson courtoise. For example, the first-person identification of author and protagonist is common to lyric poetry and to such works as the narrative Roman de la rose and the prose Bestiaire d'amours. The dedication of a work to the narrator's lady, rather than to a patron, further contributes to the lyrical quality of many a romance and dit, as does a focus on the narrator's amorous adventures. And the lyrical narrative manifests the self-reflective quality of courtly lyric. For the trouvére, to sing and to love are complementary facets of a single activity, and the song records the event of its own making. Similarly, a lyrical romance or dit frequently records the event of its composition or transcription, and sometimes both, portraying poetic composition as a form of love service.

This transposition of lyric thematics and discourse into narrative format poses certain paradoxes. Most obvious is the conflict between narrative progression and resolution on the one hand, lyric stasis and openendedness on the other. This issue has been explored in recent work on the Roman de la rose.* Equally important is the issue of medium. Whereas the twelfth-century romance narrator explicitly presented himself as heir to a long-established written tradition, the twelfth-century trouvére was explicitly a singer, whose songs derived from his own personal experience rather than from books. The evolution of thelyrical romance and dit entailed a progressive redefinition of lyric poetry as a written medium and of lyric discourse and thematics as appropriate to a writerly narrative format, as well as the identification of an explicitly written literary tradition conjoining lyric and narrative poetics.

To trace a movement from performative to writerly poetics requires not only close attention to poetic conventions but also an understanding of the status of vernacular writing in the later medieval period. It must be remembered that, throughout the medieval period, writing retained a certain dimension of orality, being understood as the representation of speech. Not only poets but even rubricators of manuscripts appropriated the language of oral declamation. Even in the late fourteenth cen-

4. See Hult, Self-Fulfilling Prophecies; Lejeune; Strohm; Verhuyck; Vitz.

tury, writing could still be described as the pale imitation of an oral original. For the translator Jean Daudin, for example, the movement from speech to script provided an analogy for linguistic translation. In the prologue to his translation of Petrarch’s De remediis unius utriusque fortunae, dedicated to Charles V, Daudin regretfully comments that eloquence is lost in translation, just as in the case of the eloquent Cicero, Demosthenes, Homer, and Virgil, “quant on lit aucun d'eulx et on ne les ot parler, une grande partie d'eulx lui est defaillant" (when one reads any of them and does not hear them speak, he is missing a good deal).5

As the visual representation of an essentially oral text, the medieval illuminated manuscript has a certain theatrical—at the risk of anachronism, one might even say cinematic—quality; it does not merely describe events but, rather, stages them. The performative quality of the medieval book is of profound importance, and I will have occasion to refer many times to this idea in the course of the present work. Writing in the second quarter of the thirteenth century, Richard de Fournival testifies to the theatricality of the illustrated book in the prologue to his Bestiaire d'amours. Commenting on the fact that the Bestiaire is constructed of speech and illustrations—parole and painture—Richard explains that the combination of the two allows for a vivid auditory and visual experience of that which is depicted: “Car quant on voit painte une estoire, ou de Troie ou d'autre, on voit les fais des preudommes ke cha en ariere furent, ausi com s'il fussent present. Et tout ensi est il de parole. Car quant on ot .i. romans lire, on entent les aventures, ausi com on les veist en present." (For when one sees a story illustrated, whether of Troy or of something else, he sees the actions of the worthy men that lived in those times, just as though they were present. And it is just the same with speech. For when one hears a romance read [aloud], he follows the adventures, just as though he saw them before him.)9

In Claris et Laris, which postdates the Bestiaire d'amours by about thirty years, Claris is described as witnessing the events that he reads of in a book of love stories: "En .I. petit livre veoit / La mort Tibé et Piramus" (In a little book he saw the death of Thisbe and Pyramus [Alton ed., vv. 162—63]). And the analogy between theater and the illuminated book is still apparent in fifteenth-century English defenses of the mystery plays, in which the dramatic performance is referred to as a “living book."?

The poets that we will examine exploited this performative quality of

5. This prologue is printed by Delisle in his "Anciennes traductions françaises”; I quote from p. 294. I am grateful to Peter Dembowski of the University of Chicago for drawing this passage to my attention.

6. Segre ed., p. 5. References to editions of works discussed are to those listed in the Editions section of the Bibliography. The first citation of a work gives the editor's last name, "ed." to indicate that the reference is to an edition, and the line or page numbers. Except where ambiguity would result, later references to the work give only line or page numbers. All translations are mine.

literature in general and lyricism in particular in various ways: some explored the notion of writing as a visual projection of song or speech, whereas others stressed the inherent differences between performance and writing. Overall, it is possible to document a general shift of focus, in the later thirteenth and fourteenth centuries, f rom lyric performance to lyric composition, with the latter defined ever more insistently as an act of writing rather than one of song or declamation. Romances and dits with lyric insertions tend increasingly to recount the genesis of the lyrics in question, rather than to describe their performance. This concern with composition is associated with a more writerly concept of the song as specifically referential, documenting a particular experience; the fiction of many a fourteenth-century dit amoureux is that of the poet-lover who uses both lyric and narrative verse forms to record, in writing, the vicissitudes of his love. As the lyric voice is assimilated to that of the narrator or writer, a new poetics is defined.

In order to designate the act of poetic creation as it is represented in lyrico-narrative texts of the thirteenth and fourteenth centuries, I have introduced the term “lyrical writing." The concept of the “lyrical writer" is exemplified in the frontispiece illustration, a portrait of the troubadour Folquet de Marseille taken from a late thirteenth-century anthology of Provençal verse copied in Italy. The poet, seated at a writing desk, holds his pen aloft and gestures expansively as though on the verge of bursting into song: lyric composition is conceived at once as an act of writing and as an inspired performance.? The image reflects an understanding of writing as the visual representation of speech (or song); indeed, this vision of the lyric poet captures the metaphor expressed by the Psalmist, "My tongue is the pen of a ready writer" (Ps. 45:1). The thirteenth-century image, however, is the opposite number of the biblical text: whereas the psalmist is a singer whose song is a figurative script, the lyric poet is a writer whose script is a representation of song. The gradual establishment of the writer's primacy over the singer will be a central theme of this study.

The lyric and lyrico-narrative tradition plays a crucial role in the emergence of a "book" culture and the definition of the vernacular poet as a writer. Central to Old French lyrico-narrative poetry is the Roman de la rose, in which the process of writing is thematized and foregrounded as a creative, rather than merely duplicative, activity. At the same time, the act of writing the Roman de la rose is presented as inspired love service, ordinarily a lyric characteristic. Profusely illustrated, carefully rubricated, the Rose was received as a book, and an important aspect of its pervasive influence on subsequent French literature is its mapping out of a poetics at once lyrical and writerly. Contemporary with Jean de

8. On analogies between speaking and writing, see Camille, "Seeing and Reading."

Meun's continuation of the Rose are the earliest surviving single-author anthologies in the French secular literary tradition, and these are devoted to either songs or dits: first-person and frequently lyrical poetry. The growth of this lyrico-writerly poetics, centered on the lyric poet as writer and author of books, reaches new heights in the fourteenth century, in the carefully arranged anthologies of lyric and lyrico-narrative verse produced by Guillaume de Machaut and Jean Froissart.

In order to discuss the poem or poetic corpus as a book, and the poet's appropriation of bookmaking processes, it is necessary to establish an understanding of just what a vernacular literary book was in the thirteenth and fourteenth centuries. To this end, I begin with a detailed examination of selected codices, and a discussion of the relationship between poetics and manuscript format and between scribal and poetic processes.? I proceed from the assumption that although some codices are disorganized miscellanies, a great many— perhaps the majority—are carefully organized literary constructs. The scribe responsible for the production of the book played a role that combined aspects of editor and performer; in the words of James Farquhar, the scribe was the “organizer and metteur-en-scène of the book."!? In examining the architectonics of the manuscript and the poetics of the visual text —comprising poem, rubrics, and miniatures and other decorative elements— I wish, first of all, to establish the paratextual and codicological features that would be manipulated by poet-compilers and poet-writers who conceived of their works as books. Additionally, I wish to address the presence of poet and scribe, respectively, and to distinguish the role associated with each.

In establishing the affinities between scribal and poetic practices, I have drawn on concepts that have already been identified and discussed by others. Daniel Poirion has stressed the importance of auctoritas, translatio, and conjointure as modes of cultural transmission in the Middle Ages.!! These principles operate at the level of both poem and codex. The authority of a classical or vernacular poet can be evoked either through a grouping of his works within an anthology or by a citation within a single text. The poet's blending of material drawn from different texts or different literary traditions—what Chrétien de Troyes called a "molt bele conjointure" (very beautiful conjoining [Erec et Enide, Roques ed., v. 14])—is analogous to the scribe's artful comfilatio.!? The appropriation of classical learning and culture by the Christian Middle

9. For general background, see Kleinhenz; Hindman and Farquhar. For a discussion of the scribal role in an analogous process, the transformation of the previously oral chanson de geste into a written tradition, see Delbouille; Tyssens.

10. Hindman and Farquhar, p. 66.

11. Poirion, "Ecriture et ré-écriture," p. 118. 12. On compilatio, see Minnis; Parkes. On conjointure, see Freeman, Poetics of "Translatio Studi” and "Conjointure"; Kelly, "Sens" and “Conjointure.”

Ages designated in the term translatio studii et imperii is practiced by poets, translators, and compilers alike.!? As we will see, a scribe could effect a conjointure through a suggestive coupling of texts; he could exploit the blend of Christian and classical materials in courtly romance in order to build a bridge, within a diverse anthology, from romances of antiquity (romans antiques) to hagiography. These concepts, normally associated with a clerkly narrative tradition, undergo certain modifications as they are appropriated into a lyric tradition that is rapidly developing its own sense of learnedness. The compilation of anthologies devoted solely to the career of a single vernacular poet entails a new kind of poetic authority, new principles of textual conjoining. In a very real sense, such compilations are authorized by Jean de Meun's enunciation of a lyric translatio that places the medieval lyric or lyrico-narrative poet in a written lyric tradition extending back to Ovid and his predecessors, Tibullus, Catullus, and Gallus.

Following the first two chapters, devoted to narrative and lyric anthologies, respectively, I will turn to the examination of individual texts. After studying a series of thirteenth-century texts in Chapters 3—6, I will return to an examination of manuscript format in Chapter 7, a study of the emergence of the single-author anthology codex in the late thirteenth and fourteenth centuries. In the last three chapters I will examine the anthologies of Machaut and Froissart.

My analysis of both codices and texts focuses on scribes, compilers, illuminators, and poets as participants in and readers of the Old French literary tradition. Admittedly, the finer points of the compiler's craft were undoubtedly lost on a certain portion of the medieval audience, who would have experienced the texts only as they were read aloud, one at a time, and who would have had no recourse to the visual elements of the book. Even the person who commissioned the book did not necessarily concern himself or herself with the literary consciousness that informed the work of scribes and artists (or the workshop master who supervised them). Poets, who frequently state explicitly that they came to produce this or that romance after having read a particular Latin book or perused their patron's library, constitute a somewhat special case. The shared concerns of poets, scribes, and illuminators may well reflect a literate consciousness of the written medium and its creative potential that was appreciated, initially, by only a small percentage of the medieval audience.

On the other hand, the sharp increase in vernacular manuscript production in the thirteenth and especially the fourteenth centuries, and the careful arrangement and execution characteristic of so many, argue

13. On translatio studii, see Freeman, Poetics of "Translatio Studii” and "Conjointure"; Kelly, "Translatio Studi"; Uitti, Story, Myth, and Celebration, especially pp. 134, 146—51, 204—5.

for a growing literacy rate and an increasing demand for books that are more than mere repositories of texts to be read aloud. The popularity of acrostics and anagrams in late thirteenth- and fourteenth-century poetry further suggests an audience that delights in the manipulation of the written word. And it is impossible to imagine that the owners of illuminated manuscripts would never have looked at the pictures for which they had paid so much, even if they were unable to read a single word. Through a visual appreciation of author portraits, recurring iconographic motifs, and general page layout, even an illiterate "reader" could have been conditioned to certain concepts of authorship, codicological continuity, or literarity, which in turn might inform his or her appreciation of texts received aurally.

Although specific information about literacy rates in the thirteenth and fourteenth centuries is difficult to come by—let alone information about the owners of particular manuscripts, virtually none of whom can be identified at all—the texts themselves offer some evidence for lay literacy. Even in the twelfth century, certain romance couples—Lavinia and Eneas, or the lovers of Marie de France's Milun—correspond in writing, and the practice recurs in various texts of the thirteenth and fourteenth centuries: the Roman du castelain de Couci, the Livre du voir dit. In Yvain, a young girl reads a romance to her parents. Froissart, in L'Espinette amoureuse, claims that his lady first won his affections by reading to him from Cléomadés. The attribution of literacy may be meant to confer a special status on the character in question, but surely some lay men and women did know how to read and write, did read books to themselves and to one another, and therefore could have been sensitive to scribal or poetic manipulation of codicological elements.

We will probably never know the full story of the reception of medieval literature by its general audience, given the lack of detailed information about the production and consumption of oral and written works alike. In the end, we are left with the texts and the books themselves as our primary evidence, and if these do not render up the secrets of their medieval owners, readers, and listeners, at least they may have something to tell us about their authors, copyists, and illustrators. It is primarily with the latter group that this book is concerned.

Part One On The Nature Of The Book In The Thirteenth And Fourteenth Centuries

Et je vous monstrerai comment cis escris a painture et parolle. Car il est bien apert kil a parole, par che ke toute escripture si est faite pour parole monstrer et pour che ke on le lise; et quant on le list, si revient elle a nature de parole. . . . Et meesmement cis escris est de tel sentence kil painture desire.

[And I will show you how this writing has illustration and speech. For it is quite clear that it has speech, since all writing is made in order to show forth speech, and in order to be read; and when it is read, it reverts to the nature of speech. ... And similarly this writing is of such a topic that it desires illustration.]

Richard de Fournival, Bestiaire d'amours

Chapter 1

Scribal Practice and Poetic Process in Didactic and Narrative Anthologies

The subject of vernacular codex organization and production in the thirteenth and fourteenth centuries is a vast and largely unexplored area; the following discussion makes no claims at being comprehensive. Its aim is to identify certain organizational principles typical of French literary codices of this period and certain aspects of manuscript format and text presentation. The implications of these codicological features for thirteenth- and fourteenth-century poets will be discussed in subsequent chapters; before a given poet's manipulation of his work as a written medium can be assessed, a certain understanding must be established of what a vernacular literary book was in the later Middle Ages and how it functioned as a poetic system in its own right. In this chapter, I set forth some basic avenues of inquiry into the architectonics of the codex and the poetics of the manuscript text.

Many medieval codices are miscellanies, containing a seemingly random mixture of didactic, courtly, and bawdy texts in prose and verse;! others, at the opposite end of the spectrum, contain but a single text. Even in those having more uniform contents, it is not always possible to discern any logic to the order of pieces. Many codices, however—many more, I suspect, than are currently recognized—are organized according to principles ranging from rudimentary groupings of thematically related texts to an elaborate overall design. Similarly, whereas some are decorated for purely ornamental and even ostentatious purposés and others lack any decoration at all, many are true "critical editions," with carefully designed programs of rubrication and illumination that clarify the structure of the book and provide textual commentary. No doubt the tastes, the degree of literary sophistication, and the financial capacities

1. For a detailed description of such a manuscript, see Lepage, "Recueil francais." See also Faral's facsimile edition, Manuscrit 1 9 1 52 du Fonds français; Faral's remarks, pp. 10-11, are relevant here.

of the manuscript owner influenced the selection of texts and the degree and type of ornamentation; and the degree of patron control over manuscript production was probably itself variable. Whether the scribe was executing a plan of his own creation or one dictated by his patron, though, is less important for the present context than an understanding of what this plan was and how it was put into effect: in either case we are witnessing the processes by which a scribe or team of scribes shaped a group of texts into a book. In the following examples, therefore, I have not attempted to distinguish patron-initiated from scribally initiated features. I analyze the evidence of the manuscripts as artifacts and the work of the scribe as it appears therein in order to arrive at a critical reading of the books themselves.

Examples of Thematic Unity: MSS Bibl. Nat. fr. 24428 and 12786

In referring to this category of manuscript organization as “thematic,” I purposely choose a term of general rather than precise meaning, for it is meant to cover a range of possibilities. The distinction of "thematic" and "narrative" organization is itself somewhat artificial, and it is used here purely as a device for imposing some kind of order on an extremely diverse field. The two examples I have chosen exhibit different kinds of organization. In both, certain themes and motifs governed the selection of texts. In MS 24428 the pieces have been arranged in a linear progression, emanating from the first text and building up to the last; in MS 12786 the pieces are grouped in loosely defined categories around a central text.