Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Code, performance and ideology : the dialogue of reception as dramatic praxis in Voltaire's tragedies Leith, Hope Mary 1993-8-29

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


ubc_1993_spring_phd_leith_hope.pdf [ 10.5MB ]
JSON: 1.0086344.json
JSON-LD: 1.0086344+ld.json
RDF/XML (Pretty): 1.0086344.xml
RDF/JSON: 1.0086344+rdf.json
Turtle: 1.0086344+rdf-turtle.txt
N-Triples: 1.0086344+rdf-ntriples.txt
Original Record: 1.0086344 +original-record.json
Full Text

Full Text

CODE, PERFORMANCE AND IDEOLOGY: THE DIALOGUE OF RECEPTION AS DRAMATIC PRAXIS IN VOLTAIRE'S TRAGEDIES  By HOPE MARY LEITH B. A.(Hons), Simon Fraser University, 1987 M. A., University of British Columbia, 1988  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in  THE FACULTY OF GRADUATE STUDIES (Department of French) We accept this thesis as conforming  THE UNIVERSITY OF BRITISH COLUMBIA April 1993 © Hope Mary Leith, 1993  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature)  Department of  ^  ci&  The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  2A PI.NA^Ill 3  Abstract CODE, PERFORMANCE AND IDEOLOGY: THE DIALOGUE OF RECEPTION AS DRAMATIC PRAXIS IN VOLTAIRE'S TRAGEDIES This dissertation undertakes to examine the dramatic prixis and the ideological systems which shape Voltaire's tragic oeuvre. This study takes the position that these texts are multi-voiced, open-ended and performance-directed. The analytical approach taken draws from semiotic, marxist and feminist critical techniques. Thirteen plays were chosen for analysis: Artêmire (1720), HOrode et Mariamne (1724, 1725, 1763), Eriphile (1732), ZalKe (1732), Adelaide  du Guesolin (1734, 1751, 1765), Zulime (1740, 1762), Mahomet (1741), Semiramis (1748), Oreste (1750), Rome sauvde (1752),  Olympie (1760), Le Triumvirat (1764) and Les Guebres (1769). Chapter I undertakes the analysis of the textual codes, largely concentrating on the language which resulted from or which perisisted despite external reaction to the text, in order to reconstruct the rules by which language operated in the tragic form. Tragedy's requirement of "noble" language and action restrict it to those who had the classical education necessary to understand and manipulate its rules, thereby establishing a gender and class privilege within the text. Chapter II begins with the premise that performance is not external or incidental but integral to the texts under examination. It details the impact which performance and performers had on the text and on public response to that  iii text. It also brings to light Voltaire's profound ambivalence towards this influence on "his" texts. The desire to control performance led Voltaire to become a "director" of his plays in the modern sense, as he sought to impose his ideas of decor, costume, staging and declamation. Chapter III defines ideology as the system or systems of belief which underlie and inform the texts. The analysis is organized around three broad areas of social organization: government, religion and the family. Special attention is paid to the status and treatment of womne within these areas. The chapter examines whether the systems revealed are static or dynamic over time, personal to Voltaire or drawn from a wider social group, radical or conservative in content. The Appendix to this study furnishes a chronological table of textual transformations for each play studied, giving the source, location, date, extent, speaker, content and function of changes.  iv  Table of Contents Abstract Acknowledgments Introduction^  1  Chapter I - Transformations in Linguistic Code:^30 the Pursuit of Noble Language in Its Relationship to Class, Gender and Literary Genre A. Noble Language in the Tragic Form B.  Linguistic Shifts in the Tragedies  Chapter II - Textual Transformation and the Dialogue of Performance: The Struggle for Acclaim and Control A.  The Tension between Performance Success and Literary Merit  B.  Authorial Control of the Performance Text  C.^The Impact of Performance on Text Chapter III - Ideology in the Tragic World: Systems 142 of Belief and Codes of Behaviour in the Political, Religious and Familial Domains A.  Government and the Social Hierarchy  B.  Religion in Its Institutional and Individual Manifestations  C.^The Family within the Social Hierarchy Conclusion^  224  Bibliography^  234  Appendix^  244  ACKNOWLEDGMENTS  To Larry Bongie and Olga Cragg, who have encouraged and guided me in the field of eighteenth-century studies from the beginning and who have watched over this project from the time it was "without form and void." To those UBC professors whose contributions were invaluable: Bruce Carpenter, Floyd St. Clair, Richard Hodgson, who as members of my committee over the years bravely waded through draft after draft; To Valerie Raoul, who introduced me to feminist criticism; To Frank Hamlin, who made linguistics a fascinating exploration of language in society; and to those in eighteenth-century studies who have through conferences and correspondence given me much-needed encouragement and criticism: Robert Niklaus, Michael Cartwright, Eric Annandale, David Jory. To the family and friends who have kept me going during nearly five years. And most of all to my partner and our four-footed "family," for their perpetual support and encouragement.  1  CODE, IDEOLOGY AND THE DIALOGUE OF RECEPTION IN VOLTAIRE'S TRAGEDIES: TEXTUAL TRANSFORMATION AS DRAMATIC PRAXIS  Au reste, mon cher ami, je suis bien loin de croire la piece finie; je ne l'ai fait jouer et je ne vous l'ai envoyee que pour savoir si je la finirais. (Voltaire to Cideville, 19 juillet 1741, D 2515) The quotation above brings together what I will argue are the two key elements in Voltaire's writing for theatre: performance and collaboration. In order to study these in a fresh way, I have surveyed the linguistic and semiotic studies of French theatre, and some of theatre in general, to develop a method of analysis which encompasses: first, text and the transformations of text; second, the circuit of communication among spectator, actor and author in performance; and third, the underlying relationship between social ideology and literary expression. Among the linguistic and semiotic analyses of theatre, those of Anne Ubersfeld and Keir Elam apply to the study of theatre the communication theories of Jakobsen and speech act theory a la Searle [1]. Deirdre Burton, as well as Reboul and Moeschler, take a purely linguistic approach [2]. Larthomas is one of the first and the most wide-ranging critics in this domain [3]. Unfortunately, most of these analyses are applied to classical [4], romantic [5] or modern [6] theatre.  2  We can no longer now "read" Voltaire's plays as 18th-century spectators did, since we lack not only opportunity to see them performed but also direct, experiential knowledge of the numerous superimposed codes (visual, linguistic, gestural) by which meaning was expressed and understood on the eighteenth-century French stage. Hence, our reading will necessarily be incomplete, unjust in a sense; all we can do is admit it. Otherwise we find ourselves as critics in an awkward position: when Niklaus concludes that Voltaire did not know how to communicate emotion in his plays [7], he does not expound the relationship between his view, however valid for present-day readers, and the wealth of contemporary sources (both in correspondence and in more formal literary criticism) which describe the spectators' profound and spontaneous emotional responses. Lion's study gives dozens of examples of these reactions, including those of Diderot, Rousseau and Pine du Deffand [8]. The circuit of communication worked exceptionally well for Voltaire's plays in their original socio-historical context. It is this past success, and its evanescence, which we will examine. Corneille, Racine, d'Aubignac, Boileau [9], the theoreticians and masters of seventeenth-century theatre, are the basis not just for Voltaire's own theatrical writing, but for eighteenth-century theatre and theatre criticism as a whole. Eighteenth-century theatre was to a  3 great extent defined, measured, revised or redirected in reference to those seventeenth-century models, whether one examines the theories of Voltaire, Diderot or Beaumarchais [10]. Because of these established norms and models, change had to come slowly if the audience was to accept it (Niklaus, 1963, p. 1229). Too enthusiastic a rejection of classical rules led to performance failure, such as was the case for Landois' 1742 one-act bourgeois tragedy in prose,  Silvie, and for Voltaire's own plays Hdrode et Mariamne (1724) and Eriphile . Voltaire's ability to balance innovation against tradition has been cited as one reason for his immediate success. Indeed, we do well to remember that he was the pre-eminent playwright of the Comedie francaise in the eighteenth century [11]. But this adeptness was also considered the reason for his later decline; to some critics Voltaire was too closely attuned to the expectations and tastes of his particular audience to retain his appeal over time [12]. Voltaire's theatre has traditionally been condemned for its emphasis on performance success, its openness to revision, and its ideological content. In this regard many critics (such as Ridgway, Vrooman, and Cartwright) follow the judgment of the magisterial nineteenth-century critic, Lion. To Lion, Voltaire's theatre is consistently inferior to that of Racine and Corneille [13]. He deplored Voltaire's  4 willingness to revise, based on friends', actors' or audiences' reactions, to judge by performance rather than by literary ideals [14], to write for immediate success rather than eternal literary merit [15]. But to the modern critic, it is the distinctive trait of all theatre as such to be written for performance and, if not, it is not a play text but rather a poem or narrative written in dialogue form. As Ubersfeld states: Le trait fondamental du discours theátral est de ne pas pouvoir se comprendre autrement que comme une serie d'ordres donnes en vue d'une production scenique d'être adressó a des destinataires mddiateurs, charges de le repercuter  a  un  destinataire-public. [16] Duvignaud argues that theatre cannot be detached from the circumstances of its production, and that a playwright is a "complice des gens de theatre," requiring "des attaches vivantes avec le monde effervescent oil s'elaborent les piëces." [17] Voltaire meets these requirements thoroughly. The concept of theatre as literary text rather than as performance [18] has been seen as detrimental to the analysis of theatre because it insists that meaning exists  independent of / prior to performance, and thereby eliminates the role of the spectator in the production of meaning [19].  5 The critical view which considers propaganda and art as distinct and ideally unrelated domains has been rejected by many modern theatre critics. Anne Ubersfeld, for example, posits that all theatre expresses an ideology, and is in a relationship of agreement or opposition to the dominant ideology of the society that produced the theatre [20]. This integration of socio-political import with the theatrical text forms an essential premise in my analysis of Voltaire's tragedies. While Voltaire's theatre does not have the unity and harmonious development of Racine's [21], Lion finds in it "une reelle unite" [22], and considers that it expresses Voltaire's own beliefs and philosophy [23], a conclusion which is echoed by Ridgway [24] and Vrooman [25] among many others. To these critics, Voltaire is always the single subject, the locutpur of his theatrical text. This assumption I intend to challenge, using the concept of "double enonciation" taken from Ubersfield's work, the refusal of subjectivity by the theatrical text: Les conditions d'enonciation ne renvoient pas a une situation psychologique du personnage^equivoque propre au theatre: le discours au theatre est discours de qui? ... Le travail du discours theatral consiste echapper au probleme de la subjectivite individuelle. [261  6 Two recent book-length overview studies on Voltaire's theatre, one by Willens and the other by Vrooman [27], continue what Ubersfeld would call the biographical fallacy [28], in that they use the theatrical texts as a primary source of information about Voltaire's own beliefs or personality [29]. Vrooman, in 1970, finds a consistency and coherence in Voltaire's major tragedies from Oedipe to Nerope [30], a single line of thought expressing Voltaire himself, which seems quite remarkable when one considers the number of revisions and editors we know lie behind these printed texts, the number of potential voices from d'Argental to Lekain we may find in them. Nonetheless this unitary view continues to be argued [31]. Both Cartwright and Conlon [32] recognize the importance of actors, of performance as such to the successful completion of the circuit of communication between the play's author and its intended audience. But the ideas expressed within it, and even the text itself (considered as a concrete object, a written or printed document) remain exclusively Voltaire's. To these critics, performance remains something that happens to a text rather than a transformation of the text from within. Within this study, the relationship between dramatic text and theatrical realization will be maintained as essential. The distinction often drawn by critics between the performed plays and those that were unperformed or even  7 unperformable, a distinction attributed to Voltaire himself [33], has as its consequence a different type of analysis for each, and, furthermore is not logically sustainable in light of the evidence in the Correspondance. Voltaire put as much effort into trying to have the Guébres performed as into la, Mort de Cdsar; les Lois de Minos went as far as rehearsals at the Comddie frangaise, as Lion admits [35]. The approach undertaken here requires us to restore the power of performance (actual or virtual) to affect text, of the role of the audience in the construction of meaning within the theatre proper and indeed in constructing the text as document [36]. By reading theatre purely as literary text, we can miss what is specifically theatrical: for example, the ability to convey meaning by ostention, as when in 54mirapis Ninias' re-appearance with blood-stained arms indicates that he has killed SOmiramis [37]. Other specifically theatrical codes for conveying meaning are intonation and proximity, both of which affect meaning through bow the line is enonciated. [38] No reading can fully restore meaning as conveyed by performance, but it should, at the least, recognize that meaning is not single or static, that a number of codes are being used simultaneously [39]. As Elam brings out, It is the performance, or at least a possible or 'model' performance, that constrains the dramatic text in its very articulation....The written  8 determined by its very need for stage contextualization, and indicates throughout its allegiance to the physical conditions of performance. [40] Although Larthomas's work is of considerable interest, his approach differs greatly from the one undertaken here in that he is concerned with those aspects of theatrical "grammar" and "syntax" which appear unchanging, which . continue to communicate, at least within the English and French traditions. My concern is with those aspects which have changed, which no longer convey meaning and which require conscious decoding. Critics frequently make passing references to audience expectations and reactions to plays, insofar as these serve to confirm literary judgments, or to give temporary undeserved merit; this study, because of its interest in how and what the texts communicated with their intended audiences, will concentrate on critics contemporary with Voltaire, such as E, C. Frdron, Charles Cone and Pierre Clement. One of the few systematic studies of theatre audiences in France, by Maurice Descotes [41], deliberately omits Voltaire, along with Marivaux and Beaumarchais, from its corpus. I wish in some small measure to fill this gap, construire les attitudes qui correspondent aux pieces mises en question par le succes ou l'echec, definir les intentions cachees qui permettent de mesurer la qualith  9 de l'attente ou des attentes propres a des secteurs de l'histoire." [42] Without this insistence on the socio-historical specificity of theatre as a genre, texts being written for a certain audience, building, or even actor, one is forced to fall back on a theory of "universal truth": Comment va se constituer le nouveau rapport entre un discours textuel creó en relation avec un public dótermine, et un public qui a change et dont ni les preoccupations ni la culture, ni l'ideologie ne sont les memes? La tendance la plus simple est de nier le problëme et de tenir que le rapport entre le discours du scripteur et la voix du spectateur se fait sur le terrain d'une nature humaine universelle, de passions eternelles. [43] But the pursuit of a "universal truth" opens the way for the critic to "privildgier non le texte, mais une lecture particuliere du texte, historique, codee, iddologiquement dêterminee, et que le fftichisme textuel permettrait d'dterniser." [44] André Helbo goes further, and speaks of a "terrorisme du texte." [45] Theatre is not a single textual voice, but a chorus: "le signe thdatral est une notion complexe qui met en jeu non seulement une coexistence, mais une superposition de signes." [46] Only the ideal spectator, in the Platonic sense of "ideal", can interpret all operant codes fully and simultaneously. The  10 incompleteness of the performance text appears necessary, the unavoidable result of choices made both by the performers and by the spectators [47]. A purely "literary" reading risks choosing to omit even more. For example it may not take into account codes other than the linguistic, like gesture, intonation and proximity, and sometimes not even those changes made at the linguistic level in response to public reaction. Particularly when reading texts widely separated from our cultural context by time and history, it is well to remember that language which seems artificial and cliché to us may well have been not only accepted but even expressive to audiences in its time. The importance of performance, of establishing the circuit of communication in theatre between text and spectator through the performers, is difficult to over-emphasize. Elam's argument about the impact of performance considerations on text seems particularly apt to Voltaire's theatrical praxis as explored in this study [48]. The spectator or reader as virtual spectator is not just a passive receiver but rather the initiator of and participant in the communication. Elam argues that performance cannot take place until the audience arrives. Voltaire's overriding concern, as seen in his correspondence, was that his plays be seen and heard, that they provoke a public reaction to which he would respond. Mere publication of the play text was but a tangential goal and, often, a quiet  11 concession that attempts to have the work performed had failed [49]. The critical view in which theatre is essentially dialogue, rather than the somewhat monologic, theatre-as-poem position, will find abundant support in the theatrical praxis of Voltaire. Since theatrical discourse has a "double subject", it offers a special situation for interpretation: Les sentiments et les emotions que le personnage est cense dprouver ne sont en fait eprouves par personne....I1 n'y a pas de signifiance de parole hors des conditions de lienonciation....Seule la situation de parole precise le sens du discours. [50] The "situation de parole" includes the conditions of enunciation, the relationships of power between the characters which determine not only the content of speech but even the ability to speak [51]. The dominant ideology, the "formation socio-historique donnee," acts not only directly through speech but also indirectly on speech by controlling access and content [52]. From Voltaire's total corpus of 27 tragedies, I have selected 13 for closer study [53]. This selection can be justified by a number of considerations including: first, the works selected encompass Voltaire's tragic works from beginning to end of his oeuvre; second, they represent the full range of potential public receptions in that the four sub-categories into which I have grouped the plays selected  12 are successes, failures reworked successfully, failures reworked unsuccessfully, and plays never performed; third, the plays chosen present a large number of textual variants and revisions, and finally the length of time over which a play is revised [54]. In making these selections, I was particularly concerned to cover as much of Voltaire's career as possible, and to include the range of public response, from the unperformed Guebres to the enduring success of  Zaire, a play which survived in the repertoire through the nineteenth century. Successes:  Zaire (1732) 44h9met (1741) Oreste (1750) Rome sauvde (1752) 0Iympie (1764) Failures reworked successfully:  Artdmire (1720) into lidrOde et Marianne (1724, 1725, 1763)  griphyle (1732) into Semiramis (1748) Adelaide du Guesclin (1734, 1751, 1765) Failures reworked unsuccessfully:  Zulime (1740, 1762) Le Triumvirat (1764)  13 Plays never performed:  Les Guebres (1769) "Success" and "failure" are relative terms both during Voltaire's career (he had very few first-night closings), and at the present day for a theatre which seems completely unperformable within twentieth- century codes of drama. The terms were initially and provisionally defined in terms of public performance during Voltaire's lifetime. But ZUlime for example, while initially a failure at the Comédie frangaise, was popular in private performance and enjoyed a revival in the early 1760s. Les Guebres, while not performed, was considered performable and even technically competent by Diderot when reviewing its publication [55]. Voltaire's tendency to rewrite in response to criticism is well documented for all his plays, as Lion's meticulous research details (56]. Although any selection from among Voltaire's tragedies is to some extent arbitrary, this particular group of plays provides the greatest possibility of reconstructing the process of textual transformation which lies behind the apparently unitary and static texts confronting the twentieth-century reader. Passing reference to other plays in the oeuvre will be made where relevant, particularly in the chapter on ideological systems. While some reference is made to manuscripts, it is not within the scope of this thesis to undertake detailed analysis of, for instance, all  14 textual variants within all manuscript versions extant, as such a task would approach that of a critical edition for each play under examination [57]. The approach adopted here is distinguished from others especially by its emphasis on text as process rather than as product, that is to say on the creative dialogue between the play text and its receiver, and by its treatment of ideological considerations as inherent to all dramatic communication. The goal of this research is a reconstruction of the theatrical praxis which produced these texts - in such quantity, over so many years, with such success. It strives, in a necessarily limited sense, to "read" Voltaire's play texts as they were read and performed in his own time, to reveal the ideologies they express or challenge, to understand more completely the codes - visual, linguistic, gestural - by which they communicated. Further, it seeks to bring out not only the codes but also the  3.pcuteurs in Voltaire's texts, the chorus of voices we are hearing when we read. These texts would not be what they are without Cideville, Thieriot, d'Argental, Lekain, Mlles Quinault and Clairon; without the critics and the censors and the fickle audience of the Comedie francaise, an audience bound by expectations of tradition even as they were bored by it; nor without the physical and practical constraints of the stage itself. The reconstruction of the  15 theatrical praxis which produced these texts requires us to rethink the texts as performance rather than as literature, as multi-voiced and open-ended. The traditional concept of author/ity is questioned; Voltaire's struggles to assert control over performance as well as over printing are simultaneous with his continual revisions, his responsiveness to advice. Voltaire's own praxis seems based on the interaction and collaboration between audience as spectator/reader/listener on one hand and author on the other, rather than composition in isolation, the author sole progenitor of his text.^It is this successful theatrical praxis, and the society in which it succeeded, that become the object of this study. My approach is distinct from what one may call the traditional by its insistence: first, on the number of other "voices" effecting transformations in the text ; and, second, on the relationship between linguistic, literary and performance codes on one hand and ideology on the other. For example, Sanderson's work, although extremely detailed, is primarily concerned with form, and does not address the issue of ideology expressed by the text and affected by the revisions [58]. Likewise Abrate's thesis Play an4 Playscript, while adding the consideration of performance and audience to a discussion of form and structure, remains largely limited to dramatic technique and, while examining how the text seeks to provoke certain audience responses,  16 does not examine the impact of audience reaction on the text itself [59]. Many critics have addressed the issue of the play text as didactic or philosophical pamphlet. However, this line of criticism tends to downplay the texts as effective theatre, and to presume the coherence and consistency of the philosophy expressed [60]. It is an essential premise of this approach that no literature and no method of literary criticism are ideologically neutral. The codes in which they are written and by which they are understood are social constructs inseparable from the belief system of the groups which use them. In this way the analysis of textual variants, and in particular their sources, provide the basis for new insights into the ideological content and didactic function of these texts. Only those reactions from readers, actors, audience and critics which provoked a concrete change in the text and thus which now permit some reconstruction of the performance are studied. This productive dialogue forms an essential element in Voltaire's dramatic praxis. This study intends to progress from textual evidence to performance interaction to ideological conclusions. In Chapter I, "Transformations in Linguistic Code: the Pursuit of Noble Language in Its Relationship to Class, Gender and Literary Genre," our concern with the dialogue of textual transformation obliges us to limit this study to those linguistic changes which resulted from or which  17 persisted in the fact of external criticism and which therefore allow us to draw conclusions about the linguistic codes operating between the text and its receivers. This chapter demonstrates that the language of Voltaire's tragic texts represents not so much an idiolect, a way of using language specific to the author, as a kind of socially defined dialect, a way of using language to indicate not only one's literary skill but also one's education and one's social status. The basic assumption of this part of the analysis is that the plays represent not a monologue but a dialogue among several "speakers". We can find reflected in the text the linguistic tastes and judgments of that circle of Voltaire's friends who read and critiqued his work, of the actors who performed them, of the censors who edited and approved them, and even of the journalists who reviewed them. The rules of "noble" language, of tragedy as a dramatic genre, were defined by French society, not to be altered easily by any individual writer. Voltaire's struggle to be creative and at the same time acceptable within this structure, the identity of those "speakers" whose voices we find preserved in the changes to the printed texts which remain, and those whose voices Voltaire sought to control and eliminate, are issues which this chapter addresses. Further, the normative effect of noble language on the tragic form, and the source, timing and importance of  18 changes at the level of language, are examined. The basic premise of Chapter II, "Textual Transformation and the Dialogue of Performance: the Struggle for Acclaim and Control" is that performance is not external or incidental to the texts under examination, but inseparable from them. Theatre, perhaps even more strongly than text intended to be read, is a situation of dialogue and interaction among senders and receivers: the author, the actors, and the audience. The meaning constructed from the performed text, and the reception afforded it by audiences, can vary greatly from those attributed to it by its readers, particularly those reading without conscious reference to a potential stage realization. However, the question of reception is one which does not seem to have received the same degree of critical scrutiny as has been devoted to theatrical form and function within Voltaire's oeuvre [61]. In this chapter, the ways in which performance and performers, both virtual and actual, have significant impact on the texts of Voltaire's tragedies is considered. Voltaire's efforts to control not only dramatic text but also performance text are indicated in the previous chapter by his opposition to editing by the actors, that is, to their changing the text by reducing it. But this continuing involvement in performance text has broad implications not only for the plays as they now appear but for  19 eighteenth-century theatre generally. One may argue that Voltaire's involvement with the stage - as an avid amateur actor and metteur en scene, his "training" of certain performers, as well as his writing contributed to changing audience expectations and standards of tragedies. He not only responded to audience reaction, but strove to guide it along specific lines. ^This second chapter examines the dynamic tension between performance success and literary merit, the struggle for control over the text during and after performance, and the concrete impact of performance on text. Chapter III, "Ideology in the Tragic Structure: Systems of belief and Codes of Behaviour in the Political, Religious and Familial Domains," defines ideology as that system or systems of beliefs which determine the proper structure of a society and which provide the norms by which citizens of that society are to behave. While some critics have concluded that certain of Voltaire's tragedies have little or no ideological content [62], this study takes as one of its fundamental premises that all theatre has some ideological function in that it exists in a relationship of reinforcement or reform to the society which produced it by the type of society it idealizes or critiques. Analysis of the ideological content cannot, however, be confined to the explicit statements which, as we will discuss, are frequently undercut and even denied by the  20 context in which they are spoken. The "situation de parole" which is inseparable from the "enoncd" itself includes the conditions of enunciation, the relationships of power between the characters which determine both the content of speech and the very ability to speak, and also the actions which arise from that speech. The dominant ideology, the "formation socio-historique donnde," is conveyed not only directly through speech but also indirectly by constraints on speech and speakers [63]. Chapter III treats the ideological system of each play as well as the relationships between them, examining what kinds of societies and individual behaviours are presented and the degree of approval or condemnation afforded them. This discussion is organized around three general areas of social structure: government, including monarchies, republics, and the role of the masses; religion, including the priesthood as a class, the status of state religion, and the portrayal of individual priests and believers; and family, including gender roles, the division of authority and duty, and the control of sexuality. Particularly in this last domain we will concentrate on the role of women in the tragic world. Transformations at the level of linguistic and performance codes form the basis for the third chapter insofar as ideology is expressed not only through explicit content but also and often more revealingly through context:  21 not only who says what, but also to whom, when, with what result. Ideological bias is also expressed through control over speech: how often, how much, and to whom a character is permitted to speak. Passivity, for example becomes a three-level term: silence, absence, and inactivity. This complex definition permits a clearer revelation of underlying beliefs and attitudes: e.g., women may be seen as largely absent, as silent presences, or as verbose but essentially ineffective adjuncts. The transformations are discussed in terms of those which appear to reflect an attempt to express the same belief more convincingly and those which reflect a change of belief. We will look for the emergence of consistent definitions and even of coherent belief systems from the texts considered as a group. Further, we will consider whose beliefs are being expressed. Often, the beliefs can be seen as those of Voltaire's circle or even of the society in which he moved rather than exclusively those of Voltaire in isolation. Hence, the belief systems may be, in common with the texts which express and operate within them, arguably the product of dialogue and consensus, a product of the society in which they appeared. In the Appendix, this study provides for each play studied a chronology of textual transformations, so that one may trace the development of the text over time, the various sources of the transformations, the characters and plot  22  aspect involved, and in general terms the content of the transformation. As well, the number of variant lines known for a particular text compared to the total number of lines in that text is given, so that the extent of transformation can be quantified.  23  Note [1] Anne Ubersfeld, lire le thëatre (Paris: Editions sociales, 1978) and Keir Elam, The Semiotics of Theatre and  Drama (London and New York: Methuen, 1980). [2] Deidre Burton, Dialogue and _Discourse; a  sociol,inguiptic approach 10 moderp drama dialogue and naturally occurring conver,sation (London: Routledge & Kegan Paul, 1980); Anne Reboul and Jacques Moeschler, Discours  thdatral et analyse conversationnelle (Cahiers de linguistique francaise #6, 1985). There is also similar work on Voltaire's contes: Carol Sherman, Reading  Voltaire's contes: a sepiotics of Oilopophical narration (North Carolina UP, 1985). [3] Pierre Larthomas, Le L'angage examatique (Paris: Armand Colin, 1972). [4] Reboul and Moeschler look at Moliére; Larthomas, at Racine and Corneille. There is material on Marivaux and Beaumarchais in Larthomas. But his work seeks to define what is "universally" theatrical, not what is specific to a certain period. Thomas J. Carr Jr., in "Dramatic structure and philosophy" (SVEC CXLCIII (1975): 7 - 48), discusses the ideological implications of certain typical plot structures, but does not consider language except to deplore Voltaire's adherence to alexandrines. [5] For example, Anne Ubersfeld on Hugo: Le Roi et le  bouffon (Paris, Corti, 1974).  24 [6] Ionesco and Pinter in Burton (1980); Ionesco in Reboul and Moeschler (1985); and a number of modern works in Larthomas (1972). [7] Robert Niklaus, "La Propagande philosophique au theatre au siècle des lumieres," SVBC XXVI (1963), p. 1229.  [8] Les Tragedies et les theories Aramatiques de Voltaire (Geneve: Slatkine, 1970; reproduction of the Paris 1895 edition), pp. 154, 255, 352, etc. [9] Nicolas Boileau - Despreaux, Art pOtique chant III in Oeuvres(Paris: Gamier freres, 1952); Pierre Corneille, les deux Discours in TWtre complet (Paris: Gallimard, 1957); Jean Baptiste Racine, prefaces to  Berenice, Iphegenie, and Phedre in Oeuvres completes (Paris: Gallimard, 1950 and 1952); d'Aubignac, La Pratique du  theatre, ed. P. Martino (Paris: Carbonel, 1927); Jacques Scherer, La Dramaturgie classique (Paris: Nizet, n.d. [1950]). The above sources give at least a summary basis for conclusions on the rules of classical dramaturgy. [10] The sources for Voltaire's theory are scattered: prefaces, Qommentaires on Corneille and Racine, articles in his Dictionna,ire philosophique, etc. See also Denis Diderot, BrItret,iens sur  le Fils naturel and De la podsj.e  dramatiquein Oeuvres (Paris: Editions de la Pleiade, 1951) and Pierre-Augustin Caron de Beaumarchais, Essai sur le  genre dramatique serieuxin Oeuvres, ed. Pierre Larthomas with Jacqueline Larthomas (Paris: Gallimard, 1988). ^See  25 Niklaus, op. cit., p. 1229 on the continuing impact of seventeenth-century norms on audience expectations. [11] John Lough, Paris Theatre Audiences in the  Seventeenth and Eighteenth Centuries (Oxford UP, 1957) and Henri Lagrave, Le Theatre et le public a Paris de 1715 a  1750 (Paris: Klinsieck, 1972). Statistics on the number of performances, tickets sold and revivals serve as a useful balance to critics like Cone and Freron, whose judgments are often tainted by philosophical partis pris. The catalogues of the Bibliothéque Nationale, of the British Library, and the National Union Catalog help to trace the printed editions and translations. [12] Jack Rochford Vrooman, Voltaire's theatre: the  cycle from "Oedipe" to "Mdrope" in SVEC LXXV (1970): p. 190. [13]  Lion, p. 267.  [14] nig, pp. 34 and 419-21. [15] 'bid, pp. 267, 299, 409. [16]  Ubersfeld, p. 258.  [17] Jean Duvignaud, La Sociologie du theatre (Paris: PUF, 1965), p. 36. [18] Lion, pp. 34 and 421. [19] See Ubersfeld, p. 121 and Elam, p. 208.  [20] Ubersfeld, pp. 265-67. [21] Lion, my translation of p. 422.  [22] Ibid, p. 42. [23] Ibid. pp. 18, 429, 474.  26 [24] Ridgway, p. 235. [25] Vrooman, p. 175. [26] Ubersfeld, pp. 249, 265. [27] Lillian Willens, Voltaire's comic theatre;  composition, conflict and critics SVEC CXXXVI (1975); Vrooman. [28] Ubersfeld, p. 250. [29] See, for example, their conclusions: in Vrooman, p. 207; Willens, p. 182. [30] Vrooman, p. 175. [31] Alain Niderst, "Tragique Voltaire," in Le Siècle  de Voltaire; Hommage aRene Pommeaq, eds Christiane Mervaud and Sylvain Menant (Oxford: Taylor Foundation, 1987) II: 701 - 706. [32) Cartwright, in his introduction to Adelaide du  Guesclin; Pierre Conlon, Voltaire's literary career froM 1728 to 1750 SVEC XIV (1961): ie at pp. 108-150. [33] Lion, pp. 351, 370. [34] Voltaire, Correspondance , ed. Theodore Besterman (Oxford, 1968). See for example the correspondence surrounding Les Guébres, January to July 1769. [35] Lion, pp. 354-56. [36] Elam, p. 34; Duvignaud, p. 7. [37] Elam, p. 29. [38] Umberto Eco, "Parametres de la semiologie thdatrale" in Semiologie de la representation (Paris: PUF,  27 1975), pp. 39-40. [39] Regis Durand, "Problemes de l'analyse structurale et semiotigue de la forme thedtralen in Sdmiologie de la  representation, p. 113. [40] Elam, pp. 208-09.  [41] Le public du _theatre et son histoire (Paris: PUF, 1963). [42] Duvignaud, p. 39. [43] Ubersfeld, p. 267.  [44] Ibid, p. 17. [45] André Helbo, avant-propos to Semiologie de 14  representation, p. 9. [46] Ubersfeld, p. 31. [47] Ubersfeld, loc. cit.; Elam, p. 49. [48] Elam, pp. 208-09. [49] The notes to the Moland edition of Voltaire's works (Paris: Gamier fréres, 1878), e.g. those for  Olympie, le Triumvirat and les Guëbres in tome VI, establish this prioritization of performance over publication. [50] Ubersfeld, pp. 139-40. [51] Ibid, p. 283.^A well-known modern example of this correlation between speech and power is the relation between Pozzo and Lucky in Beckett's Waiting for Godot. Within eighteenth-century drama, Voltaire's Nanine provides illustration of this point in that the heroine is virtually silent through the final scenes of the play, while the count  28 and her father decide her fate. [52] Ubersfeld, pp. 287, 296-97. [53] In chronological order: Artdmire, }erode et  Mariamne, Eriphile, Zaire, Adelaide du Gqesclin, Zulime, Mahomet, Semiramis, Rome sauyee, Oreste, Olympie, is Triumvirat, les Guebres. In contrast to the plays chosen, offers only 40 lines of substantive textual variants and some variation between contemporary editions in the inclusion of stage directions and author's notes in a text of 1426 lines, leaving aside such changes as singular/plural of the type on/nous, replacement by synonyms or functional equivalents (prepositions and determiners), and syntax reordering. Merope provides slightly more than 100 lines of variants (of which 80 exist only in reference to one manuscript) and thirteen lines quoted from Briphile in a 1402 line text, according to the most recent critical editions: Alzire, ed. T. E. D. Braun (1989) and Merope, eds. Jack R. Vrooman and Janet Godden (1991) in the Voltaire Foundation edition of the Complete Works published by the Oxford University Press. Neither of these plays shows evidence of substantial revision over time, and thus do not offer the material for analysis to be found in plays such as  Adelaide or Herode et Mariamne or Zulime, which have hundreds of lines of variants and substantial reworkings of plot and character over periods of thirty years or more. [54] I am working largely from the Moland edition of  29 Voltaire's works, as of the plays selected only Michael Cartwright's Adelaide du Guesclin (1985) and Eva Jacobs'  Zaire (1988) have appeared to date in the new Voltaire Foundation edition. These two editions will of course be consulted. Extensive reference is made, further, to the Oxford edition of the Correspondance edited by Theodore H. Besterman (1968). [55] Haydn Mason, "Diderot critic of Voltaire's theatre," in Le Siecle e NOmmage A Rene yomeau, eds. Christiane Mervaud and Sylvain Menant (Oxford: Voltaire Foundation, 1987) II: 636. [56] Lion, gp. cit., passim. [571 Anne Sanderson's lengthy articles, "Voltaire and the problem of dramatic structure: the evolution of the form of Eriphyle" and "In the playwright's workshop: Voltaire's corrections to Irene" in SVEC CCXXVIII (1984): 97-170 illustrate this point.  [58] Ibid. [59] Jayne Halsne Abrate, Play and Playscript;  Dramatic Structure ip Six Voltairean TKagedies, Ph. D. thesis, Purdue, 1983. [60] I.e. the works already cited by Ridgway, Vrooman and Niderst. [61] Patrice Pavia, Languages of the Stage - Essays in  the Semioloqy of Theatre (New York: Performing Arts Journal, 1982), p. 70.  30 Chapter I - Transformations in Linguistic Code: the Pursuit of Noble Language in Its Relationship to Class, Gender and Literary Genre  Theatrical text offers special problems to a linguistic analysis in that theatre is a heterogenous genre, neither wholly written nor wholly oral, combining text with visual representation, sound effects and even music. Suzanne Langer's distinction between discursive and presentational forms was designed to assist in the semiotic analysis of art as opposed to literature [1]. Theatre, however, which is both discursive or linear in that information is conveyed by means of text, and presentational in so far as the essential unit of meaning is the performance as a totality, requires analysis conducted on both registers. Langer's division of "  meaning-situations" into general types or modes: language,  ritual, myth and music [2), while no doubt valid and useful, serves to emphasize the complexity of the theatrical situation. Theatre can combine simultaneously elements or functions of all four modes by the use of costumes, sets, sound effects and music, by its role as a social ritual indicating the intellectual or financial status of the spectator, and by its power to express that society's myths about divinity, morality, and rationality. Although analysis here will be largely limited to text by our distance from performance, the remaining elements of  31 performance and social ritual will be brought forward, and the gaps left in the printed text by their non-realization indicated. Analyzing the message communicated by a play text must involve this consideration of performance rather than treating the text as exclusively literary. The theoretical basis for the linguistic analysis proposed might best be categorized as drawing from structuralist, semiotic and marxist approaches. Reference will be made in particular to the theatrical studies of Anne Ubersfeld, Keir Elam and Jean Duvignaud, of which the first two draw from speech-act theory. Their work on the sociological and ideological implications of theatre is introduced here, but will also inform the third chapter dealing with ideology in the theatrical text. This analysis in grounded in the premise that language is fundamentally a social phenomenon, inseparable from the community that defines it and uses it, and necessarily bound up in the ideology of that community. In this regard it can be called a marxist approach. Volosinov expresses the view as follow: The actual reality of language-speech is not the abstract system of linguistic forms, not the isolated monologic utterance, and not the psychophysiological act of its implementation, but the social event of verbal interaction implemented in an utterance or utterances....Verbal communication can never be  32 understood and explained outside of this connection with a concrete situation. [3] The nature of the situation and of the audience determines the forms of utterance and the behaviours which accompany and inform it (ibid, p. 63). Theatre offers an extreme situation of social interchange, with dialogue occurring between author and performer as well as with the spectator within the framework of supposedly real dialogue between characters. Volosinov deliberately extends his definition of verbal interaction to include printed texts; all forms of verbal communication are bound to the society in which they were produced. As I have argued in the introduction (pages 8 - 10), it is the context, the situation of communication, which has changed for today's readers of the texts under study. Plainly put, we no longer "speak" the codes through which these texts originally communicated. Our concern with the dialogue of textual transformation obliges us to limit ourselves in this section to those linguistic changes which resulted from or which persisted in the face of external criticism, and which therefore allow us to draw conclusions on the linguistic codes operating between the text and its receivers. As indicated in the introduction (p. 17), I will demonstrate that the language of Voltaire's tragic texts represents not so much an idiolect, specific to Voltaire, as a kind of socially  33 defined dialect, using language to indicate not only one's literary skill but also one's education and one's social status. It is my hypothesis that the plays represent not a monologue but a dialogue among several "speakers". We can find reflected in the text the linguistic tastes and judgments of that circle of Voltaire' friends who read and critiqued his work, of the actors who performed them, of the censors who edited and approved them, and even of the journalists who reviewed them. The rules of "noble" language, of tragedy as a dramatic genre, were defined by French society; they were not easily altered by any individual writer. Voltaire's struggle to be creative without alienating his audience, the identity of those "speakers" whose voices we find preserved in the changes to the printed texts which remain, and those whose voices Voltaire sought to control and eliminate, are issues which this chapter will address. In eighteenth-century France, tragedy was the "noblest" form of theatre, indeed inferior only to the epic in the full literary hierarchy. Its authors were required to write in an elevated style, largely defined by the seventeenthcentury theorists Boileau and d'Aubignac and for which Racine had become the model. The purification and the exaltation of language in tragedy were part of a broader effort to raise French to the status of Latin and Greek, and to endow it with the prestige of a truly "literary"  34 language. Influential in this process was the Grammaire de  Port-Royal, as well as the various prounouncements of the Acaddmie francaise, which published its first dictionary in 1694. If language is inseparable from the society which uses it, then one can expect to find reflected in language the power structures of that society. Access to language, and control over it become ideological issues. Existing class hierarchies and gender inequities will inevitably be expressed in the literature of that society, and can be demonstrated by analysis of that literature. Feminist critics (like Daly, Spender, and Penelope) have assessed the seventeenth-century pursuit of a "pure " or "noble" language in various countries not merely as a nationalistic desire for equal status with the classical languages, but also as part of the consolidation of patriarchal authority. This dual motivation seems particularly plausible in France where it coincided with a larger political shift to absolutist monarchy under Louis XIV. [4] The task of bringing the French language under control, of giving it the status previously reserved for classical languages meant bringing it under male control - that of the grammarians and the academicians: The construction of grammars and grammatical speculations in the west have been male-only enterprises from the beginning, studies dominated by  35  philosophers concerned with rhetoric, on the one hand, and ethics on the other....Both countries [France and Italy] had already established academies charged with keeping their languages pure and the learned men of England weren't going to sit by while rival nations purged their languages of foreign influence and decided rules of correctness that would forever protect their linguistic identities from the corruption of the dialects of the poor and the lower classes. [5] Penelope further makes clear that the explicit purpose of the Acadómie francaise was to "cleanse French" of plebeian or foreign or "prêcieux" elements. Women had been key players not only in the "precieux" movement, with women writers like Mlle de Scudery, but also in the classicizing reaction under Malherbe. However, only men sat in the academies, wrote the dictionaries and grammars, taught or studied in the colleges. The role of socially prominent women in defining "le bon usage", while considerable, was arguably subordinate, particularly in view of even their limited access to education. Roseann Runte does claim that in this period "literary and artistic fame and fortune depended on the judgment of women." [6] But she concludes, "Although women attempted to break out of the family circle, they did not succeed in being cast in new roles." (p. 151) The task allotted to them seems to have been to influence and encourage through the salons, that is,  36 a social and oral sphere of activity in which they could regulate the propriety of what men could say or write. Actresses, on the other hand, may offer an exception to this observation; the stage may give women lacking in social status and classical education an opportunity to influence and interpret at least the nobler forms of literature. It is in fact typical of patriarchal societies, according to such critics as Daly and Penelope, that women are made responsible for correctness of language while they remain excluded from mastery of it: It is well known that women excel in the study of prescriptive grammars....Granting this, however, Jesperson [in Language: Its Natures Development and  Origin (1921 Otto) p. 249] didn't hesitate to turn men's inferiority into a plus for them! ...Men have used our alleged linguistic conservatism against us -in other words, our ability to use the rules of languages they've developed -- at least since Cicero....If women speak Standard English, we're unimaginative and stupid. (Penelope, pp. xix-xx) Seventeenth and eighteenth-century women were in fact conscious of their role as guardians of good taste and good grammar [7]. Nonetheless, literature and education remained male domains throughout the seventeenth and eighteeth centuries in France. The now familiar division between  37 nature and culture, intuition and reason along the line female/male is made even by female writers: Nous allons aussi surement a la verite par la force et la chaleur des sentimens, que par l'etude et la justesse des raisonnemens et nous arrivons toujours par eux, plus vite au but dont il s'agit, quo par les connaissances. (Mme de Lambert, ibid, p. 97) It is striking confirmation of this sexual division of language that, indeed, one praised this higher form of language, the French of epic poetry and tragedy, by categorizing it as masculine: the Mercure, in approving Voltaire's imitation of Greek style in Ertphile, chooses to say "la diction en est male." (March 1732, p. 562) The critic Freron likewise uses this gendered approbation in speaking of Voltaire's Tancrede, which he describes as painted by "un pinceau si brillant, Si male, si harmonieux." [8]  Eighteenth-century French tragedy, as a genre demanding considerable classical education and verbal skill from its writers, thus seems likely to prove "masculine" not only in the language it uses but also in the beliefs it expresses. While the questions of underlying ideology properly belong in a later chapter, we wish to indicate from the beginning our intention to discuss issues surrounding the access of lower-class persons and particularly of women (or female  38 characters) to language on stage and in society, and in this chapter the degree to which women had impact on the Voltairean text, both as a written document and in performance. There were no French women making a living or a reputation as playwrights (or indeed taken seriously as writers of any kind) during the eighteenth century, as Showalter's article indicates [9]. Those women writers we do know of were mainly either noblewomen writing for themselves or a small circle such as Mme du Deffand and Mme de Lambert, or popular novel writers such as Mme de Graffigny and Mme Riccoboni. Of plays written by women in the period, only Mme de Graffigny's Cênie (1750), a drame  bourgeois, comes to mind as acheiving anything like success in performance [10]. Graffigny's second play, La Fille  d'Ariptide, was a disaster; Mme du Boccage's Les Amazones and Mme Denis' La Coquette punie were little more than curiosities. This situation is in strong contrast with that in Britain, where Aphra Behn and Susannah Centlivre lived by their playwriting before Voltaire produced his first play. Showalter concludes that the "institutional privilege of the theatre disenfranchised women in fact if not in principle, and in denying them the possibility of writing for the stage [which was the most lucrative field for writers of the period] effectively disbarred them from writing for a living." (p. 97) However, French women could hardly be faulted for this marginalization, as they had little if any  39 access to the kind of education, the grounding in Latin and in classical theory, required to write "correctly" in the higher genres like tragedy. Authors and critics, like grammarians and academicians, were men. Theatrical performance, as opposed to composition, takes a written text and places it in an oral context. Therefore, it may offer women as actresses greater opportunities to control or at least affect text. Voltaire's career as a dramatist was marked throughout by his collaborations with the leading actresses of his time: Lecouvreur, Quinault, Gaussin, Dumesnil, Clairon. He sought their advice, gave them credit for a play's popular succes, wrote and rewrote roles to suit them, coached them on acting and criticized not only their declamation but also their costumes, gesture, and decor. While one can argue that printed text remains male-dominated, and certainly Voltaire did everything in his power to retain personal control over printed texts, performance text may prove to have been an arena in which women could contribute to and even manipulate the "verbal communication". Audience reaction, however, at least that of the parterre, which was most direct and active, largely remained controlled by men. The leaders of the ca•ales, the parterre itself, the aristocrats on stage, those in the best position for immediate and interventionary reaction to performance (hisses, laughter, quips, interruptions) were all men.  40 This chapter will look into the sources of the advice and criticism on correctness of language that Voltaire sought, followed, disputed or ignored, and to determine whether any implications of class or gender bias emerge. Linguistic accuracy was unquestionably a priority for tragedy in the French classical mode. Boileau's ideal is a level of expression so refined that it is capable of discussing any subject without offending social biensdances: D'un pinceau dèlicat l'artifice agrdable Du plus affreux objet fait un objet aimable. [11] Sophocles is credited with raising Greek tragedy to its "hauteur divine" because he: Accrut encor la pompe, augmenta l'harmonie, Interessa le choeur dans toute liaction Des vers trop raboteux polit l'expression. (Ibid) So high a standard was established in the seventeenth-century that Racine himself was criticized for "low" or "common" speech. Berdnice in particular was attacked for its inclusion of phrases like: "Eh quo!! seigneur, vous nlètes point part!!" and "Remettez-vous, madame, et rentrez en vous-méme." [12] Although Voltaire is generally described as an imitator of Racine, it is perhaps worth noting that the memorable rèpliques of Voltaire's theatre are of this deliberately simple type, with "Zaire,  vous pleurez" being the best known example. In the eighteenth century, even those critics who  41 extolled Crebillon's playwriting over that of Voltaire had to admit that the language of Crebillon's plays did not meet the classical standard. Desnoiresterres described it as "rocailleux, incorrect, barbare" [13]. Grimm, in the  Correspondance “tteraire, while reviewing Crebillon's Cataline, exclaimed "Quel barbare frangais us parlent, tous ces Romains, et surtout le prince de l'dloquence latine!" [14) The journalist and critic Pierre Clement, no devotee of Voltaire's work, attacked Crebillon's Ca“line on linguistic grounds: La versification est tres-defectueuse, pleine de termes populaires, de phrases barbares, de constructions louches, de duretes, de tours et de nombres prosaiques. [15] Clement had similar criticisms for another of Crebillon's plays, gerces (I: 159). Voltaire was of course eager to set himself above his rival in matters of linguistic purity. The ideal of noble language pursued in the Voltairean tragic text is a demanding one. Noble language must be harmonious to the ears, but absolutely correct in grammar, elevated in vocabulary and clear in meaning. Too much reasoning and eloquence leave the audience cold; the overly ornate is pompous. But the overly simple is equally inappropriate. Within a restricted and largely abstract vocabulary (we will note how Voltaire was criticized for using terms such as  42 "ciment," "crie" and "voluptueux"), the tragic playwright must produce "ces sentences admirables, placdes avec art dans les dialogues intóressants." [16] The rules about rhyme were equally rigid, and attempts to innovate were also subject to severe criticism. Mahomet, generally viewed as one of Voltaire's best plays, was dismissed by the abbó Le Blanc in the following terms: "Pour la Versification elle est fort indgale & les Rimes en sont ndgligêes A son ordinaire." [17] Voltaire's use of vers croj.sés in Tancrede provoked considerable debate among critics (ed. Moland, V: 494). Voltaire, while seeking to develop an individual style, did not protest against these limits on language, as Diderot did; on the contrary, he revered and defended them. He was very sensitive, and very vulnerable, to charges that in his pursuit of the ideal of noble language he imitated the French classics. His early plays in particular are full of paraphrases and even quotations from Racine, less often Corneille and Quinault [18]. But we should also recognize that Voltaire, in the course of a long career developed a reputation for verbal brilliance: "une diction si pure et si enchanteresse." (Corrpspondange litteraipe, V: 256). Even when the critic disliked the play, he admitted the excellence of the poetry. The Abbe' Le Blanc, for example, conceded in describing Brutus: "Malgre les Beaux Vers pourra bien tomber des mains de la plupart des lecteurs,"  43 (p. 142) and likewise Eriptiile: "C'est avec de beaux vers, de beaux morceaux la plus mauvaise Tragedie que J'aie encore vile." (ibid., p. 152) Furthermore, it was a frequent criticism of the later plays that they did not meet the standard Voltaire had set in his own earlier work:  [Semiramis] peche egalement par la conduite et par la versification qui est des plus faibles, pour ne pas dire mauvaise. On ne reconnait pas là Voltaire.  (Correspondance litteraire, I: 267) En general, la versification s'est trouvee eteinte; l'on n'y [dans Olympie] a pas reconnu ce colons qui caracterise tous les ouvrages de M de Voltaire. [19] Transformations in the play texts at the level of pure language (not incidental to a change in plot, as when Adelaide is deleted from les Freres ennemis or the high priest from Eriobile) allow us to reconstruct in Voltaire's tragic text this balancing act between innovation, individuality and the traditional norms. We are better able to recognize that he often reached his own goal of the memorable phrase. In beginning at the linguistic level we can, moreover, build up stronger evidence for more profound changes - shifts in characterization and plot, in implied or explicit ideology - and take some steps towards reconstructing the lost performance texts of these plays, giving them back some of the dynamic life they had in the eighteenth century.  44 Language-based criticism was not only frequent in assessments of Voltaire's own work, but also in his judgments of his classical predecessors and of his contemporaries. The ideal of noble language was not an external standard to which Voltaire adhered against his better judgment in order to succeed, but was a matter of strongly held personal belief. As he wrote to d'Argental (25 September 1764, D 12100), tragedy must be distinguished from other literature in particular by its quality of language, which must be "pur et vif". His criticism of other playwrights affords many examples of criticism directed solely at the quality of expression, the lapses from classical standards in the plays, such as in the "Dissertation sur les DrinciDales tram:Idles anciennes et modernes": Par luelle etrange corruption se pourrait-il faire au'on souffrit Darmi nous ce nombre Drodiaieux de vers dans lesffuels la syntaxe. la Dropriete des mots. la justesse de fiaures. le rhythme sont eternellement violes? (ed. Moland. V: 192) Ce rien n'est Das francais. et sert a rendre la phrase plus barbare....Si c'etait votre aveu aui me fit est prosaiaue, plat et dur, mdme dans la prose la plus simple.^(D. 193) Both comments refer to Crdbillon's Electre. He returns to this type of criticism in the Commentaires sur Corneille  45 [201: on Horace : Ne vont pas sans tristesse. Voltaire ruled that it was an "expression familiere dont il ne faut jamais se servir dans le style noble." (p. 284) We find further examples in his critique of Polyeucte:  Vous ne savez as ce que c'est une femme est du style bourgeois de la comedie. (p. 291)  Dont il est poss. Expression impropre, vicieuse; on ne peut dire etre poAsede des yeux.  S'il ne vous traite id i d'entiere confidence. Cola n'est pas francais; c'est un barbarisme de phrase. These criticisms of Crebillon and Corneille could be dismissed as mere self-serving, examples of Voltaire publicly diminishina his rivals to elevate himself. But we find the same type of critique in his private correspondence: "Dieu me aarde de voler vains fantames d'Etat fin Rodosune ii. ii!.. .Plus le us ce Corneille. plus )e le trouve le pére du galimatias, aussi bien que le !Dere du theatre." (to d'Araental, 25 Sept 1751. D 4579) Furthermore. he applied similar criteria to Frederick of Prussia's poetic diction. where he would have no motive to be over-critical: EnchaIner dans les fers est un oldonasme; enchainer seul suffit. On ne dit point faire l'or; on dit faire de l'or....Vous dites la haine embrasee! Ce mot est impropre. (17 March 1749. D 38931  46 Even those critics, such as Freron and Coll& most unlikely to flatter Voltaire seem to hold to the same ideal of tragic language, to apply the same rules, attacking him when he falls below it but acknowledging his frequent success in meeting that high standard: C'est une elegance de podsie egale, pour ainsi dire, a l'eloauence de prose de l'Orateur Romain....Les autres personnages parient aussi le lanaaae aui leur est nropre.^(Armee litteraire (17561 II: 142) La versification en est brillante. rarement neallaCe. pleine sur tout de cette harmonle. de cette musiaue transcendante. P. Clement on Semiramis. I: 1G21: This same critic savaaelv attacXed another of Voltaire's traaeales: Les vers sont negliges. ii v en a de mauvais de toute espece; de chevili4s. de iouches. de Polteux. de durs & insupportables a l'oreille de iaches. de ois aue prosaYques & aui ne sont pas 'Ilene trancois. (on Nanine. I: 192) The Abb6 Nadal was first disappointed Mr the tailure ot own 74ariamne in 1724 and then revolted Pv a r'hyme entin  asmoneen which disappeared from the text of Voltaire's successful 1725 fferode et Marianne prior to printing (Desnoiresterres. I: 317). Voltaire made no secret of his predilection tor pushina tne limits of acceptable languaae.  a characteristic noted in an earlier citation from tne Abbe  47 Le Blanc (p. 10): Vous nous perdez tous trois; le vous en averti qui rime a dementi, il rime trés-bien; il est permis d'Oter l's aux verbes in ir. (to the Count d'Argental, 25 Sept 1751, D 4579) This propensity was so well-known that critics began to describe certain types of errors in other writers' works as characteristic of his style: "Aire et tonnerre. mauvaise rime Voltairienne." (P. Clement. II: 30) Certain kinds of textual shift, however. appear to offer little matter for our analysis. There seem to be no ideoloaical overtones, no clue to performance, no response to external reaction, in a chanae from "sait connaitre" (1733) to "salt respecter" (1736) and back to "salt connaitre" (1775) in act II. scene iii of Zaire or from "voix tombante" (1733. 1736) to "voix tremblante" (1775) in act V. scene vi of the same play r201. Similar examples could be given for almost any Voltairean nlay in this study. as in les Guebres. where we find shifts such as "nous servons" to "J'obeis" (I. iii; VI: 510) and "Ii etait bon. sensible, ardent, mais genereux" in the place of "Emoorte. mais sensible; ardent. mais genereux" (IV. i; VI: 545) where perhaps a loss of rhetorical finesse in the replacement of the Parallel structure may be observed. The printed editions of Olympie are in many instances thick with purely typoaraohical alternations: ses for les. la for ma.  48  frdmiriez for fremirez (where it is not clear a tense shift was intended). One cannot even be assured who is responsible for these, whether it was the author, editor or compositor. Anne Sanderson's studies of Irene and Eriphile focus largely on the manuscript variants, from entire speeches to shifts in choice of adjective or in singular for plural [22]. She finds that the more mechanical revisions were often incidental to a specific change in plot or character (p. 169), as in the plays currently under study the removal of the high priest from Eriphile or the elimination of Varus from Herode et Mariamne. Sanderson finds in her manuscript analysis evidence for her conclusion that Voltaire was not as effortless a versemaker nor as careless a playwright as he has frequently been classed (pp. 165-66). Her meticulous work on Irene and Eriphile gives us a basis for judging what the frequent references in Voltaire's correspondence to having "refait" or "refondu" certain scenes or acts might entail in textual transformations as opposed to the equally frequent references to "polishing" a play text. But it is those changes in characterization, plot and even ideological slant which precipitated both drastic and slight changes in text, and which can be reconstructed from those changes, which interest us. It is thus possible to argue shifts in performance or ideology from very small linguistic changes, so opening up  49 opportunities for a good deal of additional analysis. For example in Zulime: (III, iv) Benassar to Ramire "Arrete" (in 1761) becomes "Demeure" (in 1775), implying that in the first case Ramire is meant to move offstage before Benassar begins to speak, and in the second that he is not to move. When in act V scene i Zulime shifts from calling Benassar "mon 'Dere" (1761) to "seigneur" (1775), there is an increased emphasis on Zulime's submission to her father's authority rather than on the emotional tie, and therefore an identifiable ideological component in the change. Many of the printed variants to Zaire lend themselves to this kind of analysis, as Eva Jacobs' new edition makes clear [23]. For example, the 1736 edition contains stage directions (i.e. in II, iii and V, x) that are missing from both the edition of 1733 and the edition encadree of 1775, suggesting an effort to give readers a sense of the performance, rather than the text of a plya they have just seen (in the case of the 1733 editions) or a text meant to be read. In the same way the 1774 edition of Olympie follows in many instances the edits made for performance, being distinguished by the number of verses omitted as well as changed by reference both to the 1769 editions and the Kehl edition (in II, iv; III, i; IV, iii, vii, viii; V, v, vi, vii). The 1775 edition of Zaire shows traces of increasing nationalism; for example, the Crusaders are referred to as "franc:4W instead of "francs" or "chretiens" (in I, iv an T1, i). pis  50 tendency to emphasize past glory as support for present patriotism was much more pronounced in the 1765 transformation of Adelaide du Guesclin. In Olympie, the change from "J'attends, puisqu'il le faut..." to "Faites venir ici ces deux rivaux cruels" (V, iii; VI: 158) illustrates how the role was shifted from the "fille de quinze ans" of Voltaire's original conception to a heroine more suited to Mlle Clairon's histrionic talents, providing some concrete evidence of the influence of stage realization on the text. Voltaire himself seems to have believed that small linguistic shifts could have considerable implications for plot or characterization. In order to make Seide's parricide in Mahomet more plausible, and in response to criticism from his "anges" on this topic, Voltaire suggested the following [the underlined words are not in the final printed text, the passages in square brackets are]: Pour ce grand attentat je reponds de Seide; C'est le seul instrument d'un pareil homicide. Otage de Zopire, ii peut seul aujourd'hui L'approcher a toute heure, et te venger de lui. [L'aborder en secret] Tes autres favoris, pour remplir ta vengeance, [zeles avec prudence] Pour s'exposer a tout ont trop d'experience; La jeunesse imprudent a plus d'illusions  51 [est le temps de ces] Seide est enivre de superstitions, [Et Seide, enivre de] edition de 1742 [tout en proie aux] edition encadree Jeune, ardent. devord du zele qui l'inspire. [C'est un lion docile a la voix qui le guide] (Act II, scene vi; to Formont, 10 August 1741, D 2525) Voltaire put similar emphasis on slight variation in advising La Noue on Semiramis (27 July 1748, D 3727) and continued the fine-tuning of Oreste for years: Pour rendre cet instinct plus vraisembable et plus attendrissant ii n'y a qu'un vers & changer. Electre dit: vient qu'il s'attendrit? Je l'entends qui soupire. Voici ce qu'il faut mettre a la place: Oreste 0 malheureuse Electre! Electre Ii me nomme, 11 soupire. (To the Count d'Argental, October 1758, D 7988) No alteration was too small to insist on, although the precise improvement Voltaire hoped to achieve is not always evident. In a list of changes sent to Cideville and Formont for a proposed edition of Eriphile (8 May 1732, D 486) we find "cette loin becoming "cet ordre," "De cet Etat tremblant" changing to "De l'Etat qui chanc6le" (I, i),  52 "Detestable aux mortels" shifting to "Deteste des morts meme" (end of act IV). Thirty years later, in Le  Triumvirat, (Julie to Fulvie, II, iv) Voltaire insists that "Je me meurs" replace "Je succombe" merely to avoid a repetition of the verb (to the Count and Countess d'Argental, 22 June 1764, D 11943). Emphasis on pure literary style does not appear to separate the text from its dramatic function; Le Triumvirat was at the time of the correction about to open at the Come-die francaise. Throughout the development of the texts, their linguistic correctness remained a major issue, and Voltaire continued to depend on the judgment of his friends to establish or confirm it: "Votre critique du vers, ont ecrit dans le sang, est tres-juste" he wrote to the Count d'Argental in September 1751 (D 4579). "Void i come je corrige en cet endroit: Achevez son naufrage; allez, braves amis, Les destins du senat en vos mains sont remis etc."  (Rome sauvee, II, vi) [24] Voltaire even complained to these friends if he considered their criticism to have been cursory, as he did to Vauvenargues with regard to Semiramis (May 1746, D 3398). Indeed Voltaire argued that this dialogue between the author and his cenacle was essential: "Dans la plupart de nos auteurs on trouve rarement six vers de suite qui n'aient de pareils defauts...parce qu'ils ont la presomption de ne  53  consulter personne, ou l'indocilite de ne profiter d'aucun avis." [25] Voltaire attributes his own dramaturgical practice to Racine's example, a favourite justification for any contentious theatrical praxis on his part: [Racine] joignait a un travail infini une grande connaissance de la tragedie grecque, une etude continuelle de ses beautes et de celles de leur langue et de la n6tre; 11 consultait de plus les juges les plus severes, les plus eclaires, et qui lui etaient sincerement attaches; il les ecoutait avec docilite. (ibid, p. 195) Voltaire himself had a surprisingly faithful circle of such well-education and sympathetic judges for his work. Evidence of such consultation abounds. With regard to Rome  sauvee, which seems to have caused him particular difficulty, he sent a hemistich, "tyran par la parole," to the Abbe d'Olivet his old teacher for approval of the trope. The master's approval was returned to the "pupil," now well into his dramatic career on the same paper (October 1749, D 4041 and 4042). Even later in Voltaire's career, during the composition of Le Triumvirat, the correspondence tells the same story: Independamment des vers raboteux dont la tragedie des coupe-jarrets fourmille, il y en a aussi d'assez incorrects qui ont echappe a la rapidite du mauvais style, comme par exemple au troisieme acte a la  54 premiere scene, ii y a << Ces fers qui ont approche du grand Pompee,>> et autres sottises pareilles. (to the Count and Countess d'Argental, 27 July 1763, D 11320) Voltaire then entered into a prolonged debate with the d'Argentals over one line, and indeed over one word in this play: "A deux voluptueux a livre l'univers." (Fulvie on Antoine and Octave; I, i; emphasis mine) To the d'Argentals Voltaire argued historical justification for the term (7 September 1763, D 11401) and promoted the value of innovation in tragic diction: "Il est beau de hasarder sur le theatre des terms heureux qu'on n'y a jamais employes. Au nom de Dieu, ne touchez jamais a ce vers." (18 September 1763, D 11422) But the line was eliminated from the text in respect to their judgment, as was another: "L'ardeur de me venger ne m'en fait point accroire," for which Voltaire pleaded (27 September 1763, D 11429) and for which now the intended location is uncertain [26]. In Mahomet, he did however successfully defend certain passages from alteration. Voltaire had considerable regard for Cideville's judgment; in one letter (19 July 1741, D 2515) Voltaire not only expressed agreement with certain criticisms, he claimed to have incorporated them without question. It is those lines of the play for which he argues against the opinions of Cideville which help to illuminate  55 the critera by which Voltaire judged tragic language: Celui qui a fait un examen Si approfondi et si juste de Mahomet est seul capable de faire la piece....I1 y a bien d'autres details dont je vous remercie; mais, au lieu de les discuter, je vais les corriger. Je ne sais ce que vous voulez dire d'un a l'invincible Omar; y a Et l'invincible Omar, et ton amant peut-etre Ce peut-étre me parait un correctif necessaire pour un jeune home qui se fait de fête avec Mahomet et Omar. Je ne trouve point le mot de ciment de l'amitie (II, v) bas, et j'avoue que j'aime fort haine inveteree; crie encore & son përe (I, i) me parait aussi, je vous l'avoue, bien superieur a invoque enor son nére. L'un peint et donne une idee precise. L'autre est vague. La metaphore des flambeaux de la haine consumes des mains du Temps (I, i) me parait encore tres-exacte L'insecte insensible (I, iv) n'est pas l'insecte qui ne sent pas, mais qui n'est pas senti. L'indigne partage me parait aussi mauvais qu'a vous. Cideville appears to have given Voltaire the same kind of criticism directed against Voltaire that he himself applied to Crebillon and Corneille: the level of vocabulary must be uniformly noble, the figures of speech logical and precise, the dialogue apt to the character and the situation  56  in the judgment of a classically educated man. As no change was too small to insist on, in the same way no term was too incidental to be defended. The noun "rivale," applied by Herode's sister Salome to Marianne (II, ii; II: 183), shocked the Count and Countess d'Argental, perhaps by an implication of incestuous affection or a suggestion of the other Herod who married his brother's wife. But Voltaire replied: C'est precisement cette rivalite dont il s'agit. C'est de quoi Salome est piqude; et une femme a qui on joue ce tour dit volontiers a son adverse partie ce qu'elle a sur le coeur. (23 December 1762, D 10885) The potential violation of bienseances was sufficiently justified, at least in Voltaire's opinion, by psychological plausibility, as was Seide's "peut-étre" in the previous exchange with Cideville. Voltaire had a notorious tendency to revise on the basis of rehearsals, which he supervised whenever possible (from the Comedie frangaise to Frederick's court), and to insist on even slight changes. The well-known anecdote of the partridge pie full of corrections to Zaire is merely representative (ed. Jacobs, p. 328). These changes, by their sheer number and the resistance made by the actors to them, seem to have been primarily at the level of language, requiring re-learning of lines rather than re-thinking of characterization. The Correspondance, in the  57 following examples, can show us what kind of retouches were done. In the case of Zu1), Voltaire took the less colourful but perhaps more reliable route of sending them through Mlle Quinault, one of his staunchest allies at the Comddie frangaise until her retirement. He even specifies that she, with d'Argental and Pont de Veyle, should have "une autoritd absolue" to criticize the play (29 October 1739, D 2102). Act V in particular was subject to considerable revision. The final lines of Zulime to Ramire: Dans ces derniers moments apprends A me connaitre Vois quelle dtait Zulime, et rougis d'être un traitre became Je t'aimais innocent, je t'aimai parricide; Je t'aime encor, barbare, et je te laisse Atide. This change was made while the play was in rehearsal (letter to Mlle Quinault, 26 March 1739, D1957); Voltaire considered the new couplet "plus passionnd, plus vrai, et moms commun"  (ibid) We might also deem it more sentimental, showing Zulime's suicide not as an act of courage and reproach for having loved someone unworthy of her, but as an act of utter self-abnegation. Neither of these couplets appears in the printed text, which was not established until after Zulime's revival in 1762. In another letter to Mlle Quinault during this same period of rehearsal, changes were made to Atide's speech in the same scene which ends: "Ah! donnez-moi la mort par haine ou par pitid!" Voltaire added:  58 N'armez point cette main si chere et si sacrde Contre un coeur qui, sans moi, vous aurait adoree; C'est votre amant, helas! S'il a pu vous trahir, S'il m'aime, si je meurs, le peut-on mieux punir? and then a new exchange between Ramire: Au nom de mes fortaits, soyez inexorable. Frappe. and Zulime: Je vais percer le coeur le plus coupable. [27] These alterations imply that the original 1740 denouement entailed Atide and Ramire offering to die at Zulime's hand for their crimes against her, before Zulime stabbed herself as the most guilty party. In the final play text Zulime asks her father to punish them, but is overcome by Atide's attempted suicide: "C'est a moi de mourir, puisque c'est toi qu'on aime." (V, iii; IV: 65) The aim of these 1739-40 corrections was thus to polish, to make the climactic scene more emotionally plausible and touching, an effort which would continue until the text was set after the 1762 revival. Voltaire's primary goal was thus not innovation but the firmly classical desire, following Racine, to "ddchirer le coeur" of the spectator. As Pomeau observed: "Voltaire cherche l'emotion thdatrale. Dans ses tragedies tout est sacrifid a la production de la scene dmouvante." [28] This extensive process of "pre-screening" and revising  59 the text before its debut was not infallible. One of Voltaire's few first-night disasters was Addleade du Guesclin in 1734, in which VendOme's replique "Es-tu content, Coucy?" provoked the two predictable responses: "Couci-couci!" and a roar of laughter from the parterre [29]. This four was so devasting and so notorious that the play was revived under different character names (to avoid the pun) seventeen years later, and not revived in its original setting until 1765. The Comódie frangaise pit was known for such witticisms, as when during the debut of Marmontel's Cléopatre, the realistically hissing snake was greeted by "Je suis de l'avis de l'aspic." [30] Voltaire's protectiveness for his texts would lead us to expect protests against linguistic changes made in performance which were not submitted to him for approval; he might welcome collaboration but not independence, response but not license. He showed particular hostility towards actors editing for performance. A suggestion from La Noue regarding Oreste was summarily dismissed in Voltaire's reply: "Ii me semble que ce terme d'adroite n'est pas assez noble, et sent la comedie." (27 July 1748, D 3727) Voltaire was equally distressed by Mlle Clairon's liberties with her role in Tancrede: Lekain m'a mande qu'il avait en vain combattu Mlle Clairon quand elle...m'etriquait le second acte auquel la derniere scene est absolument necessaire, quand elle  60 ecourtait ses fureurs. (to the Count and Countess d'Argental, October 1760, D 9360) Voltaire had coached Mlle Clairon in some detail at the time of the play's debut (January 1750; D 4095, 4098, 4099, 4104). Indeed, these earlier letters assumed her compliance with the "father's" wishes: Le role d'Electre est certainement votre triomphe; mais je suis pare, et dans le plaisir extreme que je ressens des compliments que tout un public enchante fait a ma fille, je lui ferai encore quelques petites observations pardonnables l'amitie paternelle. (12 January 1750, D 4095) He then goes on to coach her in fine points of declamation, for example: Dans votre imprecation contre le tyran: L'innocent doit périr, le crime est trop heureux, vous n'appuyez pas assez....L'impetueuse Electre ne dolt avoir, en cet endroit, qu'un désespoir furieux, précipité, et éclatant. Au dernier hemistiche pesez sur crk. (ibid) Voltaire evidently believed that he should be able to shape even the details of the performance, in order that the audience see and hear exactly the character he had written. This later letter demonstrates that he expected her to continue under his paternal authority.  61 He was further incensed by the cutting of "J'en suis indigne" from Zulime's role in Mlle Clairon's revival, seeing in it the loss of what was in his opinion the character's great line, equivalent to Orosmane's "Zaire, vous pleurez" or Mahomet's "Ii est donc des remords" (to the Count and Countess d'Argental, 26 January 1762, D 10282). With regard to the earlier printed edition of the same play, he fulminated against J'abjure un lache amour qui me tient sous sa loi. for^  vous ravit ma foi.  (V, iii; to the Count and Countess d'Argental, 15 August 1761, D 9945) He went so far as to blame Freron in order to indicate the strength of his dislike for the expression. It is clear from the correspondence that the Zulime and the  aire performed in 1762 with Mlle Clairon were not as we  now read them. Voltaire felt that the actors were crippling  Zaire, but does not give details (to the Count and Countess d'Argental, 28 September 1762, D 10734). The end of Zulime seems to have been particularly subject to editing by the actors of the Come:lie frangaise, without affecting the plot or the characterization substantially (see letters to the d'Argental 28 September 1762 D 10734, 6 February 1763, D 10985, and 13 February 1763 D 10999). Voltaire objected strongly to these edits: Je vous demande en grace, quand vous ferez jouer Zulime  62 Mlle Durancy, de la lui faire jouer comme je l'ai faite, et non pas come Mlle Clairon l'a joude....La maniere dont les comódiens de Paris jouent cette scene est de Brioche. (D 10999) In particular, Zulime's final lines, as established by Mlle Clairon in 1762, distressed Voltaire. The Mercure (janvier 1762, II: 202) gave the text of Zulime's last speech: A la fin j'ai rempli mon devoir. (A son 'Dere) 0 vous, seul des mortels regrettd par Zulime, Souvenez-vous de moi, mais oubliez mon crime. (A Atide) Je meurs sans vous hair. (A Ramire) Ramire, sois heureux Aux depens de ma vie, aux depens de mes feux. It was this closing line which drew Voltaire's attention, and, as we have seen previously, Voltaire's protest was based on linguistic considerations: Comment ces malheureux ignorent-ils assez leur langue pour ne pas savoir que cette repetition, aux depens, c'est une suspension, que la phrase n'est pas finie, et que cette terminaison, aux depens de mes  feux, est de la derniere platitude? (to the count and countess d'Argental, 19 December 1766, D 13746)  63 Indeed, the printed text for this passage, while incorporating elements of this performed text, varies considerably. The offending line is retained but moved back into a previous speech where, apparently, its lack of conclusion would be less noticeable (IV: 64). Zulime's dying lines were now : A la fin j'ai rempli mon devoir. Je l'aurais dü plus tft...Pardonnez A Zulime... Souvenez-vous de moi, mais oubliez mon crime. (V, iii; IV: 65; suspension points are in the text) There was a similar grammatical criticism of the performed AdelaXde of 1765. Voltaire insisted (to Lekain, 29 November 1765, D 13010) that two verses be changed in the printed edition: Gardez d'être rdduit au hasard dangereux Que les chefs de l'Etat ne trahissent leurs voeux. This couplet offended him by its semantic incoherence, apparently the result of "coupures" by the actors. He sent the correction - an expansion from one sentence to two using the same elements - directly to the widow Duchéne, the printer (30 November 1765, D 13015): Tous les chefs de l'Etat, lassós de ces ravages, Cherchent un port tranquille aprés tant de nauf rages. Ne vous exposez point au hasard dangereux De vous voir ou trahir ou prOvenir par eux. Olympie seems to have suffered likewise at the hands of  64 the Comddie francaise, a text freed from Voltaire's direct supervision by his residence in Switzerland in a way the early plays were not. The Moland edition indicates 77 lines cut in performance, with nearly all these cuts followed in the 1774 Paris edition which also escaped Voltaire's supervision. Of these lines, 44 are taken from the roles of Antigone and Cassandre (the two rivals for Olympie), and 27 from Cassandre alone. The edits lack political motivation; it was not to please the censors that Cassandre's outbursts of remorse were curtailed in I, ii and III, i. A more forceful conception of the character and a fear of wearying the audience with discourse not advancing the action appear much more likely motivations. It is thus clear from the examples given that Voltaire continued to seek the seventeenth century's pure and elevated language in his tragedies. He seems to have relied on his friends' and his old teachers' judgment in matters of linguistic correctness, appearing to lack confidence in his own in isolation on the correctness of a word, phrase, or figure of speech. He was, however, capable of ignoring or disputing their judgment, and seems to have had scant respect for the opinion of actors in this regard, and indeed for their performance skills generally. Innovation was extremely difficult in all aspects of eighteenth-century dramaturgy, not least in the linguistic. Apostolides views this resistance to change as evidence of  65  the social value and function of theatre, emphasizing theatre as a form of ceremonial: En ce sens, nous ne devons pas interpreter les régles d'unite du poeme dramatique come tine contrainte arbitraire imposee aux auteurs par quelques pedants, mais bien plutft come un ensemble de recettes, connues et admises de taus, qui visent a renforcer l'aspect <<acte>> du ceremonial....La representation thóátrale s'impose avec tant de force que, meme si le public salt que tout est simulacre, ii supporte difficilement la mise en scene de certaines images interdites  a  sa conscience. [31]  The storm of criticism which surrounded Marivaux' expansion of comedic vocabulary is well-known, and his language was frequently dismissed as an artificial jargon [32]. Voltaire's own limited innovations were greeted by the same resistance. Rehearsal, in which Voltaire could hear the text, in which an actor could protest difficulties in diction or breathing, offered Voltaire an ideal opportunity to polish the language, and incidentally annoy the actors with a plethora of alterations to memorize. Not least to Voltaire himself was the tragic text a fundamentally open-ended one, constantly subject to revision and external judgment. In examining whom Voltaire consulted in this domain of linguistic correctness and effectiveness, we note that he  66 sought and frequently followed the advice of a small and relatively constant circle of friends, his cdnacle. As noble language was a key element in tragedy's literary prestige, Voltaire consulted those whose tastes and education reflected his own background in the classics. The only woman evident in this circle, particularly after the loss of Mme du Chatelet, was the Countess d'Argental. Voltaire speaks of her contribution in a very revealing way: "C'est a vous que j'en Buis redevable, c'est A votre goat, l'interêt que vous avec pris a l'ouvrage,  a  vos rdflexions,  aussi solides que fines." (18 October 1760, D 9327) "Le goat" falls into the feminine jurisdiction of sentiment, in opposition to male reason: "Le Goat est d'une grande êtendue; il fait appercevoir d'une maniére vive et prompte, sans qu'il ne coate rien a la Raison, tout ce qu'il y a voir dans chaque chose." [33] As a woman, Mme d'Argental's opinions spring as much from instinct as from intellect. Thus gender can be seen as a limiting factor for influence on Voltaire's play texts. Unfortunately, there is little documentation of Mme du Chatelet's role if any in the development of Voltaire's tragic texts, beyond the dedication of Alzire to her, and her known participation in the private performance of some of his plays, such as the now fragmentary Thdrêse (ed. Moland, IV: 259). ^Those who have studied the intellectual relationship between Voltaire and Mme du Chatelet, such as Wade and Vaillot, tend to  67 conclude that her influence drew Voltaire away from theatre and towards science and philosophy. As Wade concluded, "Only in the drama does the influence of Cirey seem negligible." [34] Her absence from the tragic text seems by her choice. As far as the contribution of actors can be discerned, Voltaire does not seem to have sought or incorporated their advice on language, and indeed opposed any shift in their sphere of influence beyond strict performance, that is to say, their use of declamation and gesture. Even in this domain, Voltaire expended great effort on coaching performance, and thus on retaining control or at least supervision of "his" text. He seems to have strongly resisted incorporating the edits made for performance into the printed texts he supervised. While actors and actresses could affect the immediate form and content of the tragic text by their interpretations and editing, and so manipulate the audience response, the enduring printed text largely conceals when it does not eliminate these contributions (as in the case of Olympie). Social class, and the consequent lack of access to a classical education, appears to bar the performers from immortalization in print within such a noble genre. At best, Voltaire might give them the credit of a dedication, as in that of Zulime to Mlle Clairon in which he says "Cette tragedie vous appartient, mademoiselle; vous l'avez fait supporter au thdátre." (IV: 6) But we must  68 keep in mind that the text which follows is not wholly the one the actress made successful on stage, and indeed is one in which her alterations have been "corrected". Voltaire's own concern with details of style and grammar justifies ours. These transformations at the level of language will make it possible for us to argue more convincingly considerable shifts in character conceptualization or in ideology, insofar as these changes occur at more than one level in the text. For example, an increased emphasis on nationalism may not be seen in an increased number of lines on the subject, but more subtly in the reiterated shift to "frangais" in describing the Crusader knights of Zaire. We have seen that linguistic purity was not the sole motive for linguistic change. Voltaire explicitly indicated that he sought to increase the emotional impact, the psychological plausibility and / or the literary distinction of a line when small changes were made: "plus passionne, plus vrai et moms commun." (26 March 1739, D 1957) Linguistic changes may prove to be incidental to larger revisions to plot or character, as Sanderson's detailed studies of Eriphile and Irene indicate. But they can also imply a change in casting as in Rome sauvee, a shift in ideological focus as in Adelaide, or a reassertion of authorial posssession of the text as in Zulime and Olympie. This perpetual alternation between the desire to collaborate and the desire to control, between  69 dialogue and monologue, in the Voltairean text constitutes one of its most fascinating elements, and one which we can see reflected at all levels. The ambivalence towards outside criticism, which is both sought and disregarded, is coloured by Voltaire's opinion of the particular source. Literary critics, performers and even audiences could be dismissed far more easily than his circle of friends and advisors, that almost exclusively male group many of whom Voltaire had known from their days together at the College des Jesuites de Louis-le-Grand. ^It is their voices, Cideville and Formont and the Count d'Argental, which we can hear under that of Voltaire, the importance of their advice and encouragement to the development of the works as they now stand which we recognize. The exclusion of women from higher education, the ambivalent position of performers in eighteenth-century France (who were both excommunicates and comèdiens du roi) and the social stigma attached to actresses (whose greed and licentiousness were presumed) serve as the context for their lack of participation in tragedy as a literary form. Without concluding that the play texts are directly autobiographical, we can state that they were written consistently from a place of gender and (relative) class privilege, expressing the world view of educated men who had access to literary recognition as well as popular acclaim through tragedy as a "noble" medium.  70  NOTES [1] Suzanne K. Langer, "Discursive and Presentational Forms," in Semiotics: An Introductory Anthology ed. Robert E. Innis (Bloomington: Indiana UP, 1985). [2] Langer, p. 106. [3] V. N. Volosinov, "Verbal Interaction" in  Semiotics: An Introductory Anthology, ed. Robert E. Innis, p. 61. [4] Maitó Albistur and Daniel Armogathe, Histoire du  fdminisme frapgais du moyen age a nos jours (Paris: Editions des femmes, 1977), p. 134. [5] Julia Penelope, Speaking Freely: Unlearning the  Lies of the Fathers' Tongues (in the Athene series, New York: Pergamon Press, 1990) p. 17.^Mary Daly,  Gyn/Ecology: The Metaethics of Radical Feminism (Boston: Beacon Press, 1978) ie at pp. 164-65 provides numerous examples of how patriarchal societies use women to enforce the social rules restricting women. Margaret Atwood illustrates this principle powerfully in The Handmaid's Tale by means of the Aunts, who teach the imprisoned handmaids to accept their roles as sexual surrogates (Toronto: McLelland Stewart, 1985). This characteristeric of male-dominated systems suggests that the function of female characters in Voltaire's tragedies is likely to be a prescriptive one, illustrative of what female behaviour in that society  71 should be. This suggestion will be examined in the chapter on ideology in text. [6] Roseann Runte, "Women as Muses," in French Women  and the Age of Enlightenment, ed. Samia I. Spencer (Bloomington: Indiana UP, 1984), p. 144 [7] Mme de Lambert, "Reflexions nouvelles sur les femmes" in Ecritures fdminines, ed. Elyane Dezon-Jones (Paris: Editions Magnard, 1983) P. 96.  [8] Annde littdraire, ed. E. C. Freron (1761) VII: 14. [9] English Showalter Jr., "Writing Off the Stage: Women Authors and Eighteenth-Century Theater," Yale French  Studies 75 (1988): pp. 95-96. ^His review of the status of women playwrights in the period concludes: "The regularity with which women abandoned the theater after one play points to a powerful deterrent at work." (p. 96) This article challenges the more positive presentation in Barbara J. Mittman's article "Women and the Theatre Arts" in French  Women and the Age of Enlightenment, ed. Samia I. Spencer (Bloomington: Indiana UP, 1984), pp. 155 - 169. [10] Showalter Jr points out that "from 1717 to 1749 no new plays by women were produced," and indicates that Mme de Graffigny's Cdnie (1750) and the failed Fine d'Aristide (1758) "were the last full-length plays at the Comedie Frangaise by a woman until the Revolution." (p. 96)  [11] Art poetique, chant III.  72 [12] Voltaire provides a long list of Racine's "faults" as part of his defense of simplicity in the preface to Les Scythes, Oeuvres completes, ed. Moland (Paris: Gamier freres, 1877), VI, p. 273. Unless otherwise indicated, all references to Voltaire's works will be to this edition and will appear in the body of the text with volume and page number. References to his Correspondence will be to the 1968 Oxford edition edited by Theodore Besterman, and will give the date, the Besterman number, and the recipient as needed for clarity. [13] Gustave Desnoiresterres, Voltaire et la societe au  XVIIIe siècle 2e ed. (Paris: Didier, 1871), III: 201. [14] Correspondance l'tteraire ed. Maurice Tourneux (Paris: Gamier freres, 1877) II: 464. [15]  lime ClinErt, I OKI aptE's  litittaires PrOdam Slathe, WV; mint cf tre 1Th al din  edition), I: 131. [16] all from Voltaire's letter to the marquis Albergati Capacelli, 23 December 1760, D 9492. [17] Hóléne Monod-Cassidy, Un Voyageur-philosophe  au Ville siècle: l'abbe Jean-Bernard Le Blanc (Cambridge: Harvard UP, 1941), p. 378. [18] Henri Lion, Les Tragedies et les theories  dramatigues de Voltaire (Geneva: Slatkine, 1970; reproduction of the 1895 Paris edition). See in particular his discussions of the early tragedies:  Oedipe, Artemire, Mariamne.  73 [19] Louis Petit de Bachaumont, Memoires secrets  Dour servir  a  l'histoire de la republique des lettres en  France depuis MDCCLXII jusqu'a nos jours (London: John Adamson, 1777), II: 35.  [20] Oeuvres completes, ed. Theodore Besterman (Oxford: Voltaire Foundation, 1968), volume 54. [21] Eva Jacobs, ed., Zaire in Les Oeuvres  completes de Voltaire volume 8 (Oxford UP: The Voltaire Foundation, 1988), pp. 457, 513. [22] Anne Sanderson, "Voltaire and the problem of dramatic structure: the evolution of the form of  Eriphyle," and "In the playwright's workshop: Voltaire's corrections to Irene," in SVEC CCXXVIII (1984): pp. 97 170 Sanderson's approach operates from a different theoretical framework than mine, and does not for example consider the ideological implications of text shifts. [23] in The Complete Works of Voltaire (Oxford: Voltaire Foundation, 1988), volume 8. [24] This correction represents an intermediate stage in the text. The edition encadree does not include either version. [25] "Dissertation sur les principes tragedies anciennes et modernes," ed. Moland, V: 194. [26] Voltaire's editor Moland speculates that the line would have appeared in I, i, spoken by Fulvie (VI: 291).  74 [27] to Mlle Quinault, 17 February 1740, D 2164, prior to the play's debut; all changes are with regard to the final scene of act V. [28] René Pomeau, La Religion de Voltaire (Paris: Nizet, 1956), p. 257. [29] Similar reactions, but to violations of behavioural norms, occurred at the first performance of  Mariamne (where she drank poison on stage): "La reine boit!" and in an earlier scene of Adelaide when Nemours appeared with his arm in a sling. [30] J. B. Clement and Joseph de la Porte, Anecdotes  dramatiques (Genéve: Slatkine, 1971; reprint of the 1775 Paris edition), pp. 209-10. [31] Jean-Marie Apostolides, Le Prince sacrifid:  Thdatro et politique au temps de Louis XIV (Paris: Minuit, 1985), p. 47. [32] See Freddric Deloffre, Marivaux et le  marivaudage, une prdciosite nouyelle, 2 ed. (Paris: Armand Colin, 1967). [33] Mme de Lambert, p. 97. [34] Ira 0. Wade, Voltaire and Mme du Chatelet (New York: Octagon, 1967), p. 193. See also René Vaillot,  Avec Mme du Chatelet, Voltaire en son temps 2, ed. René Pomeau (Oxford UP, 1988).  75 Chapter II - Textual Transformation and the Dialogue of Performance: the Struggle for Acclaim and Control  Performance is not external or incidental to the texts under examination, but inseparable from them. Theatre, perhaps even more strongly than text meant to be read, is a situation of dialogue and interaction among senders and receivers: the author, the actors and the audience. The meaning constructed from the performed text, and the reception afforded to it by audiences, can vary greatly from those attributed by its readers, particularly those reading without conscious reference to a potential stage realization. However, the question of reception is one which does not  seen  to have received the same degree of  critical scrutiny as has been devoted to theatrical form and function [1]. We will see how performance and performers, both virtual and actual, have had considerable impact on the texts of Voltaire's tragedies. Voltaire's efforts to control not only dramatic text but also performance text were indicated in the previous chapter by his opposition to editing by the actors, to their changing the text by reducing it. But this continuing involvement in performance text has broad implications not only for the plays as they now appear but for eighteenth-century theatre generally. One can certainly argue that Voltaire's practical interest  76 in staging - as an avid amateur actor and metteur en scene, his "training" of certain performers, as well as his writing contributed to changing audience expectations and standards for judging of tragedies. He did not only respond to audience reaction, he strove to guide it along specific lines. Xeir Elam has provided a useful framework for the analysis of performance dynamics, emphasizing rather than downplaying the role of the spectator: Theatre is a situation of communication initiated by the spectator, whose mere arrival and presence sets up the process. [2] Patrice Pavis concurs: "From this point on, the mise en scene is no longer (or at least no longer entirely) an indication of the intentionality of the director, but a structuring by the spectator of materials presented." [3] Elam further reminds us that theatrical communication is multilayered; the dialogue between actor and spectator is overlaid by that between character and character. Various systems of communication: verbal, gestural, visual and aural operate together. Everything that happens on stage is potentially meaningful, not just what is spoken or pantomimed [4]. Our present distance from the sociocultural context of these eighteenth-century plays of course increases our difficulty in understanding them: The performance text is dependent for its encoding and decoding both on a flexible number of systems and on a  77 set of codes more or less common to the sources, performers and audience. [5] The theatre as a building had its own impact on text as well as performance: the number of spectators, their closeness to each other and to the stage, the size and shape of the stage [6]. We have already noted the interactive role played by the French parterre - all male, standing closely packed at the edge of the stage - in performances at the Comedie frangaise (page 38). The eighteenth century saw the development of increasingly elaborate decor and effects (strongly influenced by opera), of the curtain used between acts and at the close, and in France of the clear stage in 1759, a reform which had a considerable impact on theatre as spectacle by permitting crowd scenes, ghosts, funeral pyres, trapdoor entrances and exits. Voltaire, especially, was quick to exploit the opportunities for spectacle in later plays like Tancrbde and Olympie, having long deplored the constraints imposed by on-stage seating [7]. Furthermore, we cannot neglect the extent to which the presence of a certain actor in and of itself may have considerable significance to the construction of meaning by the audience. In Voltaire's theatre, popular success was frequently attributed to or seemed genuinely due to the presence and talents of particular performers: Mlle Gaussin in Zaire, Mlle Dumesnil in Merope, Mlle Clairon in  78 Sdmiramis, Zulime, and Olympie and Lekain in Adelaide and Mahomet. This "meaning" constructed by the audience must also include the associations the audience might bring to the text because of its beliefs or expectations of the author, clearly an important factor in the popular success or failure of Voltaire's plays. A certain intellectual parti pris - for or against the philosophes - generally appeared to underlie a championing of Crdbillon's genius over Voltaire's, for example by Frdron, Colle, or Marivaux. The cabales pro- and anti-Voltaire were frequently given credit for determining the general public's reaction to a Voltaire play, whether to prevent its debut, cut short its run, or unduly prolong its run. Plays that succeed in the context of performed theatre are in many cases not those given literary acclaim at the time of performance or in later critical judgments. Audiences, as opposed to critics who work primarily from printed text, can have different criteria and construct meaning from performance text in different ways. Eighteenth-century critics, including Voltaire, were aware of this dichotomy, and frequently expressed a certain contempt for mere "popular" success. With regard to the direct impact of performance on text, Elam argues that it must be presumed for any dramatic text: Literary critics have usually implicitly or explicitly  79 assumed the priority of the written play over the performance....But it is equally legitimate to claim that it is the performance, or at least a possible or 'model' performance that constrains the dramatic text in its very articulation.. ..The written text.. .is determined by its very need for stage contextualization, and indicates throughout its allegiance to the physical conditions of performance.^[8] An eighteenth-century French critic put it more briefly: "Tout drame est essentiellement fait pour 'etre mis, ou du moms imagine au Théâtre." [9]  I. The Tension between Performance Success and Literary Merit Voltaire, while never ceasing to strive for theatrical popularity, was well aware that an initial favourable reception did not necessarily imply critical accolades: Je suis bien loin de m'enorgueillir du succes passager de quelques representations. Qui ne connait l'illusion du th6atre? Qui ne sait qu'une situation interessante, mais triviale, une nouveaute brillante et hasard6e, la seule voix d'une actrice, suffisent pour tromper quelque temps le public? Quelle distance immense entre un ouvrage souffert au theatre et un bon ouvrage! (D 517, Voltaire writing in the Mercure on the debut  80 of Zaire, August 1732)  Public reception of Voltaire's tragedies varied widely over the years and the results were often not what Voltaire or his friends would have predicted, or what the twentieth-century reader might expect. For example,  Artemire, despite Adrienne LeCouvreur's encouragement of the writer and her personal performance in the play, failed miserably. As the Mercure predicted, the author hesitated to preserve in print the evidence of a public disgrace: Come le Public bien inform6 scait [sic] d'avance quelle a etd l'avanture de l'une...Si cependant l'Auteur d'Artemire s'avise de faire imprimer sa piece, peut etre nous aviserons-nous d'en donner un extrait. [10] We note that Le Mercure did not hesitate to be spiteful towards a novice playwright. Voltaire reportedly jumped up in his loge during the opening performance and lectured the catcalling parterre on its bad judgment; the play was revived for a second performance only at the insistence of the Regent's mother. Nonetheless, it initially attained eight performances, scarcely a complete disaster [11]. However, it was never subsequently revived, and only the text of Artemire's role survives.  Mariamne, another vehicle for LeCouvreur, had excellent advance publicity in 1724: Attendue du Public avec tant d'impatience...[on a ose]  81 exiger le double du prix ordinaire....On Préta beaucoup d'attention pendant les trois premiers Actes, et dans une partie du quatrieme^[Puis sont survenus] ces tumultes Si ordinaires...le cinquibme Acte [in which Mariamne drank poison and died on stage] fut le plus maltraite ...[On a crie] la Reine bolt! (le Mercure, March 1724, pp. 529-30) The Mercure had had to base its extract on one performance, in which the fifth act could not be heard over the parterre's hoots. As if the jest "la Reine bolt!" were not enough, another followed: Ii est d'usage qulaprés une Tragedie, on donne une petite Comedie. On joua, ce jour-la le Deuil. AussitOt quelqu'un s"ecria: C'est le deuil de la Place nouvelle. Ce mot plaisant decida la chute de la Place. [12] Voltaire's attempt at innovation in the poisoning scene had been firmly rejected. But the play was not unsalvageable: "L'année suivante, l'auteur rechangea le denouement, & elle eut alors beaucoup de succbs." (J-B Clement, p. 522) Voltaire learned from the audience's initial rejection, as he would throughout his career, and the audience rewarded him: Les aplaudissements qu'on a donnes a vostre Mariamne ont retenti jusqu'icy....La reussite de vostre pi6ce a  82 ete complette....Ceux qui les ont ecrites parlent avec eloge et de la verite des caracteres et de l'elegance de la diction. us paroissent frappes entr'autres endroits et du portrait ressemblant des femmes et des fureurs touchantes d'Herode. (Cideville to Voltaire, 13 APril 1725, D 229) While public reation was rarely consistent or predictable for any author, few playwrights were as provocative as Voltaire. Response might not be favourable, but it was usually strong. Zaire, now considered one of his best tragedies, was at the time of its debut "beaucoup critiquee" and "encore plus applaudie" (Le Mercure August 1732, p. 1828). Voltaire had the rare privilege of critiquing it himself (pp. 1829ff). Despite his already impious reputation, audiences and censors came to consider the play so pious that it, rather than the traditional "Christian" tragedy Polyeucte, was the final play performed when the theatre closed for Lent in 1734. Adelaide, likewise, was "aussi extraordinairement applaudie que sevbrement critiquee par une trbs nombreuse assemblee...mieux entendu, plus goate et plus applaudi & la seconde representation...apr6s quelques changements faits par l'Auteur sur les observations du Public." (Le Mercure, January 1734, pp. 239-40) Despite what Le Mercure reported, it did not, however, approach Zaire's enthusiastic reception until Lekain's 1765 revival. Indeed, Le Blanc saw little  83 difference between the two performances, despite the reputed extent of the alterations: L'Addlaide de Voltaire fut cruellement sifflee a la ire Representation. Ii l'a corrigee depuis mais elle...sera toujours une Piece avec de grandes beautes, sans caracteres, sans moeurs....I1 a fallu refaire les trois derniers actes. Ses Amis lui disoient aussi de rechanger les deux Premiers. [13]  Zulime, too, was badly received until Mlle Clairon revived it in 1762. All this provides evidence for Elam's contention that the actor can be of more importance than the play text in attracting a favourable audience reception for the performance. One can also see in this phenomenon some explanation of why most of these tragedies disappeared from the repertoire in the early nineteenth century, with the loss of those actors who had made them successful and changes not only to public tastes but also to acting technique. Critics of the period not only disagreed on the merit of a particular play but also on the nature of the public reaction it received. Rome sauvde, according to Pierre Clement, se joue avec succes; car c'est le Parterre qui fait les succes....I1 n'y a peut etre pas de Piece de Mr. de Voltaire plus radieuse que celle-ci...[avec] tout l'dclat de son coloris....Le role de Cicdron a dtd  84 universellement applaudi; celui de Catilina lui est entierement sacrifie; celui d'Aurelie, femme de Catilina, a de grandes beautes; le plus brillant de tous est celui de Cesar. (II: 239-40) But Freron reported of the same performances that "Voltaire lui-meme convient que sa tragedie paroit plutOt faite pour etre lue par les sgavans que pour etre vue par le Parterre. On ne peut qu'approuver ce jugement." [14] Oreste , after initial difficulties and revisions, became as popular as the Crebillon Electre it presumed to correct. Mlle Clairon, as Electre, made of it a personal triumph in 1761. Voltaire saw in this success evidence of the capriciousness of audience reaction: Je ne sais plus comment la nation est faite; elle souffre une Electre de quarante ans qui ne fait point l'amour et qui remplit son caractere; elle ne siffle pas une piece ou il n'y a point de partie carree: s'est donc fait dans les esprits un prodigieux changement! (17 July 1761 D 9902) Oreste also offers an illustration of how the byzantine intrigues within the Comedie frangaise, while largely separate from the literary cabals and from what one might call public opinion, were nonetheless significant as far as the choice of plays, casting and revivals of plays were concerned. The number of performances, while useful as a counterbalance to critical opinion, is not an infallible  85 indicator of audience preferences, insofar as it could be increased or decreased by actors' preferences, based on the length or quality of their own roles. Voltaire's Oreste was unexpectedly dropped from the Comedie frangaise repertoire after its 1762 triumph; the explanation offered was that Brizard had a better role in Crebillon's Electre and as a full member had the power to "empècher qu'une bonne tragedie de M de Voltaire ne füt sur le repertoire." [15] When  Oreste was revived for the debut of Mlle Desperrieres, in the late 1770's, the play "n'avait ete joue qu'une fois depuis quatorze ans, au debut de mile Durancy." (La Harpe, X: 389) The reception afforded Mahomet offers a similar history of peaks and valleys. It was driven off the stage in 1742 by accusations of impiety (rather than by popular indifference or critical attack), but was enthusiastically received when revived only ten years later, without having become either more pious or less provocative: Son cher Proph6te a ete donc rejoue a la fin, et trouve horriblement beau. (P. Clement, II: 159) Beaucoup de gens tres eclaires pretendent que c'est celle de ses pi6ces oü ii y a plus de beautes de detail, et des iddes plus sublimes. (Le Mercure November 1751 p. 142) Its goal, "rendre adieux le fanatisme," was apparently not seen as a threat to established religion (ibid) Even Le  86 Blanc condemned the play on literary rather than ideological grounds: "Le Pretendu miracle qui fait le denoament de la Piece ne m'a paru que pueril.... [Elle n'est] pas la plus mauvaise Tragedie que je connoisse mais la plus extravagante." (p. 378) In this revival, Lekain appeared as Seide, one of his first roles at the Comedie francaise, and "impressiona vivement la salle entiere par la force de son jeu" (Desnoiresterres, IV: 191), perhaps the first indication of his later ability to make his role the play's focus.  Semiramis, like Artemire and Mariamne long before, was eagerly anticipated by the public; the theatre was booked weeks in advance. But the public proved to be most interested not in text strictly speaking, but in the increased visual spectacle, the ghost and the mechanically produced thunder (le Mercure, September 1748, p. 224) The ghost, an unmitigated disaster in Eriphile where it could hardly make its way forward through the spectators on stage, remained a contentious innovation for the critics: Ii a voulu donner du spectacle, il a rassamble tous les prodiges, tout le merveilleux qu'il a pu; mais, malgre tout cet appareil, le spectateur froid a juge sans retour la piece et l'a raise au rang des plus mediocres tragedies.... Elle ne sera jamais qu'un conte de revenants. [12]  87 Aux premieres representations de cette Piece, on badina beaucoup sur l'Ombre de Minus.... [Azema says] Quoi! tous les morts en cet affreux sejour, Pour nous persecuter reviennent-ils au jour? (act V) On ne s'appercut pas a la premiere representation, du ridicule que ces deux vers repandoient sur la Piece; mais a la seconde, ii en resulta un éclat de rire en choeur dans le Parterre. L'Auteur n'a eu garde de les laisser a la troisieme. (J-B Clement, pp. 162-63) But on the whole the audience disregarded the critics and welcomed the innovation: "Semiramis est Si peu tombee qu'elle aura quinze ou vingt representations si l'Auteur ne la retire." (P. Clement, I: 103) Voltaire did not embrace all innovation. He expressed misgivings about the taste "tant soit peu anglais" of Lekain appearing with bare bloodstained arms in a later revival, as he would oppose Mlle Clairon's desire for a scaffold in Tancrbde (1760), but Lekain had read correctly the shift in audience tastes towards brutally vivid spectacle (9 August 1756, D 6965). This increased audience desire for visual spectacle culminated, as far as Voltaire's theatre is concerned, in Olympie, with its act V funeral pyre on which Olympie throws herself. Its Comedie francaise debut occurred on 17 March  88 1764, and it was still playing when le Mercure published a notice in April, mentioning "la beaute et la splendeur de son spectacle" (p. 195). Thanks to a lengthy analysis in the May 1764 issue, an act by act summary, we learn precisely when the temple doors were opened and closed during the performance - more innovation in staging that took advantage of the clear stage (pp. 170-91). Its success lay, according to the Mercure's analysis, where that of all true theatre does, in the number of spectators attending (p. 192). Voltaire himself argued the necessity of visual impact: "Tachez de parler a la fois aux yeux, aux orellies, et a l'Ame; on critiquera mais ce sera en pleurant." (D 11811) D' Alembert himself, we learn, did not think much of the play, but he reported that the performance was all that Voltaire could ask: Son grand succes...des changements heureux. Le role de Statira et celui de l'hierophante sont beaux, celui de Cassandre a des moments de chaleur...celui d'Androgide et d'Olympie m'ont paru faibles; mais Mlle Clairon y est admirable au dernier acte. (6 April 1764, D 11814) The growing role of spectacle, and the rise of certain actors as stars capable of carrying a production, would seem to reduce the importance of the text per se, the text as a piece of poetry written in conventionalized "noble" language. But only audiences appeared to put less emphasis on the prestige of text; writers and critics, who were often  89 the same people, remained loyal to the traditional elevation of literary above visual elements. When popular reaction did not correspond to critical judgment, the public simply had to be wrong. Voltaire and his critics were acutely aware of the persistent gap between literary merit as they judged it and the reception afforded by the general public. The process by which the Comódie frangaise selected their repertoire appeared to operate by a third set of standards. The critics, as we see in Voltaire's article on Zaire in Le  Mercure, have a general tendency to give more weight to their own "literary" judgments as more objective and more lasting. Critical accolades, however, were not sufficient to keep a play on stage. One of the most striking illustrations of this point in Voltaire's theatre is provided by Eriphilg. Le Mercure praises it at length: Pleine d'harmonie, et d'elOgance dans les Vers et de pensees nobles et dlevdes, la diction en est male ... les images, les reflexions, les maximes neuves et hardies^extrémement applaudie par de nombreuses assemblaes....Eriphile est de la composition de l'illustre M de Voltaire, connu dans l'Europe come le seul Poete Epique de nos jours. (March 1732, p. 562) The review calls Eriphile a "Tragádie dans un gout entibrement nouveau (p. 563).. .le goilt Grec" (p. 564),  90 quoting in full Eriphile's confession: "cet age fatal et sans experience..." (pp. 564-65) and Alcmdon's speech on birth: "ce qui m'accable et qui me desespere" (pp. 570-01). Yet the play was not revived or printed during Voltaire's lifetime. Indeed, journal criticism seems to have had comparatively little effect on popular reception. Independent of the critical debates over plot structure and plausibility, Zaire's popular success was assured by the quality of performance. In Voltaire's opinion, there was never a play so well done as Zaire at its fourth night (D 515) In fact, Voltaire fretted that its success was attributed to Mlle Gaussin's beauty, to the acting and the spectacular costumes, rather than to his own writing (D 526). It was an accusation which would be levelled at others of his plays as well. As the years passed, Voltaire seems indeed to have developed a contempt for audiences that could be dazzled by mere spectacle rather than educated reason. In this he approached a belief that mediocrity would fare better with this public, as for example his Trimvirat: Je crois que Si Mlle Dumesnil jouait bien Fulvie, et Mlle Clairon pathetiquement Julie, la piece pourrait faire assez. (16 December 1762, D 10843) Je m'etais follement imagine que la chaleur de la representation sauverait mes fautes....Si la piece  91 telle qu'elle est aujourd'hui, dtait bien jouee a Fontainebleau, elle pourrait reprendre faveur. (20 August 1764, D 12056). The long and tumultous career of Zulime demonstrated, to Voltaire's satisfaction at least, the disassociation of performance success from literary merit: Je commence a esperer beaucoup de succes de cet ouvrage aux representations, parce que c'est une piece dans laquelle les acteurs peuvent deployer tous les mouvements des passions et une tragedie dolt etre des passions parlantes. (12 March 1740, D 2180) [Elle] peut reussir parce qu'on y pane continuellement d'une chose qui plait assez genóralement; mais ii n'y a ni invention ni caracteres ni situations extraordinaires; on y aime a la rage, Clairon joue et c'est tout. (on the 1762 revival, D 10253) The 1762 revival was in fact a considerable popular success, largely attributed to Mlle Clairon in the title role [17]. The successful revival of Adelaide in 1765 again confirmed Voltaire's contempt for public opinion as a reliable indicator of a play's worth: Vous vous etes donc mis, monseigneur, a ressusciter les morts? Vous avez deterre je ne sais quelle Adelaide morte en sa naissance, et que j'avais empaillee pour la deguiser en Duc de Foix. Vous lui avez donne la plus  92 belle vie du monde. (to Richelieu D 12886) Quand vous miapprites, monsieur, quion jouait a Paris une Adelaide du Guesclin avec quelque succés, j'etais trbs loin d'imaginer que ce filt la mienne....I1 y avait plus de trente ans que j'avais hasarde devant ce public une Adelaide du Guesclin....Elle fut sifflee des le premier acte, les siff lets redoublbrent au second. (September/October 1765, D 12909) Voltaire referred to Nemours' sling and the cannon shot as provoking whistles, with the shout of Couci-couci the final blow [18]. He described le Duc de Foix as Adelaide "devenue plus mauvaise, [elle] reussit assez." (ibid). This last  Adelaide was, in Voltaire's eyes, not even an improvement on the original: us liont representee telle quills l'avaient donnde en 1734.. .et elle a &be acceuillie avec beaucoup diapplaudissements; les endroits qui avaient ete le plus siffles, ont ete ceux qui ont excite le plus de battements de mains....Tour a tour sifflees et bien regues, les opinions ont ainsi flottó dans les affaires serieuses, comme dans les beaux arts et dans les sciences. (ibid) But we find a different account in the memoirs of Bachaumont who, as a spectator of the revival, recognized the extent of changes made, although he believed Voltaire had been responsible:  93 Depuis le succes du Siege du Calais...M de Voltaire a juge a propos de rapprocher de nouveau l'dpoque de sa Tragedie, pour la rendre plus interessante & de la restituer sous les premiers noms....Le succes a ete complet: le coup de canon a fait le plus grand effet. La marche rendue plus rapide, l'interet plus pressant, un grand nombre de beaux vers ajoutes, des noms plus illustres & chers...joint aux beautes dont l'ouvrage etoit déjà rempli, a transporte les spectateurs. (II:^256) Not consulted by Lekain before the revival, Voltaire refused, however, all responsibility for the performance text: Je ne connais plus du tout cette Adelaide dont vous me dites tant de bien: ii y a trente ans que je l'ai oubliee. Ii plOt alors au public de la condamner; plait au public d'aujourd'hui de l'applaudir, et il me plait a moi de rire de ces inconstances. (D 12918) The knotty question of whether or not the 1765 Adelaide was actually identical to the original is greatly simplified by the new critical edition prepared by Michael Cartwright (Oxford, 1985). Cartwright argues in his introduction to the texts that the 1765 version represents a compilation of the previous ones. A comparison, using Cartwright's tables of concordance, shows that the 1765 printed edition differed considerably from that of 1734, with at least 400 lines  94 changed or omitted. The plot and characterization remain, however, largely unchanged: two brothers pursue the same woman, the one who finds the other in his power resolves to kill him, a tragic denouement is averted by a faithful friend. Voltaire's assertion is, thus, justifiable but incomplete, and it suggests the beginning of his efforts to bring the play text back under his control, minimizing, for example, Lekain's contribution. It is of course in the very nature of theatre that audience reception is variable and transitory, even as the dramatic text itself is not static, but renewed by each fresh performance. Voltaire's not always successful efforts to change audience standards - in favour of "Greek" models and visual pageantry - may indicate that his resentment of public opinion, like that expressed by his critics, might stem from the audience's resistance to their didactic influence rather than from the inherent inconstancy of public opinion. From the time of Artemire, Voltaire sought to "educate" public tastes by his plays; he was willing to alter according to audience reaction, but not to accept the audience's judgment as decisive, for example in his reworking of Mariamne into Herode et Mariamne, and of  Adelaide du Guesclin into Le Duc de Foix. The history of Voltaire's plays in performance is therefore one of tremendous struggle to influence and even control public opinion for or against the play, and of the  95 great wars of the cabals and the critics. Voltaire's editor Moland, for example, recounts the battle over Oreste, in which the playwright complained to the play's patroness the Duchess du Maine that her absence from the debut had been one reason for the public's lukewarm reception. He persuaded her to attend the second performance, and attended himself with his "partisans" in the parterre (see the introduction to the play, Moland V: 74 - 75). According to La Harpe, Voltaire's career was a continual series of such triumphs and defeats in a perpetual conflict with his chief rival Cróbillon and his literary enemies: Brutus fails and Zaire succeeds, Adelaide is avenged by Alzire, the failure of Zulime and the banning of Mahomet are balanced by the triumph of Mdrope (La Harpe, IV: 388 - 89). After initially encountering resistance from critics and audiences, Voltaire's versions of Cróbillon's themes ended up largely replacing the originals in the repertoire: Sdmiramis...pleine de beautd supdrieures et vraiment tragiques fut sifflee a la premiere representation Oreste fut encore plus mal recu....Le Catilina et la Semiramis de Crdbillon sont dans l'eternel oubli; la Semiramis de Voltaire est en possession du theatre; son Oreste y est applaudi, et sa Rome sauvde est sue par coeur de tous les amateurs de la belle poósie. (La Harpe, XI: 83)  96 Voltaire, perceiving himself as having to struggle against the cabals as well as the censors, resorted to semi-transparent pseudonyms and elaborate games of concealment and denial around the authorship of some plays. This practice began as early as 1736, with L'Enfant  prodique, and was typical of his later plays like  Le,  Triumvirat and Les Guebres: Le jeune auteur des Guebres West venu trouver; il a beaucoup ajoute a son ouvrage....Mais tous ses soins et toute sa sagesse ne desarmeront probablement pas les  prétres de Pluton. On etait pres de jouer cette piece a Lyon; la seule crainte de l'archeveque...a rendu les empressements des comediens inutiles. (to the count d'Argental, 30 August 1769, D 15855) Those who disagreed with Voltaire's ideas or disliked his plays had a firm belief in the existence and vigour of the cabal which strove to keep his plays before the public. Cone believed that Voltaire's partisans had bought the success of Oreste by paying a claque (I: 120, 125) and "ont remue ciel et terre pour qu'on reprit [Mahomet]; ils sont plus fanatiques de Voltaire que Seide ne l'est de Mahomet." (I: 352) Writing from this cynical perspective, Freron deplored Voltaire's pretense of anonymity. Le Triumvirat, for example, might have been better received as coming from the great Voltaire rather than from some unknown youth. He pointed out, with barbed politeness, the number and  97 enthusiasm of Voltaire's admirers in Paris: "On est Si prevenu en faveur de tout ce qui sort de sa fertile plume." (Armee litteraire (1767) VII: 94) He advised that Voltaire should always admit authorship, so that his work would be judged by his reputation rather than its own merit (ibid.). Bachaumont made much the same comment on the appearance of Les Scythes: "Elle est déjà affichee sous le nom de Voltaire, ce qui lui 6te toute ressource de la nier, et sembleroit devoir lui en garantir le succés." [19] Even the Correspondance littóraire, usually favourable to Voltaire, concludes with regard to Le Triumvirat: M de Voltaire eut tort de garder ainsi l'incognito. Si les heros n'ont pas besoin d' n'en est pas ainsi de certains enfants faibles qui ont besoin de la gloire de leurs pbres pour etre toldres. (VII: 210) The tension between these opposing forces seems to have been roughly balanced, however. The records of the Comedie francaise indicate that Voltaire's work was profitable and frequently revived; he was, in fact, the Comedie's most successful living playwright during the eighteenth century [20].  II. Authorial Control of the Performance Text Voltaire's struggle for authorial control over the text - in its language, its reception and its printing - has already come under discussion in this study (pages 59-64).  98 Another aspect of the struggle - control over the performance text - can now be added. Voltaire was not content to have a voice in casting and to attend rehearsals, as was the right of any playwright at the Comedie francaise. Voltaire's resentment of actors' edits and the lengths to which he would go to ensure the acceptance of his own revisions are indicative of the extent to which he involved himself with the performance of his texts. In a very real sense Voltaire became an influential eighteenth-century metteur en scene - staging his plays in his own or his friends' homes, acting himself, choosing costumes and decor, coaching professional actors. In this regard, although it is not a comparison Voltaire sought, he resembled Moliere the actor-director-playwright rather than his chosen model Racine, who had a distinctly more limited notion of the playwright's role. There is not doubt that Voltaire's interventionist style of playwriting had considerable impact on eighteenthcentury French stage production, increasing the emphasis on historically plausible costume and visually striking decor. The theatre was moving away from Racine's almost abstract simplicity. Voltaire's taste for spectacle was sometimes even seen as a lapse from classical taste: A la fin du mois [apparaitra] la Semiramis de M de Voltaire, avec tout son spectacle....I1 y aura de la Magie et surtout du tonnerre, car M de Voltaire l'aime  99 beaucoup; il en a mis jusques dans la Merope pour augmenter la terreur et la pitie. (P. Clement, II: 93) But on the whole, eighteenth-century critics viewed spectacle as one of his major contributions to tragic theatre: Des effets plus profonds, plus puissants, plus varies a tirer de la terreur et de la pitid; des moeurs nouvelles a etaler sur la scene, en soumettant toutes les nations au domaine de la tragedie; un plus grand appareil de representation...les grands verites de la morale. (La Harpe, IV: 337 - 38) Voltaire's efforts to control performance text included a lifelong enthusiasm for staging plays himself. Several of his earlier plays were done at the home of Mme de Fontaine-Martel, among them Eriphile (Desnoiresterres, I: 440). Voltaire was pleased to have "attendri" and "fait verser des larmes," but he feared that act V as it stood could ruin the plays, due to the great difficulty in producing a plausible ghost (to Cideville, 3 February 1732, D 459). In a performance of Zaire, again at the home of Mme de Fontaine-Martel, Voltaire played Lusignan, one of his favourite amateur roles (to Cideville, 27 January 1733, D 564). The Lille performances of Mahomet by La Noue's troupe in 1741 offered a special situation, in which Voltaire is  100  able to influence a provincial troupe and use their performance, rather than that of a private one under his direct control, in order to polish his work for Paris (see d 2415, 2483, and 2495). It was an experience that had implications for his ongoing working relationship with the Comedie, when La Noue later joined the Paris company. Voltaire's home in the Rue Traversi6re became almost a private school for theatre arts (see Desnoiresterres, III: 372ff). The young Lekain appeared there, under Voltaire's supervision, before he was accepted by the Comedie francaise, in a performance ofMahomet in 1750. The following year at the Comedie he played Seide to La Noue's Mahomet, later advancing to the lead role. The house in the Rue Traversiere also saw performances by Mme Denis and Mme de Fontaine, who played the lead female roles in a private production of Zulime (Desnoiresterres, III: 405). These performances were also used to attract favourable publicity for upcoming Come-die francaise debuts. The Nouvelles  litteraires reported that in Mahomet "le principal acteur, nomme Le Kain, a montre un talent distingue," and that the new version of Zulime was now "ni froide, ni faiblement ecrite" (I: 436). Rome sauvee was also put on at the Rue Traversiere, and both it and Zulime were played at Sceaux in 1750 before the Duchess du Maine. Mme Denis continued to play Zulime, just as Voltaire liked to play Lusignan, Mohadir and Ciceron even after the couple were established  101  at les Delices. Voltaire was no less active a director during his stay in Prussia: Jules Cesar, Rome auvee done with himself as Ciceron, the special adaptation of Adelaide as Les Fréres  ennemis (with no female role), and even Zaire with Voltaire as Lusignan of course (see Desnoiresterres III: 453ff and the correspondence of 1751). Voltaire sent revisions to  Rome sauvee based on the Berlin performances to the Comedie frangise for the revival in 1752. The patriarch's life in Switzerland continued to be graced with private performances of old favourites and new works in progress; continuing revisions resulted. The visits of his pupils presented special theatrical occasions. When Lekain visited les Delices in April 1755, Zaire was performed for the Tronchin family, with Voltaire as Lusignan. Voltaire marked another Lekain visit, more than fifteen years later in 1772, by arranging performances in Geneva of the actor's major roles: Arzace in Semiramis, VendOme in Adelaide, Mahomet. In attending these, Voltaire saw Semiramis done on a clear stage for the first time (Desnoiresterres, VII: 430), as he had not been to Paris after the stage was emptied of spectators by Lauragais' reforms in 1759.  Zulime was a frequent choice for Voltaire's own stage, where he worked out his extensive revisions. The name of the play's main character was particularly subject to  102 change: Zulime became Fanime, and then briefly Menime. The goal was always to send the play back to Paris and to see it succeed there: Nous l'avons joude a Lausanne [Zulime]....Tout ce que je souhaite, c'est qu'elle soit aussi bien joude A Paris, me n'ai jamais vu verser tant de lames. (to Thibouville, 20 March 1757, D 7208) Si vous voulez vous amuser, conduisez cette Fanime avec le fidele d'Argental....Si la piace est bien joude, elle pourra amuser votre Paris. (to Mme de Fontaine, June 1757, D 7290) Nous la jouames hier, et avec un nouveau succes. Je jouais Mohadir; nous dtions tous habilles come les maitres de l'univers. (to the Count d'Argental, 26 February 1758, d 7652) Cette Fanime vous fait fonder en lames, du moms Mme Denis fait cet effet. (to the Countess d'Argental, 13 October 1760, D 9306) These performances were not simply entertainment en famille. Voltaire staged a successful performance of Zulime before the Duke de Villars, and the intendants of Bourgogne of Languedoc. He then reported to Mlle Clairon that her role of Zulime was now "plus decent et par consequent plus attendrissant"; the suicide in the last act "fait un effet terrible" (16 October 1760, D 9317; see also D 9331, D 9346 and 0 9350). He was evidently seeking her support for a  103 revival at the Comedie. Modestly, he described himself to friends as having played "assez pathetiquement," adding, however, that he had failed to attain the "sublime de Mme Denis" (To the count d'Argental, 1 November 1760, D 9372). But despite this triumph, Voltaire was not ready to send the play to Paris; he was still struggling with the stage management of act V, and trying to decide whether Zulime should kill herself with her own dagger or with Atide's (to the Count and Countess d'Argental, 26 November 1760, D 9425). Even after the 1762 Paris revival of the play, Voltaire remained much involved, trying to direct the difficult final scene from a distance - sending instructions to the count and countess d'Argental for this production (February 1762, D 10311 and D 10314), and later for Mlle Durancy (19 December 1766, D 13746).  Olympie, which was the last Voltairean tragedy to achieve popular success, was first performed, complete with funeral pyre, at Voltaire's home in March 1762. He reported to the Duke de Villars that "la pitie et la terreur ótaient au comble. Les larmes ont coule pendant toute la pi6ce." (25 March 1762, D 10388) As with Zulime, Voltaire was not directing for the mere pleasure of performance; these private theatricals were an essential part of his praxis, his workshop for new ideas: Je fis jouer cette famille d'Alexandre le jour que je vous envoyai le quatribme acte; je m'apergus que  104 Statira, en s'evanouissant sur le theatre, tuait la piece....Je vis encore clairement que le duel propose a. la fin du troisieme devenait ridicule au comencement du quatrieme. (to Chauvelin, 17 October 1762, D 10770; see also D 10746, D 10748 and D 10754) Olympie was done at Schwetzingen for the duchess of Saxe-Gotha on 30 September 1762, under Collini's direction but with Voltaire's detailed advice: Ce qu'il y a de plus necessaire, c'est que l'actrice chargee du role d'Olympie soit tres attendrissante, qu'elle soupire, qu'elle sanglote....I1 faut au moms deux ou trois secondes en recitant: Apprends...que je t' que je m'en punis. (30 August 1762, D 10682) Voltaire also discussed with Collini various techniques for staging Olympie's spectacular death, according to the stage machinery available (ibid). Even after Olympie's debut in Paris, Voltaire believed he could do better. He directed a performance in Geneva in 1766 and reported: "Elle n'a jamais eu un si grand succes." (3 November 1766, D 13644) We note from the letter that this Olympie used Cramer's text, published under Voltaire's supervision, rather than the Paris text published by the widow Duchene. Virtually from the beginning of his working  105 relationship with the Comedie francaise, Voltaire strove to persuade actors to do his play his way. He coached the young Mlle Dangeville as Tullie in Brutus, one of her first roles, and praised her efforts in spite of the play's lack of success (Desnoiresterres, I: 417). More experienced actors, however, resented Voltaire's interference; we have already mentioned the ploy of the partridge pie devised in order to persuade Dufresne to accept more alterations to Zaire. Bachaumont reports that "les comddiens sont excddds de ses extravangances, et disent qu'ils ne veulent plus jouer de ses pieces. ..[& cause des] corrections qu'il fait a chaque reprdsentation, et qu'il les force d'apprendre." (I: 152) But Voltaire's interventionist approach was supported by many. Voltaire was pressed by his friends to come back to Paris in order to supervise Rome sauvde and ensure its success (Desnoiresterres, IV: 191). When the comddiens refused to accept the author's corrections to that play as relayed by Mme Denis, she appealed to Richelieu directly, as the government minister responsible for the Comádie, to enforce them (ibid.). In its review of Tancrede, the Correspondance litteraire deplores Voltaire's absence: C'est un grand inconvenient que le podte soit cent lieues du theatre oa il est joud. Je suis persuade que si M de Voltaire avait pu assister a la premiere representation de sa piece, il ^rendue admirable,  106 pour l'effet, A la seconde. (IV: 292) In Semiramis, Voltaire sought to re-introduce the innovation of a ghost without offending his audience's expectations for tragedy as Eriphile had. It was a step beyond his previous reworkings, the second Mariamne and the  Duc de Foix (from Addlalde), where he completely removed similar "unacceptable" elements. He hovered over the production, and attended rehearsals: "us Wont fait pleurer, us Wont fait frissonner." (27 June 1748, D 3678) He fretted over the spectre's costume, feeling that white would be more imposing than black (15 August 1748, D 3732). The spectators on stage so impeded the ghost's entrance that he sought police intervention (see letters to the marquis d'Argenson and to Berryer, 30 August 1748, D 3737). He was unhappy with the debut, particularly since Sarrasin and La Moue had rejected his advice. He reportedly watched the play in disguise to see from direct experience what revisions were needed [21]. But the Fontainebleau performances offered him an opportunity to impose his revisions on the actors, including "cent vers nouvellement corriges" for Mlle Dumesnil as Semiramis. Voltaire continued to criticize the decor, especially Minus' tomb (to the count d'Argental, September 1748, D 3761 and D 3766). Upon the play's revival, Voltaire relied on Lekain to give  Semiramis the visual impact he had originally wanted: "observer le costume, rendre l'action theatrale, et etaler  107 sur la scene toute la pompe convenable." (4 August 1756, D 6959). The revival, while doing little to improve  Semiramis' standing as mediocre literature, represented a tremendous personal triumph for Lekain. Voltaire came to rely on Lekain to uphold his wishes among the actors: "Recommendez bien au fidele Lekain d'empecher qu'on n'Otrique l'dtoffe, qu'on ne la coupe, qu'on ne la recouse avec des vers welches." (22 June 1764, D 11943) In Mlle Clairon, Voltaire found another talented if not always obedient pupil. He instructed her, for example, on the fine points of Electre in Oreste (January 1750; D 4095, 4098, 4099, 4104), a play he considered easier to perform than its contemporary Rome sauvde (to the duchess du Maine, 2 January 1751, D 4085). He went into great detail in his advice: Pressez sans ddclamer quelques endroits come: Sans trouble, sans remords, Egisthe renouvelle... Dans votre imprecation contre le tyran: L'innocent doit pdrir, le crime est trop heureux, vous n'appuyez pas assez....Au dernier hdmistiche pesez sur cri.... La nature en tout temps est funeste en ces lieux. Vous avez mis l'accent sur fu....Vous ne sauriez trop ddployer les deux morceaux du quatrieme et du cinquieme actes. (12 January 1750, D 4095) and  108 Ii y en a deux qui exigent une esp6ce de declamation qui n'appartient qui& vous....que la voix se deploie d'une maniére pompeuse et terrible. (c. 15 January 1750, D 4098) referring to passages in act IV, scene iv and act V, scene vi. He concludes: Tout le sublime de la declamation dans ces deux l'accablement de la douleur & l'emportement de la vengeance...les reproches, les sanglots, l'abandonnement du desespoir...voil& ce que vous mettez dans votre role. (ibid.) When the play was successfully revived, Voltaire continued his efforts to direct the performance, advising that Oreste's fureurs would be more difficult to convey than Mile Clairon's as Amenaide in Tancrbde because he is alone while she was before the assembly of nobles (16 October 1760, D 93a8'). The playwright complained also about Clairon's editing of act II and of her fureurs in acts IV and V of Tancrede (28 October 1760, D 9360). Despite this distressing independence, he wanted her cast as Elextre, with Mlle Dumesnil as Clytemnestre (22 December 1760, D 9485). On the whole, Voltaire was pleased with the revival, "une tragédie grecque sans amour," with the denouement strengthened from the 1750 version (see the introduction to the Moland edition and the correspondence of July 1761). To modern eyes, Mlle Clairon might seem more independent than  109 Lekain, more concerned with the quality of her own performance than eiyh the quality or integrity of the Voltairean text; we have noted her tendency to edit, here and in her revival of ZUlime. While Voltaire saw Lekain as his ally at the Comedie, his trust was not absolute. He expressed on one occasion his wish to print Olympie so that nni Lekain ni Mlle Clairon ne mutileront mon ouvrage" (25 April 1763, D 11174). Mistrust prOved to be justified; Lekain went so far as to stage an unauthorized revival of  Adelaide in 1765, without submitting the text to Voltaire until after the first performances. The history of Adelaide du Guesplin, in text and in performance, is exceptionally complicated. As far as the Comedie frangaise is concerned, we have the 1734 original, tepidly received, and the 1752 puc de Foix version, a fair success, before the Lekain version of 1765. There is also  Les Fréres ennemip (1751) done in Prussia with no "Adelaide" role, and an unpublished manuscript called Alamire with an even more indistinct historical setting, which appears to be contemporary with it. When informed by Thieriot that an  Adelaide had been staged to popular acclaim, Voltaire almost immediately began to reassert authorial control over this unknown version: J'ai retrouve ici, dans mes paperasses, deux tragedies d'Adelaide; elles sont toutes deux fort differentes, et probablement la troisieme, qu'on a joude la Comedie,  110  diffbre beaucoup des deux autres. (to the Count and Countess d'Argental, 17 September 1765, D 12887) In this letter, Voltaire asked the d'Argentals to send him tune copie bien exacte, afin qu'en la confèrant avec les autres je pusse en faire un ouvrage supportable a la lecture, et dont le succês fut inddpendant du merite des acteurs." (ibid) That last phrase seems key to Voltaire's attitude towards performance: the text should succeed primarily as his literary text, with the actors' contribution subsumed and minimized insofar as possible. Initially Voltaire protrayed himself to Lekain as merely curious: Je me borne A obtenir une copie de l'Adelaide que vous avez fait jouer. Je voudrais surtout savoir si le duc de Nemours est reconnu rival de son frbre au troisieme ou au quatrieme acte. (12 September 1765, D 12895) He requested the performance text (23 September 1765, D 12898), and informed Lekain that "la copie quo vous m'envoyez", which presumably is the text of the successful revival, "est pleine de fautes; je les corrigerai de mon mieux." (8 October 1765, D 12922) The version of the play presented before the court must be his: "Ii sera trés-ndcessaire qu'elle soit reprêsentde a Fontainebleau avec les changements essentiels que j'y ai faits." (11 October 1760, D 12930) Lines from the Duc de Foix, which  111 the d'Argentals had missed from the Paris production, were sent to Lekain, presumably for Fontainebleau as they do not appear in the 1765 printed edition: Les quatres vers que vous regrettez, et qui commencent: II faut a son ami montrer son injustice,  sont déjà restitues, et je les ai envoyes a Lekain, qui je vous prie de faire tenir ce nouveau brimborion. (23 September 1765, D 12899) Although he must have known otherwise, with the three versions in his possession, Voltaire now began to claim that the 1765 text was none other that that of 1734: La belle reception qu'on fit a cette Adelaide du Guesclin!...Le plaisant de l'affaire, c'est qu'il n'y a pas un mot de change dans la piêce autrefois sifflee et aujourd'hui applaudie. (19 October 1765, D 12944) What is clear is that the two stage devices which had provoked laughter in 1734, Nemours in a sling and the cannon shot, were reinstated by Lekain and accepted in this production. Finally, Votaire downplayed both Lekain's judgment in choosing this play to revive and that of the public in welcoming it. The play did not read well; it therefore lacked literary value: Quant a la pauvre Adelaide, elle ne me parait pas Si heureuse a la lecture qu'a la representation. Je vois bien que vos talents l'avaient embellie. L'edition a  112 beaucoup de fautes qui ne sont point corrigees dans l'errata. (29 November 1765, D 13010) In short, we may conclude that in Voltaire's judgment it was the printed edition, not the performance, which would last and which would form the only basis for assessing intrinsic worth. Actors, as the letter states, by their talents "gild" the text so as to persuade the audience to overlook its faults. In the same way Voltaire took back control of Zulime after Mlle Clairon's triumphant revival in 1762. While the performances continued, he persisted in sending revisions and instructions on "la maniere absolument necessaire dont ii faut jouer la derniere scene." (6 February 1763, D 10985) He cast doubt on the wisdom of printing the play, despite its current success: La piece ne se vendra guere....Comment d'ailleurs la donner au public? sera-ce avec la coupures qu'on y a faites?...Ces nuances delicates dchappent aux spectateurs, et sont remarquees avec degoft par les yeux severes du lecteur. (1 February 1762, D 10301) and concluded: Au reste, le debit de Zulime est un tres-mince objet, et je doute qu'il se trouve un libraire qui en donne cinq cents livres. (13 February 1763, D 10999) The line between an actor's influence on the author and his independence from the author seems very clearly drawn for  113 drawn for Voltaire. Performance was certainly the goal of all his theatrical work, but adapting to the needs of performance by no means implied relinquishing authorial control. Although he could obviously never be entirely successful, Voltaire strove to make the performances as much "his" as a purely literary text could be "his". For Voltaire, the role of author expanded to encompass director and even performer. This broad involvement with the presentation and interpretation of theatrical text certainly increased Voltaire's impact on theatre in eighteenth-century France, an involvement he shared with his contemporary comedic playwrights Marivaux and Beaumarchais.  III. The Impact of Performance on Text Voltaire became notorious for the extent to which performance could affect his texts. As the widely quoted aphorism attributed to Fontenelle put it, "Voltaire est un auteur bien rare, il fait ses pieces a mesure qu'on les joue." [22] The first stage in this dialogic process was the private reading, which in Elam's terms we can place under the heading of virtual performance. The purpose of these private readings was not primarily to create publicity for a forthcoming play; Voltaire sought, rather, critical response from his favorite audience, well-educated friends. It was a technique typically employed for his earlier plays; it was not often feasible once he had left Paris.  114  Artdmire was first presented at the home of Adrienne Lecouvreur; an initial positive response subsequently brought the play into production (Desnoiresterres, I: 183-84). Voltaire, as the correspondence makes clear, could choose who would be present at such readings; we have, for example, the letter inviting Moncrif to a reading of  Eriptiile  (3 January 1732, D 451). Zaire was read at the  home of Mme de la Rivandiere who wept (the desired response); Voltaire further invited the judgment of Cideville and Formont (3 August 1732, D 507), as he had with regard to Eriphile: "le parterre jugera Eriphile en dernier ressort; mais je veux qu'auparavant elle soit jugee par vous et par M de Cidevillle." (18 April 1732, D 480) The playwright's interest in performance extended to these private readings, and he was proud of his skill in conveying the emotional impact of his texts: "Je lus hier Adelaide. Je n'ay jamais tant pleure fly tant fait pleurer." (27 July 1733, D 638) This technique was not wholly abandoned even after Voltaire left Paris, although his opportunities to read for well-informed and influential listeners were reduced: J'ai lu cette piece [Olympie] a M le duc de Villars .... On fondait en larmes a tous les actes; et si cela est jouet, bien joue...avec ces sanglots etouffes, ces lames involontaires, ces silences terribles, cet accablement de la douleur... qui  115 passent des mouvements des actrices dans l'Ame des ecoutants. (24 October 1761, D 10090) Not only did the audiences at these readings respond to Voltaire directly, they also discussed the plays among themselves, in very much of a workshop atmosphere: J' ay Recu Adelaide de mr Dubocage....J'ay Lu cette piece Rapidement, elle m'a fait extrémement pleurer mais vous savez que je suis un grand Larmoyeur. (Formont to Cideville, 20 November 1733, D 682) Je suis charmee qu'Adelaide vous plaise. Elle m'a touchee. Je la trouve tendre, noble, touchante, bien dcrite et surtout un cinquifte acte charmant. Elle ne sera pas joude Si tot, la pauvre petite Dufresne se meurt....Elle etait tres capable de faire valoir son role et la petite Gossein le jouerait pitoyablement. (Mme du Chatelet to Jacques Francois Paul Aldonce de Sade, December 1733, D 689) For these earlier plays, we see that Voltaire's notion of an ideal response seems to have been emotional rather than intellectual, to provoke tears rather than philosophical discussion. He was still obviously following the Racinian vision of tragedy as something "qui dechire le coeur." As Voltaire expressed it in his preface to Merope, "Le grand point est d'emouvoir et de faire verser des larmes." (ed. Moland, IV: 197)  116 Fontenelle's flippant comment had at its core a very significant observation; Voltaire's involvement in the play text did not end when it was accepted by the Comddie frangaise. The text continued to be manipulated and altered, as Voltaire responded to audience reaction and struggled to control and incorporate actorial changes. Taking the plays in chronological order, the failure of  Art4mire, based on a recent novel by Mme de Fontaines, in 1720 seems to have led to an adaptation of the same irrational husband/ dutiful wife plot structure into a setting with a historical basis well known to the audience;  Herode et Mariampe (1724). Interestingly, where a twentienth century reader would likely see a homosexual subplot in the rivalry between Artemire the wife and Pallante the "favori" for Cassandre's trust, neither Voltaire nor his audience seem to have interpreted it so. We note also in support of this interpretation that the role of rival in Berode et Mariamne went to a female character, Salomd, and that the d'Argentals are shocked to hear her describe Mariamne as a rival for Herode, emphasizing the sexual undertone to the struggle for power (23 December 1762, D 10855). This first Mariamne also failed to find acceptance, due to the flagrant violation of biens4ances in having a tragic character drink on stage. But then the Comedie frangaise audience was confronted by the abbe Nadal's attempt at treating the same subject in February of  117 1725, and suffered his version for only four performances. Voltaire took advantage of this failure, which he may even have encouraged (see a letter to the abbd Nadal, 20 March 1725, D 226), to offer his revised  MArigane.  The audience  was pleased by his incorporation of their criticisms: Le succes qu'elle a aujourd'hui le console assez [Voltaire a] retravailld sa Piece avec tant de soin qu'elle n'est pas reconnoissable.  (le Mercure, March 1725, p. 803) The March issue of the Mercure gave a new resumd of the play (pp. 801-25), and reported that: Le spectateur a le coeur attendri, L'esprit satisfait, et il est souvent en admiration... [Elle est] tres bien reprdsentde, les Acteurs voulant rdpondre A la beautd de l'ouvrage et au plaisir qu'il fait au public, se surpassent A l'envi les tins des autres. (pp. 825-26) Its success was such that numerous parodies followed (May 1725, pp. 1007 - 09), and the Comddiens frangais stopped performing Mariam= only after 18 performances "pour la redonner l'hyver prochain." (p. 1009) Between the 1724 and the 1725 versions, we see the direct impact of performance in that Mariamne's death was moved offstage, and only reported to Hdrode. No text of the original denouement seems to have survived. However, we find in the revision an accompanying loss of political  118 content. In act III, scenes between Hdrode and Varus the Roman consul, and between Hêrode and his adviser Mazael, in which Herod's status as a subject king of Rome is at issue, were redone. In act V, scene i, Mariamne no longer deplores her unhappy ancestry (last of the previous dynasty of kings), but rather her fatal beauty as the cause of her fate. In act V, scene vii, Herode curses his jealousy, and not Jerusalem and the Jews. On the whole, the plot dynamic is reduced to Mrode's vacillation between obsessive love and violent jealousy. The political role of their marriage, to legitimize Herode's rule over Palestine, is consistently downplayed. When Voltaire returned to  Marianne during his revisions  of Olympie, his previous affection had waned: "je l'ai trouvee plate et le sujet beau; je l'ai entiérement changdte." (21 July 1762, D 10597) The major innovation was reflected in the character of Soheme, an Asmonean nobleman distantly related to Mariamne: "une espéce de janseniste, essdnien de son métier, que j'ai substitud a Varus" (to the Count and Countess d'Argental, 7 August 1762, D 10636). The character Varus, serving no purpose once the Roman consul versus subject king tension was removed in 1725, should have been eliminated earlier. Calling SoMme a Jansenist not only suggested the rigid uprightness of his character, but also gave a topical flavour to the plot by alluding to the long struggle between Jansenist and Jesuit in France. There  119 were no structural changes to the play. Sohéme functions as Varus did to double the rivalry between Mariamne and Salome in that Salome seeks to deprive Mariamne both of his affection as well as of Herode's trust, and to emphasize Mariamne's virtue when she refuses to flee from Herode with him. Even so long after the success of 1725, Voltaire was obliged to meet audience expectations: Mariamne's exit in act II, scene v was considered lacking in motivation. Voltaire defended its plausibility (to the Count d'Argental, 10 November 1762, D 10797)^Curiously, this 1762 revision, while the last in Voltaire's lifetime and therefore the definitive version, did not receive the same degree of popular approval that the 1725 version had.  Eriphile offers a difficult problem to the textual analyst. According to Niklaus, the so-called "Lekain" manuscripts indicate the performance edits, presumably those of 1732 preserved by the Comedie as the play was never revived during Lekain's career [23]. The correspondence of 1732 throws light on the type of changes which were made. Voltaire's primary concern was to increase the effect of the ghost in act IV i and ii and in the final scene (May 1732, D 486 and D 490). He in fact decided to remove the high priest, and to keep the ghost despite public reaction, hoping thereby to increase the tragic interest (to Formont, 25 June 1732, D 497). This change is substantiated by the manuscripts underlying the Moland edition of the play (II:  120 505 - 530). The high priest had a largely expository role: 86 lines in 6 scenes of 3 acts, of which 56 lines come in act I. His removal had little impact on plot, and the task of exposition was given to active characters. But in  Semiramis, which was a reworking of the same plot, the role of the high priest was increased: °roes has 173 lines, including the closing speech. This type of wise old priest also reappears much later in Olympie, the role Voltaire liked to play himself, a possible clue as to why such a passive character was in the original version of Eriphilpe.  gulime offers similar difficulties in that it was not published in 1740 after the original performance. The 1761 pirate edition which preceded Mlle Clairon's revival, and which Voltaire disavowed, was accepted as a variant text by Moland. It varied dramatically from the later definitive text, and included a new ending in which Atide and not Zulime commits suicide. This ending does not seem to have been performed; Mlle Clairon and later Mlle Durancy both died tragically and conspicuously in the title role. But it was the denouement, in which Atide proposes suicide and Zulime commits it, which most troubled Voltaire, a scene which provoked the greatest number of revisions and instructions on performance. The interrelationship between text and performance is greatly developed in Nahomet. From the first stages of writing, Voltaire was concerned with the difficulty in  121 casting and acting the play, and frequently consulted the actress Mlle Quinault for advice. Revisions were made to the end of act IV, where Beide must kill his father before discovering the relationship, in an attempt to make Beide a more consistent and sympathetic character. But Voltaire continued to express doubts to Mlle Quinault that a suitable actor could be found to play Mahomet within the Comódie francaise; Dufresne, who would want the role, would be disastrous in it (16 February 1740, D 2163). ^Voltaire doubted too, as he did for his Roman plays, the play's potential for popular appeal because Maholpet was not about love. It was, he writes, "la pace des homes" whereas  Zulime was for women, who set the tone for theatrical success (19 April 1740, D 2200). The current vogue for Destouches' and La Chaussee's bourgeois comedies would have to pass if the play were to have any chance (3 June 1740, D 2218). Furthermore, Mahomet as a text appealed to the mind rather than to the senses: the actors alone would have to evoke the audience's sympathy (to the count d'Argental, 19 January 1741, D 2408) Voltaire almost seemed to rely on Mlle Quinault's professional judgment regarding the performability of the play:  Ii faut, pour l'honneur de vos predictions, qui j'envoie quantite de changements...Ce sera a vous, mademoiselle, qu'il devra sa fortune. (6 January 1741, D 2395)  122 Mahomet was tried out in Lille with a provincial company under La Moue so that Voltaire could judge his play properly and learn from audience response without running the risks of a Paris debut: "La representation m'eclairera encore, et me rendra plus severe." (to the count d'Argental, 7 April 1741, D 2459) These four performances are described as "une sentence de juges inferieurs qui pourrait bien etre cassee A votre tribunal." (to Cideveille, 27 May 1741, D 2488) While not conclusive, the response in Lille was an extremely useful tool for Voltaire. He made not more but rather fewer alterations in the text than he had thought would be required: J'aurais voulu pouvoir retrancher l'amour; mais l'execution de ce projet a toujours ete impracticable, et je me suis heureusement apercu, a la representation, que toutes les scenes de Palmire ont etd tres bien recues; [act IV when they innocently discuss killing their father] tout cela faisait au theatre un effet que je ne peux vous exprimer....Cette scene est aussi neuve qu'elle est touchante et terrible....Je n'ai vu personne qui n'ait pense ainsi, a la lecture et a la representation. (to Cideville, 19 July 1741, D 2515) Voltaire did contemplate changing the motivation for Palmire's entrance in act V, scene i. It seemed implausible that Mahomet would summon her at a moment when his whole  123 revolution is in jeopardy. In a letter to Cideville, Voltaire proposed that she enter "pour lui demander la grace de son frere; alors les bienseances sont observees, et cette action Blame de Palmire produit un coup de theatre." (19 July 1741, D 2515) The final text for this scene used the idea of Palmire seeking Selde's pardon, but introduced her entrance rather than making it the proposed coup de theatre. When at last Mahomet came to Paris, Voltaire had revised it so extensively on the basis of provincial performance that, as he writes to the Count d'Argental, "je le crois plus interessant que lorsqu'il fit pleurer les Lillois." (19 January 1742, D 2584)  Mahomet is one of the few cases where Voltaire conceded that the actor was capable of transcending weakness in the text to create a convincing illusion of sincerity. In the conclusion of the play, the prophet expresses remorse for his crimes and then stifles it for the sake of power. Voltaire suggested an alternative to d'Argental prior to the Lille performances (19 January 1741, D 2408) in which the repentance is more thorough: Delivre-moi du jour, mais cache a taus les yeux Que Mahomet coupable est faible et malheureux. This version was not incorporated, and Voltaire continued to be dissatisfied. But he had to concede that Lekain's performance in the 1751 revival succeeded admirably in concealing the flaw which his authorial powers could not:  124 "Ii y a un malheur a ce Mahomet, c'est quill finit par une pantalonnade; mais Lekain dit si bien: Ii est donc des remords!" (to the count d'Argental, 28 August 1751, D 4557)  Sdmiramis, while a great success in performance, underwent considerable revision after its initial run. The  Mercure indicates that before the revival in March 1749 lines were added to the scenes in act IV between Sdmiramis and her son Arzace, that Assur was made a more dignified villain if not a more adept one, and that the ending was improved by Semiramis' abdication in her son's favour and Assur's entrance in chains before Ninias (Le Mercure, April 1749, p. 205) There is no such denouement in the text as it stands now. The printed ending is the one in which Lekain was later so successful, in which Arzace/Ninias emerges from his father's tomb with bloodstained arms, having accidentally killed his mother (V, vi).  SOmiramis narrowly avoided Addlaide's fate, to be ruined by a witticism. The well-known jest "Place l'ombre!" was provoked by an incident in act IV. Furthermore, Azóma's lines in Act V: Tous les morts, en cet affreux sejour Pour nous persdcuter reviennent-ils au jour? elicited laughter at the second performance, and were removed at the third (P. Clement, I: 103). The revisions made seem to have little to do with critical objections to the play, and indeed did nothing to  125 quell the controversy, actively pursued by Piron in particular, around the play's literary merit: "Semiramis a ête remise au theatre avec cinquante corrections qui ne corrigent rien." (P. Clement, I: 60) Critics complained that the public seemed immune to the implausibilities in plot which so offended them: Assur and Semiramis confide their schemes of murder to their servants, Azema is a useless character, the ending is predictable. Nonetheless, the attendance remained high (le Mercure, September 1748, p. 228), and the play was an even greater success in Lekain's revivals, especially after the cleared stage permitted more spectacle and a more dramatic entrance for the ghost.  Rome sauvee, which Voltaire considered difficult to perform, and more suitable for "messieurs de l'universite" than the Comedie francaise audience (to Mme Denis, 12 August 1749, D 3975), was subjected to considerable revision both in anticipation of and reaction to performance. Its earliest performances were at Sceaux and at the Prussian court (see correspondence of September 1750). The reworking largely centered on the character of Aurelie, Catilina's wife (see correspondence of September to December 1750), whose entrance before the assembled Senate in act IV was made more of a coup de theatre (13 July 1751, D 4518) Indeed, as the only female character, the role of Aurelie had to be given more plot importance in order to please French audiences. Voltaire's original conception of her  126 character, gentle and obedient like Zaire, gave way before the d'Argentals' criticism (13 November 1751, D 4604) and the casting of Mlle Clairon; Aurelie eventually became more "romaine" (7 October 1752, D 5037). The process of revision went on for months, Voltaire despairing of the "role ingrat et hasarde d'Aurelie." (8 January 1752, D 4760) The play, despite praise from critics such as Clement, was neither long-running nor frequently revived.  Oreste, like Semiramis, was revised during performance but not drastically enough for the critics. At Mlle Clairon's request, her scene of exposition with Iphise in act II was shortened, without reducing Electre's lines (12 January 1750, D 4095). We note that the final text of this scene is even shorter than that in the 1751 printed text, indicating further editing. In addition, Voltaire reduced the amount of text in those difficult scenes in which Electre calls down the gods' vengeance and in which Oreste is pursued by the Furies, where gesture seems more effective than words. But in Clement's opinion, little was accomplished: Vous me demandez la difference de l'Oreste de la seconde representation a celui de la premiere: quelques longueurs de moms et un récit de plus [in act V announcing Oreste's victory] (P. Clement, II: 274) The staging of act V, in which Oreste must kill both Egisthe and Clytemnestre, was changed at the third performance.  127 Instead of stabbing both with one blow, Oreste stabbed each separately, an alteration which did not eliminate the perceived awkwardness for the actor (P. Clement, II: 275). There were, however, considerable revisions made for the 1761 revival of Oreste. Act II, scene i, for example, shifts to a more even division of dialogue between Oreste and Pylade; later in the same act Egisthe's attack on religion is abridged (ed. Moland, p. 158). The process of dialogue between audience and author was especially prolonged for Olympi,e, beginning with the flamboyantly successful performance at Voltaire's home in March 1762, in accodance with his maxim: "Pour le connaitre, il a fallu le faire jouer." (to Collini, 23 April 1762, D 10424) The test run was necessary to prove that the staging, especially the pyre, would work, and that the plot could not be altered in essence. It also convinced Voltaire of the play's worth: "le spectacle le plus singulier et le plus grand tableau qui'on ait jamais vu au theAtre." (to the count and countess d'Argental, 26 April 1762, D 10429) Voltaire admittedly realized that the character of Olympie could not rely for its emotional impact and plausibility solely on the text; the words had to be coloured by the performer: Ce qu'il y a de plus necessaire, c'est que l'actrice chargee du role d'Olympie soit tres-attendrissante, qu'elle soupire, qu'elle longues pauses,  128 de longs silences. (to Collini, 4 September 1762, D 10688) The actress needed to be particularly skilful in speaking Olympie's final line: "Apprends...que je t' que je m'en punis," in act V (  bid). Further, in terms of  structure, the division between acts III and IV, involving the placement of Statira's death and of the duel between Cassandre and Antigone, was the focus of persistent uncertainty [24]. Voltaire encountered such difficulties with the Comedie frangaise's refusal to be directed by him in the staging of  Olympie that he contemplated printing the play rather than losing control of it to the actors (to the count and countess d'Argental, 25 April 1763, D 11174). But the desire to be performed triumphed and the play appeared, although the actors did ask him to rewrite act V (5 March 1764, D 11749).^The 1774 edition largely followed performance edits and revisions, and indicates that the roles most affected were those of Cassandre and Antigone. Cassandre's remorse, and the exposition of his involvement in the rebellion and the death of Alexander, are expressed in the minimum number of lines (I, iii, II ii and iv; III i, passim in act IV esp v-viii; V, v-vii). One speech by Cassandre may illustrate the type of reductions made: [Sostene] De la religion la fureur animee....  129 Vous fait un crime affreux, un crime a detester, De posseder la fille, ayant tud la mere.  [Cassandre] *Les reproches sanglants qu'Ephese peut me faire, *Vous le savez, grand Dieu! n'approchent pas des miens. *J'ai calme, grace au ciel, les coeurs des citoyens; *La mienne sera toujours victime des furies, *Victime de l'amour et de mes barbaries. *Helas! j'avais voulu qu'elle tint tout de moi, *Qu'elle ignorat un sort qui me glagait d'effroi. *De son Fibre en ses mains je mettais l'heritage *Conquis par Antipatre, aujourd'hui mon partage. *Heureux par mon amour, heureux par mes bienfaits, *Une lois en ma vie avec moi-méme en paix; *Tout etait repare, je lui rendais justice. *D'aucun crime, apres tout, mon coeur ne hut complice; J'ai tue Statira, mais c'est dans les combats, Clest en sauvant mon pere, en lui prétant mon bras.... [act III, scene i in Moland VI: 125-26; the asterisks indicate lines omitted in performance] Although the edits are scattered throughout the play, less than 100 lines are involved; what is lost is largely further exposition. From these elisions we may deduce that the actors, unimpressed by Voltaire's reputation as a writer, were concerned chiefly with sustaining the action and the spectacle. Despite his more than forty years of writing  130 successes, he was not, in their eyes, expert in the business of production. Voltaire's only recourse against them was to retire the text from performance and put it into the domain of literature by printing it, as he had done with Rome  sauvee, and would do with les Guebres and les Scythes. Voltaire's use of audience response as a source for textual revision was not acclaimed as sound playwriting technique during the eighteenth century, however sensible it may seem to twentieth-century readers. Voltaire himself, while incorporating elements from performance and performers, retained the status of sole author. The impact of individual performers was considered to be on theatre itself, on the choice of repertoire, and not on the text strictly speaking. The actors' contribution would seem to be lost when the play passed out of active repertoire. Yet we can detect it with careful reading. Voltaire's persistence in playwriting despite the failures of Artemkre and the first Mariamne may owe something to the encouragement of Mlle Lecouvreur, whose opinions and abilities Voltaire respected so highly (Desnoiresterres, I: 183-84). For certain plays the merits of performance, conceived as separate, were viewed as compensating for the faults of text, ie. Zulime and  Semiramis. Even Merope, one of his greatest successes, could be criticized so: "Les representations de Merope...ont fait beaucoup d'honneur  a  M de Voltaire, et la  131 lecture en fait encore plus a Mlle Dumesnil." (attributed to Fontenelle, Moland IV: 175) Voltaire conceded the merit of this type of argumentation: Je doute qu'elle reussisse a la lecture autant qu'a la representation; ce n'est point moi qui ai fait la piece, c'est Mlle Dumesnil...qui fait pleurer le parterre pendant 3 actes de suite. Le public a...mis sur mon compte une partie du plaisir extreme que lui ont fait les acteurs. (4 April 1743, D 2744) A performer was given credit either for transcending a bad text or for what may be called technique. As the  Couespondance litteraire concluded in reviewing Olympie: Tout le monde l'a jugee assez mauvaise A la lecture; mais elle vient de paraitre avec beaucoup de succtIs sur la sCene....Ce succés, auquel le respect qu'on dolt a un grand home et le faste du spectacle paraissent avoir la principale part, ne rend pas cette piëce meilleure aux yeux des gens de gout. [V: 479] Mlle Clairon , for example, was instrumental in the revival of &aim in 1762, which the Mercure describes as "tellement diffórente de celle qui avoit ete representee [en 1740] ...pas moms different d'une Tragedie imprimee sous le titre de Zulime ii y a quelques mois." (January 1762, p. 180) Although the play was not reported in the Mercure on its debut in 1740, in 1762 it was reviewed at length (pp. 181 - 202). Its success was seen as that of Mlle Clairon in  132 the title role: [On remarqua] la verite d'expression que donne l'Actrice a ce dernier vers [et] le pathetique de cette scene [that of Zulime and Ramire on religion] [qui ótait] toujours interrompue par des applaudissemens .... Le role de Zulime est sans doute si dominant, qu'il eclipse presque tous les autres. (p. 203) Mlle Clairon was, the Mercure notes, litres applaudie aux representations comme elle móritera toujours de l'etre," and acclaimed as unsurpassable as a performer (p. 205). But this success did nothing to raise critical opinion of the play, often seen as one of Voltaire's weakest. Bachaumont, for example, considered Zulime an "enfant indigne de sa plume" (I: 12); Mlle Clairon had taken full advantage of "la magie de son jeu [pour] faire disparoitre [sic] les defauts de son role." (I: 13) In the same way Lekain, whose name is as inextricably linked with Voltaire's playwriting as Silvie's was with Marivaux, won plaudits for his acting, but not for his efforts in bringing plays into the repertoire and reviving them. Lekain's first role in Paris was as Titus in Brutus at the Rue Traversiere, his first Comódie frangaise success was as Selde, his last performance was as VendOme (La Harpe, IV: 449). He was a student of Voltaire's theatrical "school", and his career was bound up with that of Voltaire's plays: "M de Voltaire dira sans doute, OU est  133 VendOme, oU est Tancrede, et le public dira, Oil est Lekain?" (La Harpe, IV: 458). His revival of Adelaide in 1765 staged without Voltaire's prior knowledge, much less involvement, was so identified with the actor that the play became known as VendOme, his role. He died "a la suite d'une representation de VendOme, la derniere oU ii ait paru et dans laquelle ii sembla se surpasser lui-meme." (La Harpe, XI: 13-14) And yet his superiority was acknowledged almost exclusively in the realm of visual performance: On n'oubliera jamais le jeu terrible & anime du sieur le Kain, chargé du role d'Arsace...sortant du tombeau de Ninus, le bras nud [sic] & ensanglante, les cheveux dpars, au bruit du Tonnerre, a la lueur des éclairs; arretd par la terreur a la porte; luttant, pour ainsi dire, contre la foudre. Ce tableau, qui dure quelques minutes, & qui est de l'invention de l'Acteur [emphasis mine], fait toujours le plus grand effet. (J-B Clement, p. 163) One could argue even that to eighteenth-century theories of performance at least, an actor did not interpret in an interactive sense, make a personal meaning out of the text. The actor merely spoke text, that "art de la bonne declamation" perfected by Lekain (La Harpe X: 507), and by Mlle Clairon, noted for her "diction enchanteresse." (Desnoiresterres, V: 393) Literary critics gave these two  134 credit for how they spoke, how they moved, even what they wore on stage: "[Lekain] est le premier qui ait eu de vóritables habits de costume .... C'est a lui et a mile Clairon qu'on est redevable du costume sur le thOatre frangais." (La Harpe, IV: 453) Acting, in short, remained in the judgment of literary critics something which is done to text, or around it, capable of varying greatly from actor to actor or from performance to performance, visible and audible to the spectator, but of little consequence to the later reader. We have discussed in some detail the impact performance and performers had on the structure and development of Voltaire's play texts, and on popular and critical reaction to the texts. We have seen the effect of casting,  biensdances and changing audience expectations on the reception afforded to plays in the examples of : Mariamne,  Rome sauvde, Zulime, Mapomet, Adelaide, 04mpie. In short, there seems considerable evidence for the view that performance was an essential and conscious element of Voltaire's dramatic praxis. Yet we must also concede that there was a profound ambivalence on Voltaire's part with regard to performance. On the one hand, he wrote for the stage throughout his entire career, from Oedipe to Irene , staged just before his death, and Agathocle, left unfinished. He expended enormous effort to see his plays performed and to help them  135  succeed. He had an enviable record of stage successes. Indeed his literary reputation during his lifetime was largely as an epic poet and tragic playwright. On the other hand, Voltaire deeply distrusted the judgment of audiences and actors as to the value of a play, and seemed to consider theatrical success transitory and arbitrary. Crowds might flock to La Motte's Ines de Castro, but Racine remained the master. Theatre, at least in the early part of Voltaire's career, was often a means to an end: a way for a young writer to make a name for himself, to make contact with influential people, to have some impact on popular opinion. As a playwright, he was receptive to external response, but unwilling to relinquish control whether during rehearsals, during performance runs, or after the run when the play might be printed. We see in Voltaire strenuous resistance to actors moulding the text even as he asked their advice on revising it. Actors were considered unreliable because they, from Brizard and Dufresne to Lekain and Clairon, were concerned only with their own roles, with the potential for personal success. They were judged to lack both the education and the inclination to appreciate tragedy as a noble form of literature. Their contributions to the text itself, as distinct from those to public reception of that text, is severely controlled, even as we found to be the case in our  136 examination of changes at the linguistic level. Access to education in eighteenth-century France, was a function of class as well as of gender. The marginalized social status of actors may lie at the root of the lack of credit given to them even by the contemporary critics, as we saw in Diderot's comments on Lekain and Mlle Clairon, where he praised their pantomime, their diction and their costumes, and in Bachaumont's inability to believe that Lekain alone could have been responsible for the 1765 Addlai4e. Actresses, despite their key importance to the development of the performance text and to its popular dissemination, appear doubly barred from having their influence credited in the printed text. It might even be argued that Voltaire's apparent reliance on the advice of actresses such as LeCouvreur and Quinault in the early part of his career was in order to elicit not their reactions to the text as literature but rather their practical help in persuading the  comddiens to accept his plays and his frequent corrections. His later relationship with Mlle Clairon makes clear that he preferred to give advice, and was singularly reluctant to trust her judgment about changes to the text, despite her many theatrical successes. The extent to which Voltaire insisted on the preeminent authority of the playwright, and strove to control all aspects of performance, is to some critics a key distinction between his still classical dramaturgy and the more  137 forward-looking theories of Diderot: Voltaire ne saurait ni abdiquer la preeminence de l'ecrivain, ni laisser, comme le souhaite Diderot, quelque liberte a l'acteur, ni alterer la nature d'un genre oil rbgnent la parole et le vers. Ii croit toujours a la tragedie; Diderot n'y voit plus qu'un genre moribond. [251 However, we must point out that Voltaire's desire to control not only text but also performance was itself an innovation, and had considerable influence on the staging of plays, in the realms of costume (from the time of Zaire and  Alzire), decor and stage effects, and particularly on acting style. The extent to which many of his successes, especially the later ones, were identified with individual performers (many of whom he had coached), may explain why his plays were scarcely to be found in the nineteenthcentury repertoire, although the twentieth-century reader might envy Louis Moland's privilege of having seen Sarah Bernhardt portray Zaire in the 18705. We conclude that under each text we have dealt with, static and linear as they may now appear, lies this dynamic struggle between the author's single voice and the many voices of actors, audiences, and critics, between a text open to reaction, revision and interpretation on stage and one closed and controlled by the author's supervision of printing, a struggle both within the text and between the  138 text and the genre in seeking the delicate balance between creativity and obedience to the rules. This struggle was integral to the process by which the texts have arrived at the form they now hold. The extent to which they have been read as expressing Voltaire's personal beliefs and history is an acknowledgment of his skill in incorporating these external influences and in reasserting control over the texts. It is a basic hypothesis of this study that all theatre is political in nature, expressing some relation, whether of opposition or support, to the society in which it appears. The multiplicity of voices, of influences and responses, which we have brought to light may lead us to expect that the ideology which underlies these plays will be complex in nature and heterogenous in source.  139  NOTES  [1]  See  Patrice Pavis, Languages of the Stage - Essays  in the Semiology of Theatre (New York: Performing Arts Journal, 1982), p. 70. [2] Keir Elam, The Semiotics of Theatre and Drama (London & New York: Methuen, 1980), p. 34. [3] Pavis, P. 138. [4] Elam, pp. 42-43. [5] Elam, p. 49; see also Ubersfeld, lire le theatre (Paris: Editions sociales, 1978), pp. 30-31. [6] Elam, pp. 62-63; see also Jean Duvignaud, LA  Sociologie du theatre (Paris: PUF, 1965), pp. 69 - 83 and Ubersfeld, p. 50. [7] Voltaire, Correspondence, ed. Theodore H. Besterman (Oxford: Voltaire Foundation, 1968), letter to Francois Tronchin, 16 December 1758, D 7983 and letter to the count d'Argental, 3 October 1761, D 10052. [8] Elam, pp. 208 - 209. [9] Pierre Clement, Cinq anndes litteraires (Geneve: Slatkine, 1967; reproduction of the 1755 Berlin edition), I: 277.  [10] Le Mercure de France, mars 1720 P. 101. [11] Gustave Desnoiresterres, Voltaire et la sociéte au  XVIIIe siècle 2e ed (Paris: Didier, 1871), I: 183 - 87. [12] Jean Marie Bernard Clement and Joseph de la Porte,  Anecdotes dramatiques (Geneve: Slatkine, 1971;  140 reproduction of Paris 1775), p. 522. [13] Helene Monod-Cassidy, Un Voyageur-philosophe au  XVIIIe siècle: l'abb4 Jean-Bernard Le Blanc (Cambridge: Harvard UP, 1941), p. 195. All citations from Le Blanc will be from this work, which contains his correspondence.  [14] Armee littéraire, ed. Elie Catherine Fróron (1756) II: 348. [15] Jean Francois de la Harpe, Oeuvres (Geneve: Slatkine, 1968; reproduction of the Paris 1820 edition), X: 390.  [16] Correspondanc- littdraire, philosophique et critique, ed. Maurice Tourneux (Paris: Gamier freres, 1877), I: 266. [17] In lie Mercure, January 1762. [18] See also the letter to the Count and Countess d'Argental, 17 September 1765, D 12887; Le Mercure was kinder to the original production. [19] Louis Petit de Bachaumont, MémoireS secrets pour  servir A l'histoire de la rêpublique des 1.ettres en France depuis MDCCLXII jusqu'a nos jours (London: John Adamson, 1777), III: 177. [20] See  the statistics furnished by H. C. Lancaster in  "La Comedie francaise 1701 - 1774: Plays, Actors Spectators, Finances," Transactions of the American  Philosopical Society 41: 4 (1951), pp. 593 - 894. [21] René Vaillot, Avec Mme du Chatelet, Voltaire en  141 Son temps 2, ed. Rend Pomeau (Oxford UP, 1988), p. 336.  [22] Journal et mdmoires de Charles Colld, ed. Honoré Bonhomme (Paris: Firmin Didot, 1868), I: 127. [23] Robert Niklaus, "Briphile de Voltaire et le thdatre d'Eschyle" in Le Siècle de Voltaire: Homage A Rend  Pommeau, eds Christiane Mervaud and Sylvain Menant (Oxford: Voltaire Foundation, 1987), II: 709. [24] See the correspondence of October 1762. [25] Henri Lagrave, "Voltaire, Diderot et le 'tableau scdnique': le malentendu de Tancrëde," in Le Siècle de  Voltaire: Hommage A Rend Pommeau, II: 574.  142 Chapter III - Ideology in the Tragic World: Systems of Belief and Codes of Behaviour in the Political, Religious and Familial Domains [1]  Ideology is herein defined as a system of beliefs which determines the ideal structure of a society and which provides the norms by which citizens of that society are to behave. Of course, much of the published analysis of the ideological content of Voltaire's tragedies has concentrated on the explicit statements made by characters, and has often arrived at the conclusion that certain plays have little or no ideological content [2]. But, as set out in the introduction (pages 15-16), this study takes as one of its fundamental premises that all writing for the theatre is ideological and political in that it exists in a relationship of reinforcement or reform to the society which produced it, so that even texts which apparently have no ideological content do take an ideological position. Among the critics who adopt this view, Anne Ubersfeld posits that all theatre expresses some kind of ideology, and is in a relationship of agreement or opposition to the dominant ideology of the society in which the theatre was produced [3].  Nor is this critical position restricted to contemporary analysis of modern theatre. Classical tragedy in France had an overtly political overtone, particularly  143  under Richelieu who sought to make of tragedy a "vehicle for the promotion of authoritarian views and a static society." [4] This is not to say that State control over tragedy through patronage and censorship amounted to dictation, but rather that the general understanding of the genre included its responsibility to consider political issues such as the nature of kingship: La mainmise progressive de 1/Etat n'implique cependant pas que le thdAtre soit un simple instrument de propagation de l'iddologie monarchique....Les auteurs dramatiques...parachevent leur tAche en posant, sur la scene du thdAtre, des problemes nouveaux qui ne peuvent pas étre dits ni resolus, ni peut-Otre même pensds, A l'interieur du systeme du droit....On n'y met pas seulement en scene des monarques hêrolques, mais l'ensemble des images associ6es A la figure du roi, depuis le prince gendreux jusqu'au tyran. [L'auteur] renvoie [au roi] par l'intermódiaire de la scene, une image de ce quill dolt ou ne dolt pas étre. [5] Thus an ideological parti pris in Voltaire's theatre can be seen not as an innovation springing from his overriding philosophical beliefs but rather as part of his adherence to classical norms. In considering the ideological content of these plays, and in particular in seeking to establish a system or systems of belief which appear to operate within and even  144 between texts, we will also address the question of whether variation in textual expression necessarily includes variation in ideological content. We will seek to determine whether the underlying beliefs are as fluid and as open to change as we have demonstrated the texts to be, subject to perpetual transformation at the levels of linguistic and visual expression through revision and performance. We may find constants within these texts at the ideological level, similarities in the nature of kingship, the role of the priesthood and the structure of the family which persist over time and from text to text. The beliefs themselves may prove to be the product of dialogue and collaboration with friends, actors and audiences, or on the other hand relatively personal and static. Certain beliefs appear to be constant, at least. Jory's conclusion, that Oeaipe contains "the image of the ideal king which remained consistent throughout Voltaire's works and life," (p. 12) can, as we develop this analysis, find abundant textual support, from Eriphile to les Guebres. Analysis of the ideological content of theatrical text must include but not be confined to explicit statements. These statements are frequently undercut and even denied by the context in which they are spoken (see pages 160, 179). The "situation de parole" which is inseparable from the "Ononcd" itself includes the conditions of enunciation, that is, those relationships of power between the characters  145 which determine not only the content of speech but even the ability to speak [6]. These relationships of power, controlling both speech and action, will provide evidence for conclusions on the nature of government and family. The dominant ideology, the "formation socio-historique donnee" in which the text was conceived acts not only directly through speech but also indirectly on speech, constraining what authors could write [7]. The extent to which the ideology of the texts supports or challenges the general beliefs of educated eighteenth-century Parisians, their first if not their only intended destinataires, becomes key to our analysis. We note that expressions of an ideology opposed to the status quo are often placed in the mouths of minor characters or villainous ones, thus lessening their modelling force because they are spoken by the less important and less attractive agents in the drama. They can also by this choice of locuteur become associated with punishment and failure [8]. Lancaster goes so far to argue that this was a deliberate technique to allay the censors, who presumably would accept the unlikelihood of the public adopting such an obviously unsuccessful stance and thus let the statement pass (I: 216). There are some cases where a notable speech is contradicted by the action of the play as a whole, as when in Eriphile Alcmdon complains that birth rather than merit is unjust as a measure of fitness to  146 command (in II, i) The audience knows, or at least suspects at that point, that his birth is royal. Thus his ability to lead is explained within the extant social system rather than proving its injustice. Zaire's famous speech on religion as an environmental or cultural phenomenon is thrown into question by her behaviour, in that she abandons immediately the religion in which she was raised to embrace the Christianity from which she was separated at birth and of which she knows almost nothing [9]. Thus context must be included in any analysis of ideological substance or intent. The discussion of ideological systems in this chapter is organized around three general areas of social structure: A.  Government, including monarchies (male and female rulers and the role of the court), republics and the role of the people;  B.  Religion, including the priesthood, the status of state religion, conversion between religions, and the depiction of individual priests and believers;  C.^Family, including the divisions of power and responsibility between parent and child as well as husband and wife, and the treatment given to gender and sexuality within the family structure: We will continue to consider the role of public response in all its forms now as it affects the ideological content or impact of a play. The ideology of a theatrical  147 text may prove to be neither constant over time nor consistent with itself, even as the text itself is dynamic. We will examine how the textual transformations could emphasize, alter, lessen or even contradict the types of societies and individual behaviours presented. But we will also consider whether the underlying ideological system, or at least key concepts and structures of that system, remain relatively constant within a text and over time. There is evidence for our hypothesis that Voltaire successfully sought to build a plausible social microcosm in his tragic plays. Contemporary critics gave him credit for accurate portrayals of historically based characters and societies. Cideville, describing the public reaction to the 1725 Mariamne, tells Voltaire how the audience was struck by the "portrait ressemblant des femmes et des fureurs touchantes d'Herode" (13 April 1725, D 229). The  CorresponOence litteraire concludes, late in Voltaire's career, that "M de Voltaire est un grand peintre de moeurs, et voila le grand merite de ses tragedies, c'est d'en presenter toujours un tableau fidéle." (October 1760, IV: 299)  Government and the Social Hierarchy The predominant type of government within the societies depicted in these plays is the patrilineal monarchy. The characteristics of the good or bad king are set forth at length in many of the plays. While the reader may now find  148 an ironic undertone in Oedipe's proclamation, "Mourir pour son pays, c'est le devoir d'un roi" (II, iv: II: 77), there is no question that the audience is meant to take the statement seriously. Philoctete's objectivity, conveyed in the lines: Un roi pour ses sujets est un dieu qu'on róvére Pour Hercule et pour moi, c'est un home ordinaire must be placed in its context of association with the semi-divine hero Hercules, outside the bounds of normal society (II, iv; II: 78). The ideal of a king sacrificing himself is taken up in Eriphile, where Alcmeon proposes single combat in order to halt the civil war (V, i). In a late play, les Guebres, divine status is given to a just ruler without any ironic implication: "Je crois entendre un dieu...Qui pane au genre humain pour le rendre plus juste." (V, vi; VI: 567) The characteristics of Voltaire's good king are relatively constant, as Jory has indicated (p. 12), and the texts do not seriously question the monarchy as a system which transfers the right to rule by male inheritance. Sareil finds this acceptance a merit in Voltaire's social criticism: C'est en acceptant toutes les regles du monde dans lequel ii vit, qu'il s'attaque aux institutions et aux abus. Jamais il ne remet vraiment en cause la legitimitd de l'ordre social, et c'est pour cela que sa  149 critique est Si gaie, et aussi qu'elle a si bonne conscience, [10] Showalter Jr. finds this complaisance throughout eighteenthcentury theatre: The temporary errors of authority figures seldom prod the heroes to revolt or the narrators to question the political structure. Female characters were held to an even more groveling acquiescence to the unreasonable demands of their masters. [11] While the Voltairean text reiterates that a king must have worth - learning, justice, tolerance, it remains within the mainstream of contemporary theatrical practice by retaining the hierarchy based on birthright. Within these texts, the good or rightful king must recognize the primacy of learning and written law above military might, as Genghis Khan does at the close of l'Orphelin de Chine: "Je fus un conqudrant, vous m'avez fait un roi." (V, vi; V: 356) Several of the unjust kings, or aspiring usurpers, are nonetheless efficient soldiers: Hdrode, Hermogide, Egisthe, Polifonte. As well, the characteristic of mercy seems key: Orosmane's last act, as he lies dying, is to order Nerestan's safe conduct (Zaire V, x). In le Triumvirat, Octave's pardoning Pompde is the first sign of his transformation into the relatively just ruler history paints as Augustus Caesar. Another central aspect of this portrayal is the  150  independence of the good king, controlled neither by his courtiers nor by the priests. The importance of independence is further illustrated within the plays chosen for study by the bad examples of Cassandre and Herode, whose malleability leads the first to be murdered, the second to kill himself. Herode himself recognizes the faults in his reign, in that he has to this point been content to "regner avec éclat, mais avec barbarie"; he intends "sur mes sujets rógner en citoyen." (III, iv; II: 196) While there remains considerable material on the nature of just rule in Herode  et Mariamne, we should indicate that much of the historical context of the play, the debate over Herode's ambivalent status as a subject king under Roman authority, was greatly reduced in the 1725 revision and eliminated from the  1762  version. Hdrode is made more of a type character, demonstrating as Cassandre does the evils of government under a king who acts from emotion rather than reason, instead of the particular politico-historical problem of the kingdom as an imperial province. These pledges to rule as a citizen should be seen as part of the attack on the power of the court, which acts as a barrier to free dialogue between the king and the people. The good ruler is not controlled by his priests any more than by his court. It is typical of the just and compassionate priest that he refuses to become involved in politics at all. In les Guatres it is stated that  151  le trOne s'humilie Jusqu'a laisser regner ce ministere impie. (IV, vi; VI: 555) The emperor arrives at the pragmatic conclusion: Les persecutions Ont mal servi ma gloire, et font trop de rebelles (V, vi; VI: 566) The phrase "en citoyen," used by Herode to describe a just reign, implies a direct relationship between the ruler and his people. It recurs in a much later play, les Gubbres: "Je pense en citoyen, j'agis en empereur." (V, vi; VI: 567) Good kings are accessible to their subjects, not only to arbitrate disputes but also to hear their concerns and take their advice. (Zaire, ed. Jacobs; I, iii, 235-38 and les Gu6bres IV, i; VI: 544) The emperor in lep GuëbreS is portrayed as preferring to take advice from apparently common people, although audiences then and now recognized in the gardener Arzemon the far from humble Voltaire (les Guebres, ibid). The central image for this relationship between the ruler and his people is that of a good father, whether it is used of Alvarez in Alzire [12] or of Zopire in Mahomet, as in Haydn Mason's analysis [13]. Mason's focus is on the implications of the paternal image in a critique of religion: It is conceived within the dominant structure of  152 paternity, true and false...the father-son theme that could still, even in the early 1740's, serve as the essential channel through which to ecraser l'infame. (1988, p. 133) There is, however, textual evidence allowing us to extend the metaphor, both within this text and to others. In  Mahomet, Zopire rules Medina with a "zele paternel" (I, i; IV: 107) Alcmdon, in coming to the throne still in his teens, gives himself this paternal status: Avec le nom de roi, je prends un coeur de pere. He faudrait-il verser, dans mon regne naissant, Pour un seul ennemi, tant de sang innocent? Est-ce a moi de donner le sacrilege exemple D'attaquer les dieux mem, et de souiller leur temple? [14] This metaphor of paternal authority illustrates the rightful use of power as one centralized in a single benevolent and just male figure. The responsibilities of rulership, like those of citizenship, are modelled on the family system. But these duties to the State take precedence over the duties owed to one's biological family. A father should sacrifice his child for the good of the country (Mahomet II, v; IV: 128). This principle also informs the plots of Brutus and 'Orphelin de la Chine). Even the promised return of his children, missing and believed dead for years, does not  153  weaken Zopire's opposition to Mahomet; he would kill them himself rather than bow to an imposter (II, v; IV: 128). Indeed, 'dame deplores her maternal weakness in trying to save her own son rather than supporting her husband's decision to sacrifice him to save the emperor's son: "Il pensait en heros, je n'agissais qu'en mere." (IV, iv; V: 341) Voltaire's last tragedy, left unfinished by his death, was Agathocle, in which he returned to the theme of the correct models of fatherhood and kingship as interdependent. But there is a hierarchy of responsibility and duty established within and across the texts, with the king as head of state, father of his country, at the apex. The key principle in this hierarchy of authority is the supremacy of the State and its laws over other duties, even those to religion and family. As the emperor in les Guebres summarizes the equitable state: Que chacun dans sa loi cherche en paix la lumiére, Mais la loi de l'Etat est toujours la premiere. (V, vi; VI: 567) The crime of lese majesty, however motivated, is uniformly punished. VendOme, in Adelaide, has no rightful authority over Adelaide and Nemours because he is a rebel against his king (II, v). To be motivated by love rather than honour and loyalty is a grave fault (III, iii). This overriding duty to the ruler forms the basis of the plot of Olympie. Although Cassandre at the time was obeying and  154 protecting his father, his participation in the rebellion against Alexandre requires that he be punished, his marriage to Alexandre's daughter prevented, hence the deaths of Olympie and Cassandre. Crimes against the rightful ruler are unforgiveable, and are spoken of in terms suggesting religious blasphemy rather than politically motivated violence. In Oreste, Egisthe is introduced as a "tyran" (I, i) and referred to as such throughout; he is constantly opposed to Agamemnon, who is spoken of in the Biblical phrase "roi des rois."  ^Egisthe and Clytemnestre in  striking Agamemnon raised "leur main sacrilege" (I, iv; V: 96) against (if we may continue the Biblical image) the Lord's anointed. This hierarchical arrangement of duties does not entail, however, that lesser duties could be shirked entirely, not even by kings. The corrupt or illegitimate ruler is as clearly marked by a failure in his duties as father as by his injustice to his subjects. Mahomet is called a "father" to the orphans he has recruited as soldiers and concubines, that is to say to his dupes and victims (II, iv; IV: 122 and III, v; IV: 135). Zopire calls him a "tyran" (in I, iv); this particular epithet will become a favourite pejorative used of those holding power illegitimately throughout the plays, and a key indication that their hold on power is precarious. Catilina is marked as a potential tyrant by his deliberate dereliction of  155 familial responsibilities : Titres chers et sacres, et de p6re, et d'epoux, Faiblesses des humains, evanouissez-vous. (Rome sauvee 1, i; V: 214) The leaders of the Triumvirat are rarely if ever described as other than "tyrans" throughout the play. Both Antoine, in abandoning his wife Fulvie, and Octave, in choosing to execute Julie's father and lover in order to marry her, reveal their lack of commitment to the whole system of values and responsibilities held up as necessary for achieving and maintaining legitimate rule. Thus, as analysis of the text shows, a relative constancy in the portrayal of model kings, both favourably and unfavourably. This theme, the nature of kingship is taken up frequently throughout the oeuvre under examination, but the idealized emperor in les Guébres bears considerable resemblance to the ideal king in Qedipe. Those textual transformations which appear to have political implications entail an increased emphasis on French nationalism, clearly seen in the variants to Adelaide and Zaire dating from the 1760's: the shift from "francs" to "frangais" as the favoured epithet for the Crusader knights in Zaire, the restaging of Adelaide with historically accurate personal names and setting. But these transformations, and particularly the renewed Adelaide which was initially staged without Voltaire's knowledge, can easily be attributed to  156 theatrical pragmatism rather than to any shift in the ideological system. The considerable popular success of Du Belloy's patriotic Siege de Calais in the early 1760s would naturally lead to imitations. This relatively constant authority system is based on the symbol of the father, on male authority at all levels of society. Realms in which women have managed to wield power inevitably fall into chaos, whether Assyrie under Sdmiramis, Argos under Eriphile, or Trdbizonde under Zulime. Zulime, in fact, refuses her father's throne to flee with Ramire (II, iv), showing that for a woman affection takes precedence over responsibility. Despite her original ability and courage, Sdmiramis has lost the respect of her subjects, who "de servir une femme en secret sont lassds." (II, iii; IV: 525) As a woman, her appearance was key to her success as a ruler: Sa beautó, ce flatteur avantage, Fit adorer les lois qu'imposa son courage. (II, iv; IV: 527) Although an active partner in the murder of Agamemnon, Clytemnestre likewise sinks into subordination afterwards. Egisthe becomes "un maitre absolu", and she no more than "sa premiere sujette." (Oreste I, v) Against historical example, these plays operate on the premise that females are not fit to govern, and are not accepted by their subjects, for example in Eriphile : "L'Etat demande un maitre" (I,  157 iv; II: 467). A woman has no access to legitimate authority to rule. Mdrope, as a good mother, refuses the throne even as regente for her son, despite the persistent danger from other cities and the threat of imminent civil war. Those female characters who have attained power are shown as unable to maintain it; the focus is on them as guilty mothers rather than as political figures. Eriphile, who was at sixteen no more than a passive accomplice to her husband's murder, and has since succeeded in keeping his murderer from the throne, is from the beginning of the play obsessed by remorse. Semiramis, the ruthless and capable Assyrian queen, is painted as devoured by remorse and no longer able to govern: "Succombant au mal qui la dechire, See mains laissent flotter les renes de l'empire." (I, i) Fulvie, in LA  Triumvirat, continues this association of sexual activity, crime and ineffectiveness established by Eriphile and Clytemnestre. She was originally a co-conspirator with Octave and Antoine, and as such not only a political agitator but independent and vengeful enough to try to kill Antoine on her own (V, ii). But she is ultimately not only unsuccessful but contemptible, because she has stepped outside the bounds of proper female behaviour, as exemplified by the virtuous and passive Julie (i.e. in IV, vii). Fulvie is reproved for her unwomanly activity:  158 Dans nos vaines douleurs D'un sexe infortune les armes sont les pleurs. (IV, i; VI: 222) The restoration of order in these chaotic realms, prone to civil war, requires the establishment of a legitimate male ruler, by any means necessary. Alcmeon, who gains the throne by accidentally killing his mother, re-establishes the male line of Amphiarails and therefore legitimizes his rule. Rather than evoking punishment, his actions are considered a source of wisdom: Que de ce jour affreux l'exemple menagant Rends son coeur plus juste, et son régne plus grand. (V, vii; II: 504) We will return to the differing retribution meted out for patricidal and matricidal actions, for the crimes of sons and those of daughters (see pages 201-202). As perhaps could be expected in a body of work meant to appeal to Paris rather than Fontainebleau or Versailles, the court surrounding the king is consistently criticized from the earliest plays. Pallante, the villain of Artemire, has succeeded in establishing such influence over Cassandre that he has become "id i le seul roi" (I, i; II: 128). He uses Cassandre's trust in plotting to murder and succeed him, bypassing succession by inheritance. Many of Cassandre's faults as a king and husband are attributed by Artemire to Pallante's influence:  159 C'est toi qui, de ton prince infAme corrupteur, Au crime, des l'enfance, as prepare son coeur. (II, Hi; II: 137) The rivalry between Salome and Mariamne is, like that between Pallante and Artemire, painted in political terms: the struggle for control over Herode. His court, like all the courts in the tragedies, is full of plots, secrets, and enemies (see II, i ^ii). The character of the courtier, from Pallante to les Guebres, remains consistently negative: unfaithful, ambitious and unscrupulous. Alcmeon's condemnation in Ertphile: Vos oisifs courtisans, que les chagrins devorent, S'efforcent d'obscurer les astres qu'ils adorent (IV, i; II: 488) is echoed by Merope: ... la foule infidele Des memes courtisans que j'ai vus autrefois ramper sous mes lois. (eds. Vrooman and Godden; V, iv, 134-36) The courtiers have been corrupted by Polifonte (I, iv, 294), and support his accession. In Semiramis, we return to the villainous courtier as a plot device; Assur, like Pallante, is seeking to take his "master's" place as king. This courtier type of usurper is less frequently seen within the plays, and appears less efficient than military leaders such as Hermogide, Egisthe, and Polifonte, who win  160 credibility on the battlefield first and use rather than join the court, which appears no more efficient than honest. The court does not actually appear in Les GuOres, but is nonetheless attacked as one reason for the apparent injustice of the emperor's reign: Ces flots de courtisans, ce monde de flatteurs ...qui laisse languir la valeur ignorde; (I, i; VI: 508) and Ce mercenaire usage, et ces homes cruels Gages pour se baigner dans le sang des mortels. (IV, vi; VI: 552) The one exception to this condemnation of court and courtiers which we may note is the character Coucy / Lisois in the various versions of Adelaide, who is able to bring reason to Vendeme's leadership while remaining subordinate to it: Vos conseils prudents [peuvent] Moderer de son coeur les transports turbulents. (ed. Cartwright; I, i, 121-2; 1734 and 1765 versions) It is part of his superiority to know when to disobey orders so that he is not VendOme's accomplice in fratricide (V, iv, 133 in 1765 and V, v, 143, in 1734). Indeed this rational disobedience, this refusal to encourage crime, which distinguishes Coucy is essential to the play's happy  161 denouement. But the general rule of these plays is enunciated in Semiramis:: "un soldat est mauvais courtisan," (I, iv; IV: 514), a conclusion emphasized by the contrast in Les Guebres between Cesene and Iradan on the one hand, loyal officers, and the corrupt court and arrogant priesthood on the other. The right government of a society appears, despite the negative influence of the court, to require some sort of social hierarchy. Only Brutus offers us the picture of a stable republic, and even here that republican ideal is personified by a single individual. The senate form of government will be examined shortly. But the monarchical form appears to be typical of these plays, and that structure entails an aristocracy. There are within the tragedies well-known criticisms of the aristocracy, as in that cited earlier from Oedipe (p. 148) and in the passage from Eriphile cited below. However, on close examination, these passages lose some of their force (see p. 148). For example, Alcmeon's speech in  Eriphile, "Qui sert son pays n'a pas besoin d'aleux," (II, i; II: 471ff) is undercut first by the action of the play, in which Alcmdon is revealed as the heir to the throne, thus explaining his extraordinary ability within the social status quo. The passage is further undermined by its repetition within Merope, where it is spoken by the villain Polifonte as part of his justification for seizing the  162 throne (I, iii). In this context, the statement is denied overtly by Merope within the scene (ibid.), and implicitly by the working out of the plot. Other plays give some weight to the traditional association between birth and fitness to rule. Mahomet's credibility is weakened by his lowly birth: "au dernier rang des derniers citoyens" (I, iv; IV: 113) A reference to Ciceron's plebeian status, as a reason for Catilina to oppose his authority, was first included and then deleted from Rome sauvee (I, 1). In Adelaide, the issue of nobility, and social expectations of how it should be portrayed, formed the major obstacle to public acceptance. Voltaire accepted that it was impossible to depict "un prince de sang", VendOme, as his brother's murderer (27 February 1734, to Cideville, D 712). This respect afforded to the nobility, as distinct from the court, explains the 1752 revision of Adelaide as the much less historical Duc de Foix (to the count d'Argental, 16 oct 1751, D 4595; and 3 June 1752, D 4902) This constraint of social expectation played a role in Voltaire's resistance to the revival of the original Adelaide (to Lekain, 20 February 1763, D 11027; and to the count and countess d'Argental 25 February 1763, D 11042) We must note that Vendeme's repentance, the denouement of all versions of the plot, included his joining Nemours and Adelaide on the French royal side. The true noble does not rebel against his king: "Bon Francois,  163 meilleur fr6re, ami, sujet fidele." (V, v, 207 1765; likewise in the 1734 and 1751 versions) Nobility, like rulership, has its responsibilities, and social stability depends on their maintenance. The Roman senate is not favourably portrayed within these tragedies. In Brutus, the first of Voltaire's Roman plays, the Senate's ingratitude to Brutus' son Titus is made a motivating factor in the plot, in order to explain why he is susceptible to the Tarquins' corruption. Voltaire follows the Shakespearian model in his adaptation of La Mort de Cesar, by making the assassination of Caesar, intended to restore republican rule, into a tragic event. In Rome sauvóe, moving to the plays under direct study, Catilina attacks the Senate as the "tyran de l'Italie" (I, ii; V: 214); indeed this key epithet is applied to the Senate throughout the play (i.e. again at II, vi). The Senate is flawed by its own power: "enivrê de sa grandeur supréme." (I, vii; V: 224) Its ingratitude, in failing to reward its subjects, is again a plot factor, as Voltaire foreshadows Cicdron's being punished for foiling Catalina's conspiracy: Ah! qui sert son pays sert souvent un ingrat. Votre merite méme irrite le senat; Ii volt d'un oeil jaloux cet éclat qui l'offense.  (I, vii; V: 224) Although in history no dictator was appointed at this crisis, Cicêron himself, defender of the republican state,  164 exclaims, "Rome demande un chef en ces calamites." (IV, vi; V: 257) Even he is portrayed as finding the Senate inadequate to protect the state. The Senate remains a weak force in the later Roman play Le Triumvirat, in which the Senate proves helpless to protect the people from the tyrants (I, iv). In Tancrëde, there is a type of senate called an "assemblee de chêvaliers" which is similarly marked by injustice and ingratitude. It banishes Tancrede and makes his goods forfeit on the mere accusation of Orbessan (I, i). Argire admits, "Je le servis injuste, et le cheris ingrat" (I, iv; V: 511) and concedes that "un senat tyrannique est id i tout-puissant." (I, vi; V: 513) Amênaide is so outraged by its condemnation of Tancrbde and herself, resulting in his dying to defend her, that she curses it comprehensively: Que l'enfer engloutisse, et vous, et ma patrie, Et ce sênat barbare, et ces horribles droits D'êgorger l'innocence avec le fer des lois. (V, vi; V: 562) There is also an organization called a senate in Mahomet, but it is subordinate to Zopire's leadership. While characters refer to it (I, iv), it is both invisible and inactive within the play. Although the good king is in direct contact with his people, within the tragedies under study the general population is generally portrayed as a mass of foolish  165 children requiring a wise father. Their fickleness and prejudices are presumed from Oedipe onwards. In this first play by Voltaire, the people of Thebes lose their respect for Laius upon his death (I, iii), which was seen at the time as an allusion to the popular reaction after the death of Louis XIV; they are equally quick to accuse Philoctete without evidence of complicity in Laius's murder (II, i; III, i-ii). In Eriphile the people demonstrate their need for a strong male ruler by throwing their support behind the suspected regicide Polifonte (I, ii and iv). He holds this support at its true worth, calling them "ce peuple infidele" (I, iv, 294). To Merope's despair, "nos peuples volages" acclaim Polifonte the murderer (II, iii, 181); however, they are equally ready to accept Egiste. Cicdron, in Rome sauvee, expects to be in turn acclaimed and rejected by the people: Je connais l'inconstance aux humains ordinaire; J'attends sans m'ebranler les retours du vuIgaire. (V, iii; V: 265) Gullibility and irrational prejudice are likewise characteristic of the masses. Zopire condemns their facile conversion to Mahomet's leading: Ce que ton peuple adore excite mes mepris.... Des plus des humains tente la foi cr6dule. (I, iv; IV: 114) Mahomet himself has no more respect for his followers. He  166  knows and uses their preconceived ideas to control them: "Les prejugeffs, ami, sont les rois du vulgaire," (II, iv; IV: 122), as in his use of superstitious credulity to pass off Seide's murder as divine intervention (V, iv). We would be hard pressed to find any evidence for "democracy" in the treatment given the mass of common people or of more republican forms of government within these tragedies. The ideal states portrayed - stable, just and merciful - are at most constitutional monarchies, where kings are limited by universal laws of reason and equity, where duty to God and family is subordinated to one's duty as a citizen. The prevalence of the father as symbol of just rule, and the interrelationship indicated among the various levels of authority, will be examined in more detail in our discussion of family social structures (beginning at p. 184). B.^Religion in its Institutional And Individual  Manifestations Within the limits of this study we cannot propose to address the complicated and contentious question of Voltaire's personal beliefs about divinity and religion. Our focus is rather religion as it operates in the tragedies, and in particular the social function and values attributed to it. As Pomeau has observed, Voltaire was labelled as anything from essentially Christian to atheist by his contemporaries and by nineteenth-century critics [15]. Current criticism usually places Voltaire as a deist  167 or a humanist: God may exist, as a Creator, but he is so distant and non-interventionist that the responsibility is thrown back on humanity to save or condemn itself. But there is considerable variation in how a particular critic emphasizes the anticlericalism or rationalism implicit in this belief, or argues the extent to which his anticlericalism became anticatholic or even antichristian [16]. The relationship between the stage and the Church in France was problematic at best during the eighteenth century. The Church was active in the censorship and even banning of plays; actors of course were excommunicates. Pomeau mentions Voltaire's tendency to depict "a la scene les sanglantes evocations du pretre assassin", and finds in Voltaire a lifelong struggle to defeat "la double obsession du Dieu terrible et du pretre cruel." [17] But neither Pomeau nor Gay analyzes the frequent portraits of idealized good priests in the tragedies, nor the fact that Voltaire enjoyed playing the wise high priest in Olympie at les Delices. Thus there appears to be some separation between what is commonly held to be Voltaire's personal belief and the belief system of the plays in the treatment of organized religion. In concentrating on the texts, we will examine the way in which the god or gods themselves are described, the nature and influence of the priesthood as an institution, the ambivalent status of conversion as both desirable and  168 condemned, and the portrayal of individual priests and of individual believers including their statements of faith. Attacks on the gods themselves are more typical of the earliest plays, such as Oesiipe, written under the Regency where frank skepticism was socially tolerated and even popular. The gods of Oedipe are unjust; Thebes suffers under their "colere inhumaine." (I, i; II: 61) The divine oracles are declared to be dubious in value and subject to abuse by ambitious priests (III, iv: II: 87). These statements of doubt arguably carry more weight because they are spoken by Major protagonists and are not denied by the development of the plot insofar as the goods do appear cruel and their oracles more harmful than helpful. The audience seems meant to approve such speeches as Jocaste's famous rejection of priestly authority, although the well-known plot requires that the priests be correct: Nos prètres ne sont pas ce qu'un vain peuple pense, Notre cródulite fait toute leur science. (IV, 1; II: 93) The attacks on faith itself, on the existence of god(s), were highly vulnerable to censorship on their appearance in performance or print. They occur less frequently the later the play or revision. The 1750 edition of Oreste contains ten lines by Egisthe on the foolishness of attempting to receive guidance from the gods (II, vi), which is reduced to two lines in the established text (V:  169 111). The first editions of Les Gubbres had an additional four lines in the opening confrontation between Iradan and the high priest (I, iii), part of their debate over the primacy of religious law; these were later edited out, possibly as part of the efforts to see the play performed. However unlikely or unwelcome their prophecies, the divine oracles are never proven wrong; indeed their fulfillment is key to the plots of Oreste and Semiramis as well. The attack on divine power in Sdmiramis is put into the mouth of the villain Assur: "C'est par la fermetd qu'on rend les dieux faciles." (II, vii; IV: 530) This line was censored by Crdbillon and only reinstated on Voltaire's complaint to Nicolas Rend Berryer, the lieutenant gdndral de  police (27 June 1748, D 3679). Any value this statement might have had as a promotion of skepticism, however, appears to be undercut by its speaker, the scheming courtier Assur. Furthermore it is denied directly by the protagonist herself, who responds to Assur's unbelief with a statement of her own contrition: Qu'on peut sans siavilir S'abaisser sous les dieux, les craindre et les servir. (II, vii; IV: 532) Although Merope complains, as did Jocaste, Par l'or de ce tyran, le grand-prbtre inspire A fait parler le dieu dans son temple adore" (eds. Vrooman & Godden; IV, iv, 209-10)  170 the high priest proves not to be corrupt, and the play ends with his benediction: Cette gloire est aux dieux Ainsi que le bonheur, la vertu nous vient d'eux. (V, viii, 295-96) Late plays go even further in their expression of belief in a divine power. Olympie and Les GuObres contain apparently sincere prayers made on stage by major sympathetic characters. Even Mahomet was initially accepted, in less sophisticated Lille, as an attack on the falseness of Islam as opposed to Christianity, rather than as Paris understood it, as an attack on Christianity "a travers l'Islam." [18] We must also note that this play presents Zopire as a sincere believer, holding to the faith in which he was born despite the practical merits of conversion (I, i; IV: 107). Nonetheless, the gods of the tragic world are not universally compassionate. Their punishments are neither consistent nor sure. If they were, there would be no tragic action, no suffering of a victim who is sympathetic without being absolutely innocent. Desvignes indicates that the suffering of a relatively innocent character is a recurrent theme within these plays (pp. 550-51). Comparative virtue is no protection from "divine" wrath, because no one can be perfect. Jocaste attacks with justification "les dieux qui m'ont forcee au crime" (V. vi;  171 II: 112), since her sins were unconscious. Zaire's Christian brother Ndrestan calls her death her just punishment for having loved Orosmane (ed. Jacobs; V, x, 172), which seems and was meant to seem excessive, the persistence of his fanatical hatred of Islam. In Eriphile, Thdandre criticises not the gods' existence but their mercy, exclaiming: Impenêtrables dieux! est-il donc des forfaits Que vos sdvèrites ne pardonnent jamais? (V, iv: II: 501) AlcmOon asks that their vengeance fall on his head for having killed his mother (V, vii; II: 504), a punishment which does not arrive in his case, though his crime seems no less than that of Oreste. It is interesting that the gods and the father Amphiarads are interchangeable as sources of retribution; this same passage attributes Eriphile's death to both forces (ibid). Not only are the gods of this world more severe than loving, they permit that at least on occasion evil should succeed. At the end of Mahomet, the charlatan triumphs; Palmire cries out in despair, "Le monde est fait pour les tyrans," and dies. (V, iv; IV: 461) At the end of Zaire only the unsympathetic Nerestan and the scheming Corasmin survive. Antigone, the only major character left alive, closes Olympie with the unanswerable questions: Dieux...  172 Qu'avait fait Statira? qu'avait fait Olympie? A quoi reservez-vous ma deplorable vie? (V, vii; VI: 164) Such faith as may exist in the tragic sphere is not without its doubts and unresolved conflicts. Such questioning, such wrestling with "the essential question of evil" in the face of "an incomprehensible universe" [19], is consistent with the whole atmosphere of rationalism and skepticism in eighteenth-century thought, from Locke and Hume to Diderot and d'Holbach. But there is a place, a need for religion in the state; an attractive and national system of belief is required for the efficient government of the masses, as Mahomet well understood (II, v; IV: 126). This restriction of religion to its social role, without a need to establish the truth of that religion, fits in with Voltaire's non-theatrical treatment of religion. Both Pomeau and Gay agree that Voltaire never wholly liberated himself from the idea that the "canaille" deserved a social religion that was more vulgar that the true faith of the philosophers, without which they had no moral restraints. [20] As we saw earlier, wise kings do not reject religion; they merely control it within their reasonable laws, a goal Voltaire promoted widely in his writings [21]. It is a maxim frequently reiterated within the plays that the priesthood not be banned, merely made subordinate to the  173 throne: Un pretre...quelque dieu qui l'inspire, Dolt prier pour ses rois, et non pas les maudire. (Oedipe III, iv: II: 88) Mahomet, as a conscious statement against fanaticism, is opposed only to the kind of religion which refuses to be bound by reason and duty: "la piece n'est au fond qu'un sermon contre les maximes infernales qui ont mis le couteau la main des Poltrot, des Ravaillac et des Chátel." (to the count d'Argental, 7 Dec 1742, D 2696) In writing it Voltaire was aware of being restricted by "l'attention a ne pas dire tout ce qu'on pourrait dire" (to the marquis d'Argenson, 26 Jan 1740, D 2148), in view of the considerable power wielded by the politically influential and interventionist Church of his time. Mahomet himself is called an "imposteur" throughout the play; there is no question of his believing what he preaches, in contrast to the misguided sincerity of Sdide. The problem of whether divine inspiration actually occurs is skirted; the play attacks instead the hypocritical pretence of religious inspiration for material ends. Badir observes that religion is simply a practical means to a political end in Mahonet, the unification of the Arabs: "ndcessaire au gouvernement de peuple...ddsormais subordonnde a la souverainete du monarque." [22] While most of these plays portray, unfavourably,  174 ancient and pagan priesthoods, and leave their audience to draw the obvious parallels to contemporary religious politics, Voltaire does on one occasion refer directly to one form of Catholicism, that which arrived in the Americas with the sixteenth-century Spanish explorers. Although no priests appear in Alzire, the action of the Spanish in conquering, forcibly converting and plundering the Americas is condemned as unrestrained greed masked as religious zeal: "Ces tyrans cruels...Qui depeuplent la terre...Dont l'infame avarice est la supreme bit" (ed. Braun; II, ii, 71-75). This outspoken criticism does not appear to have provoked any serious response, perhaps because it was masked by the expedient conversions of both the cruel Gusman and the pagan Zamore in the denouement.  Les Guebres is very outspoken on the dangers of elevating religion to the status of government and permitting the persecution of "alien" sects, as leading to injustice, violence and arrogance. The denouement restores the primacy of secular authority under the emperor. A similar plot structure was used in Les Lois de Mipos, attacking the evils of theocracy and persecution as producing a bloodthirsty and inflexible priesthood in control of an unjust society. Individual priests are not portrayed with the same sense of cynicism and criticism that was directed against organized religion. They are frequently used as sources of  175  knowledge and experience in the exposition of plots. As Voltaire himself phrased it in a note to Olympie, "Les heros agissent, et un grand-pretre instruit" (VI: 127, footnote to III, ii). We find this type of character in  Oedipe, the original Eriphile, Merope, Semiramiq and Olympie. we should point out that in the first private performances of Olympie, Voltaire played the high priest himself. There are within the tragedies many priests held up as good examples. But their authority is consistently limited to moral issues. The high priest in Oedipe, although attacked by Oedipe as "sacrilege" (III, iv; II: 88), is sincere and accurate in his prophecies, and takes no part in the political disputes. °roes in Semiramis, who is essential to the revelation of Arzace as Minias the lost heir, is portrayed as devout and apolitical: Renferme dans les soins de son saint ministere, Sans vaine ambition, sans crainte, sans detour, On le volt dans son temple, et jamais a la cour.... Moins ii veut etre grand, plus il est revere. (I, i; IV: 510) The high priest in Olympie, speaks of the alliance of faith and law against violence (II, v), and accepts the limitation of his ministry to religious matters: Me preservent les cieux de passer les limites Que mon culte paisible a mon zele a prescrites!  176 Les intrigues des cours, les cris des factions, ... N'ont point encor trouble nos retraites obscures:... Les debats des grands rois... Ne sont connus de nous que pour les apaiser. (III, ii; VI: 127-28). To make the point even more strongly in the printed text, this twelve-line speech is accompanied by a three-page footnote criticising the barbarity of priests in the Old Testament. The same kind of apolitical priest is set forth as ideal in Les GuébKes: Des homes de paix Des ministres cheris, de bonte, de clemence, Jaloux de leurs devoirs, et non de leur puissances ...par les lois soutenus Et par ces memes lois sagement contenus Loin des pompes du monde enformes dans leur temple. (V, vi; VI: 566) Iradan, in his confrontation with the high priest, compares that prelate unfavourably with the same ideal: Les pontifes divins, justement respectes Ont condamne l'orgueil, et plus les cruautes. (I, iii; VI: 510) Le vieil Arzemon, seen by contemporary readers as a portrait of Voltaire, repeats the same lesson: Eux qui de la pitie devaient donner l'exemple, Eux qui n'ont jamais da penetrer chez les rois  177 Que pour y temperer la durete des lois. (V, ii; VI: 560) Thus the portrait of the ideal priest in these texts appears to be as consistent and enduring as that of the good king. There are also certain examples of what the priest should not be, as there had been for kings. The high priest in 1s Gubbres violates the limits on his role most clearly in abrogating to himself the powers of a king: "Nous seuls devons juger, pardonner, ou punir." (I, iii; VI: 510) He, like others illegitimately exercising power, is labelled a tyrant, by Cesene (I, ii) and Iradan (II, i). He is as insincere as Mahomet, a priest for worldly motives of wealth and power, and unlike Mahomet is punished with more than remorse: Ii blasphemait ses dieux qui l'ont mal defendu, Et sa mort effroyable est digne de sa vie. (V, iv; VI: 562) As in the examples of kings, the texts give us right and wrong examples in order to make clear the nature of the ideal being presented: compassionate, apolitical and reasonable. There are nonetheless several characters in these tragedies for whom sincere religious belief is painted as a sympathetic characteristic rather than a tragic fault. The restraint of belief by reason and compassion is implicit in these favourable portraits. Alvarez in Alzire is one of the  178 earliest examples. He tries to bring his son into the same conception of religion: J'en ai gagne plus d'un, je n'ai force personne; Et le vrai Dieu, mon Ills, est un dieu qui pardonne. .... Tout exces merle au crime. [23] When Gusman repents, he returns to his father's God of pity and reason (V, vii, 211 - 226). This kind of faith is exercised within the bounds of reason, mercy and social responsibility. As Iradan expresses it in Les Guebres: Un sujet gouverne par l'honneur Distingue en tous les temps l'Etat et sa croyance. (I, i; VI: 506) Faith which displaces reason leads to evil, and is strongly condemned, notably but not exclusively in the character of Seide (Mahomet III, vi). Cassandre and Olympie, when they take their oaths before the altar, actually pray on stage. These prayers are lengthy expressions of belief attested by their having passed through the mysteries (occupying all of I, iv, 39 lines), and bear close resemblance to orthodox belief in Voltaire's society. Cassandre invokes: Dieu des rois et des dieux, etre unique, eternel!... Qui punis les pervers, et qui soutiens les justes.... Confirme, Dieu clement, les serments que je fais! (VI: 107-08) Olympie implores:  179 Protegez a jamais, ô dieux en qui j'espere, Le maitre genereux qui m'a servi de pbre, Mon amant adore, mon respectable dpoux; Qu'il soit toujours cheri, toujours digne de vous! (VI: 108) However, these prayers go unanswered within the action of the play because of the characters' sins against duty to king and father, namely Cassandre's participation in the rebellion against Alexandre and Olympie's inability to subordinate her love for Cassandre to her duty to her father. The hierarchy of duty and the inevitability of punishment we noted earlier are still operational. Statira, Alexandre's widow, holds retribution against Cassandre as a higher duty than that of her vows as a priestess, the state above religion (II, ii and iv), and is not rebuked. Another unexceptionable statement of faith within these plays occurs in the very late play Les Guébres. Arzame, while considered an infidel by the priesthood, believes in a recognizable God: Dieu qui le fit.. .Dieu son seul auteur Qui punit le me-chant et le persecuteur. (I, iv; VI: 513) This passage establishes that she worships a spiritual being and not the material sun. The list of duties imposed by her faith are hardly iconoclastic: obedience to parents, fidelity to kings even the unjust, protection of innocence,  180 justice, generosity, and compassion (ibid). By no means an isolated incident within these plays, Arzame's confession closely resembles the faith of the Chinese in l'Orphelin de la Chine, meant to contrast with the chaotic violence of the Mongols and defined as: "le droit foi de l'hymen,^l'honneur, la justice, le respect des serments." (IV, iv; V: 340) We can even include in this credo a belief in life after death; in the face of imminent execution, Arzame anticipates: Par lui persecutde, avec toi reunie, J'oublierai dans ton sein les horreurs de ma vie. (III, v; VI: 542) However, while faith may have a legitimate place in the social structure, subordinate to one's duties to the state, conversion between faiths has a dubious status. As Scott-Prelorentzos observes, even Zaire is an ambiguous statement on tolerance, in that religion is made to demand the destruction of an otherwise desirable human relationship [24]. To abandon the faith of one's birth, one's father, is rarely rewarded. While sincere and unpunished conversions do take place, notably in Alzire, they occur solely on the basis of persuasion by a father figure, either Alvarez (ed. Braun; I, i, 129-134) or Mont&ze (I, iv, 255-62). Alzire makes use of the Spanish declaration in the Americas that "la loi pardonne a qui se rend chretien." But Zamore refuses with Alzire's support: "Renoncer aux dieux que l'on  181 croit... 'est le crime d'un lache." (V, v, 142 and 185-86) We are meant to understand that he does not actually convert in the play (V, vii, 256; see note). Mahomet further illustrates the fundamental injustice of conversion by force or as a result of military conquest (II, v; IV: 125); Zopire is to be admired for refusing to recant the faith of his fathers (I, i; IV: 107). Zaire is called Christian because she was born of Christian parents, but she has no knowledge of the faith and has grown up in a Muslim environment (I, i). The impact of her often-quoted speech on diversity of belief which conludes "l'instruction fait tout" (ed. Jacobs; I, i, 109) is undercut, however, by the plot, in which birth and not environment is the key factor: The calm philosophic phrases of the first scene do not stand up in the face of the revelation of her origins and the importunings of her father and brother.^[25]  Lusignan appeals to his daughter in heavily weighted terms: Songe au sang qui coule dans tes veines: C'est le sang de vingt rois, tous chretiens come moi. (II, iii, 344-45) Her conversion from the Muslim faith in which she has grown up is swift (II, iii between lines 332 and 378). Her father's faith and his approaching death are used to pressure her to renounce Orosmane 4nd be baptized (III, iv).  182 She calls on the God she knows only as "Dieu de mon 'Dere" (a significant epithet) for help which does not appear to come (III, vi, 234 and V, iv, 73). While this play was not censored as anti-religious during the eighteenth century, it raises for present-day critics serious questions on the superiority of one religion over another.^Zaire falls within the paternalistic system by returning to the faith of her fathers, but the text does not otherwise uphold the orthodox view of a spiritual reward for conversion. Zulime's proposal to convert to Christianity in order to marry Ramire is not even taken seriously within the play; it is suggested in one scene (II, ii) and not therafter discussed. She returns at the end to the authority and faith of her father (V, ii). Thus we observe conversion is portrayed not as a matter of personal belief but rather as an act which involves an entire social unit. The individual must remain under paternal authority, whether this involves a change in belief or fidelity to the creed into which one is born. Religion can therefore be seen to take its place within the social hierarchy, imposing duties that are necessary but subordinate to those imposed by the State and which ideally fall within the bounds of reason, compassion and paternal authority. The State's control over religious expression and over the priesthood as an institution should be absolute. The substance of belief in the plays, however, bears  183 considerable resemblance to that which would be acceptable in French society of the eighteenth century. True skepticism, doubt of the existence of a god rather than of His mercy or willingnes to intervene directly in human affairs, appears much less often in those plays written after the end of the Regency. The considerable powers of the Catholic Church in French society, demonstrated by the banning of Mahomet and Les Guebres, to give only examples from the plays under study, were doubtless a force to be reckoned with for a writer such as Voltaire as well as a target for his persistent criticism. A writer who wished to have his texts published or performed had to keep them within the limits set up by secular and religious authority. The relationship between religion and family, particularly the authority of the father over belief, which we have indicated here, will be developed in the next section.  C. The Family within the Social Hierarchy We have spent considerable time examining the prevalence of the father as the actual and metaphorical representative of authority at all levels of society, from emperors to gardeners. Haydn Mason's contention that in  Mahomet the primary struggle is between good and bad father figures can with profit be extended to many other tragedies in the Voltairean oeuvre. [26] It has been made clear that within these plays female characters have little if any active role to play in the  184 sphere of government or religious institutions. Even Statira is as a priestess under the authority of the high priest (Olympie II, ii) and draws such limited power as she exercises from her status as the widow of Alexandre. The paternalistic model of authority at the State level requires a similar structure within the family itself. One can find a wide range in Voltaire's writing with regard to the role and value of women in society, from the almost feminist tone of the preface to Alzire, in which he praises Mme du Chatelet's erudition, to the prevalence of women as pathetic victims in his theatre, which Lucette Desvignes has pointed out in her article in the Revue des sciences humaines[27]. As Albistur and Armogathe conclude: L'attitude de Voltaire envers les femmes est, elle aussi [come celle de Diderot], entachee d'ambigultd. Certes Voltaire s'est elevê a plusieurs reprises contre les injustices dont les femmes sont victimes n'est-ce pas pluteit parce qu'elle est un étre faible opprime, que parce qu'elle est vraiment une &gale privee par l'usurpation masculine de ses droits? [28] Indeed, with the exception of Condorcet, the philosophes offer little in the way of evidence for a belief in equality of the sexes, according to the Histoire du feminisme. Nonetheless Voltaire's theatre in eighteenth-century performance is marked by productive working relationships between the author and the strong, creative actresses of the  185 period - Lecouvreur, Quinault, Clairon. These actresses found many of their star roles in Voltaire's plays, which may explain their enthusiasm in encouraging, critiquing and promoting his work. The implications of these roles are less straightforward now, insofar as they can be seen as normative or prescriptive, idealizing in some way desirable or undesirable behaviours within the family and within the society as a whole. The questions which arise are whether these female characters are powerful or passive within the family, whether they act from reason or sentiment, and whether they are rewarded or punished differently from their male counterparts. Desvignes has indicated the Racinian tendency to martyrdom in Voltaire's female characters [29]. No such tendency can be observed in male characters; the most frequently occurring types are the young men struggling to control emotions or overcome destiny - Lekain's star roles of Arzace, Vend6me, and Orosmane - or the wise old men, fathers or priests, who were Voltaire's own favourite amateur roles. The theatre under study expresses considerable ambivalence towards female sexuality, and develops a complex relationship between sexuality and maternity, particularly in the inconsistent treatment of incest, an interrelationship suggested and summarized in Voltaire's nickname for the niece with whom he had a affair, Hmamann. It is a commonplace of Voltairean criticism that his theatre was a  186 medium for propagating his ideology. This ideology could be more conservative and patriarchal than many critics believe, especially in this domain of domestic organization. The philosophical ideals of freedom of thought and belief were not always extended to women. In the microcosmic world created of his tragedies, female sexuality is a force to be feared and controlled whether by external paternalistic authority or by internalization of the values and duties imposed by that authority. Male sexuality, however, attracts no such concern. Such duties imposed on women are frequently impossible or internally inconsistent, from which dilemmas Voltaire often draws his tragic plot. Death becomes the favorite resolution to the impossibility of living as a woman (Hdrode et Mariamne, Zulime, Rome sauvee, Olympie), when it is not the favoured punishment for conscious or unconscious crimes against those duties (Zaire, Eriphile, Sdmiramis, Oreste). The tragedies of men are not usually based on such dilemmas, but rather on the relentless pressure of destiny. The situations of incest in these plays highlight this strict code of behaviour applied to women, and illustrate the greater dangers of unrestrained sexuality in women than in men, through the inconsistent treatment of the men and women involved. When a female character expresses love, it is usually in terms of subservience to the beloved. Men are painted as  187 struggling against this "enslavement", and resentful of it: Ii vaut mieux sur mes sens reprendre un juste empire; Ii vaut mieux oublier jusqu'au nom de Zayre. [sic]  (ed. Jacobs; III, vii, 325-26) But Zaire exults, Je ne connais que lui, sa gloire, sa puissance; Vivre sous Orosmane est ma seule esperance. (ed. Jacobs; I, i, 25-26) and tells him, J'ai par dessus vous ce plaisir si flatteur... De tenir tout, seigneur, du bienfaiteur que j'aime. (ed. Jacobs; I, ii, 220-22) This term "bienfaiteur" is also used in situations where the female dependence is on a father figure, suggesting the blurred line between affection for a male parent and a male lover, as between Arzeme and Iradan in Les Guebres. When Zulime describes herself as "fidele a mon epoux et soumise a mon maitre" (II, ii: IV: 28), the association between love and self-abnegation is heightened by the fact that Zulime is a reigning princess and Ramire a slave. The woman in love is not an independent agent, but an extension of that loved man's identity and ambition, and so an accomplice to all his actions. The duty and love owed by a daughter to a father are spoken of in the same hierarchical terms. This control over the daughter's sexual desire and expression by the father,  188 may be illustrated first by the number of attempted and actual forced marriages in the texts. Jocaste speaks of both her marriages as forced on her: Deux fois de mon destin subissant l'injustice J'ai change d'esclavage, ou plutOt de supplice. (II, ii; II: 72) To her example we can add the alliances of Eriphile to AmphiaroUs, Semiramis to Ninus, Alzire to Gusman, Artemire to Cassandre, and that proposed of Olympie to Antigone. The power relationship between mother and daughter is germane to only one text of those under study, Olympie, in which the mother, Alexandre's widow Statira, acts as a channel for absent paternal authority. Most often the daughter is considered only in terms of her struggle to reconcile two male systems of authority, paternal and marital; the mother is absent. Such is the case for Zaire, Zulime, Alzire, Palmire, Aurelie, Amenaide, Arzbme and Irene. There is no text in which both parents are active in a conflict with an adult child. The stormy relationship between Clytemnestre and Electre shows to what extent loyalty to the father permits disrespect towards the mother. The father's authority is unquestionable within this depiction of social roles. This emphasis on the nature and extent of parental authority is usually explained in biographical terms by Voltaire critics [30]. In the articles cited above, Mason and Spica both point out the  189 duality of paternal authority in Voltaire's oeuvre, parallel to that of the priesthood and the divinity, in that each is seen as either vengeful or forgiving. The play texts under examination, however, concentrate on the legitimacy of paternal authority as a basis for social order, whether the individual father is just or unjust. Idamd describes the Chinese empire, an ideal rational and just society contrasted with Mongol brutality, as first "fonde sur le droit paternel, sur la foi d'hymen" and then "sur l'honneur, la justice, le respect des serments." ^(L'Orphelin de la  Chine IV, iv; V: 340) To a daughter, her parents are "du dieu que nous vive image." (V: 339) In the same way, Arzeme's confession of belief begins "qu'on soit soumis aux lois de ses parents." (Les Gu6bres I, iv; VI: 513) Paternal authority over children appears not as a deplorable aspect of the ancien régime, but as a necessary component of the philosophes' ideal society. This submission of daughter to father is not weakened or impeded by years of separation. For example, Zaire's first word to her father on learning of their relationship is "seigneur", symbolizing her immediate recognition of his authority (ed. Jacobs; II, iii, 319). Knowing that her love for Orosmane will be blocked by a return to Christianity, she still asks "mon 'Dere, cher auteur de mes jours, parlez, que dois-je faire?" (ibid., lines 375-76) She admits no doubt of her decision:  190 Quoi, je suis votre soeur, et vous pouvez penser Qu'A mon sang, a ma loi, j'aille renoncer? (ed. Jacobs; III, iv, 79-80) Indeed, despite her expressions of regret and anguish, her resolution remains firm through the rest of the play. Alzire extends her filial duty beyond abandoning Zamore to marrying the brutal Gusman at her father's order: Je sais ce qu'est un 'Dere, et quel est son pouvoir; M'immoler quand ii pane est mon premier devoir.... Mes yeux n'ont jusqu'ici rien vu que par vox yeux Mon coeur change par vous abandonna ses dieux. (ed. Braun; I, iv, 251-56) Palmire, raised in the belief that Mahomet is her de facto father: "Mahomet m'a tenu lieu de 'Dere" (I, ii; IV: 111), obeys him as such : "un dieu qui m'epouvante" (IV: 110). The texts may have indicated reservations about the priest's ability to represent divine authority over secular rulers, but not the right of husbands and fathers to that divine status. In many cases, however, the father role, as do those of certain kings, serves to exemplify justice, compassion and the reasonable exercise of authority. In this regard we can cite Alvarez, Zopire, and the foster father old Arzemon in Les Guebres. Gusman's faults in Alzire stem from his rejection of his father's example; his dying repentance reinstates that ideal of compassion and justice:  191 Mon ame fugitive, et prete  a  me quitter,  S'arréte devant vous...mais pour vous imiter. (ed. Braun; V, vii, 213-14) Zopire, separated from his children by war, acts as a wise and generous foster father to them without realizing their identity. His virtuous nature, in their ignorance of the relationship, acts as a serious obstacle to Mahomet's command to kill him (IV, iii). In his last appearance, he implores the gods for the return of his children (IV, iv). He, like Zaire and Clytemnestre, receives his request literally: Rendez-moi mes fils a mon heure dernibre Que j'expire en leurs bras; qu'ils ferment ma paupi6re. (ibid., IV: 148)  As he lies dying, he forgives them, even as Sómiramis forgives Arzace and Eriphile Alcmdon. Let us point out, however, that Zopire, unlike the mother characters, dies innocent, and the true responsibility for his death lies with Mahomet. Old Arzemon, in Les Guebres, is in his careful upbringing and wise advice essential to the happy outcome, and serves to contrast with Iradan and C6sene, who sacrificed their children in obedience to their military orders. He, the foster father and humble gardener who advises the emperor, was considered a portrait of Voltaire from the play's appearance in print.  192 No female character more vividly illustrates the absolute and compelling nature of paternal authority over a daughter than Electre. Her time on stage is devoted to mourning her father, castigating her mother, and beseeching the heavens to send Oreste to avenge the murder  (Oreste  I,  ii; II, vii; III, iv). The marriage proposed between Electre and Plisthbne, Egisthe's son, which compounds regicide with a kind of incest, is so unthinkable that it disappears from the plot even before Plisthene's death is reported. Agamemnon is afforded a semi-divine status, and his murder is sacrilegious. Electre never mentions Agamemnon's sacrifice of her sister Iphegdnie. Apparently that killing was not sufficient to permit insubordination. She invokes the furies in the names of Oreste and Agamemnon (IV, iv), and offers no hope of her mother's salvation: "Peut-elle r6parer les malheurs qu'elle a faits?" (V, ii; V: 144) Nonetheless, as the representative of her father's spirit crying out for vengeance, her own hands are clean of her mother's blood. Oreste alone carries that guilt: "Qu' avez-vous fait, cruel?...Quoi, de la main d'un fils?" (V, ix; V:^154-55) Zulime may defy paternal authority, but not without regret and contrition. Her freedom and independence are seen as a failure on Bênassar's part (II, iv), and resolution requires the re-establishment of his authority.  193 From the opening lines of the play she expresses contrition for her reckless action: Je n'ai plus de patrie Je l'outrage et je l'aime; il est assez venge. (Zulime I, i; IV: 11). When she is brought before her father, she uses the language of a subject: "Seigneur, mon souverain, j'ose dire mon p6re." (II, iv; IV: 31) Not even her love for Ramire can destroy her filial devotion: Et ce coeur, tout brillant d'amour et de colbre, Tout forcen6 qu'il est, volt un dieu dans son pêre. (V, ii; IV: 58) Again we see reiterated the divine status of paternal authority over the daughter. Aurdlie, like Zulime, suffers for her flouting of paternal control and for choosing a husband freely: Ma perte fut certaine au moment oil mon coeur Recut de vos conseils le poison seducteur. (Rome sauvde III, ii; V: 239) Her unwitting involvement in Catalina's conspiracy against Rome and in her father's death drives her to suicide (IV, vi). The father's power extends from control over sexuality to control over belief; the daughter remains faithful or converts to a religion as he does, as we saw in the cases of Zaire and Alzire. When Mahomet boasts,  194 Votre coeur a-t-il pu, sans etre dpouvante Avoir un sentiment que je n'ai pas dicte? (III, iii; IV: 133) he is expressing no more than normal paternal influence over a daughter. Only when that authority, whether of father or lover, has lost its legitimacy does the female character resist or argue. Once Zaire has decided that filial duty outweighs her love for Orosmane, she can say to him "Vous, seigneur! vous osez me tenir ce langage!" (ed. Jacobs; IV, vi, 308) Mahomet's power over Palmire is broken when she learns he is not her father, and she repulses him: "Imposteur teint de sang, que j'abjure a jamais!" (V, ii; IV: 158) VendOme's status as a rebel negates any legitimate (as opposed to practical) power he might have had over Adelaide, and she never treats him with the respect his actual power might deserve. Among these self-consciously virtuous and obedient daughters, Amenaide, in Tancrede, stands virtually alone as independent. She was reared by her mother alone and betrothed to Tancréde in exile. On her mother's death, she finds herself forced to be self-reliant: Je me vis seule au monde, en proie a mon effroi ...n'ayant d'appui que moi. (I, iv; V: 509)  195 She has but late come under her father's control, and does not have Zaire's swift docility: Je vous ai consacre mes sentiments, ma vie; Mais pour en disposer, attendez quelques jours. (V: 511) Argive, in view of his insufficiencies as a father, grants an unparalleled concession: Ma fille, je n'ai plus d'autorite sur toi. J'en avais abuse, je dois l'avoir perdue. (IV, vi; V: 549) No other father, not even the unreasonable and ambitious fathers of Irene and Zulime, makes such a confession of failure. It cannot be concluded, however, that Amenaide is rewarded for her independence of spirit. Rather it becomes the instrument of her downfall, preventing her from explaining the misdirected letter either to her father (in IV, iv) or to Tancrede (in IV, v). Her silence is made the cause of Tancrede's death, and thus of her own heartbreak. The duties of marriage allow no exceptions to the wife, whether the marriage was against inclination or the husband is brutal and even criminal. There are no tragedies of the husband's failures causing his punishment, only those of jealousy and domestic violence, namely Herode et Marianne and Zaire, where one is meant to pity rather than condemn the excess of emotion, as even Nerestan does: Faut-il qu'a t'admirer ta fureur me contraigne,  196 Et que dans mon malheur ce soit moi qui te plaigne? (V, x, 229-230, the closing lines) Semiramis, however, speaks for a number of Voltaire's heroines when she confesses: Plus les noeuds sont sacres, plus les crimes sont grands. J'etais epouse, Otane, et je suis sans excuse. (I, v; IV: 517) While the father has virtually divine authority over the daughter, that power passes with the same prestige to the husband over the wife. Although Eriphile was married at sixteen without consultation or affection, her husband is a "demi-dieu dont je fus la coupable moitie" (Eriphile I, iii; 11: 466), and her complicity in his murder the crime she cannot escape. Clytemnestre, although active in the murder of Agamemnon, sinks into subordination afterwards; Egisthe becomes "un maitre absolu" and she no more than "sa premiere sujette." (Oreste I, iii; V: 100) Artemire, faced with a violent and irrational husband to whom she was married against her will and who killed her father, accepts that it is her duty to obey and respect him, to take no action against him or even to save herself from him: Le ciel qui me poursuit me l'a donne pour maitre. Je connais mon devoir.  (Artemire 1, ii; II: 130. See also II, i) The same theme of the virtuous wife and the abusive  197 husband is taken up in Mariamne: Malgre ses cruautes, malgre mon desespoir, Malgre mes interets, j'ai suivi mon devoir. J'ai servi mon epoux, je le ferais encore. (II, v; II: 186) Mariamne is so insulted by the insinuation that she had betrayed Herode with Soheme that she refuses to lend colour to it by supporting the uprising or leaving the palace with him (V, ii-iv). Thus the uprising is crushed and she is executed (V, vi). Indeed, Voltaire was so concerned with holding Mariamne to a standard of virtue that he eliminated evidence of her affection for Varus (before this character was changed to Soheme), because "elle ne servait qu'a justifier sa condemnation et par consequent a diminuer la compassion qu'on doit avoir pour elle." (October 1723, D 171) The social values of his audience would not permit such a lapse from the norms of wifely behaviour. Thus the textual transformation acts to strengten, not challenge, conventional norms. Nor is the 1762 Mariamne the last staging of this theme. Voltaire's last complete play,  Irene (1778), again draws its tragic dilemma from the wife's inescapable duty to an unloved and unjust husband. Widowhood, however, is not in these plays the woman's emancipation from male authority, as comedic tradition might lead us to expect, but rather a new set of restrictions. Chief among these is the strong bias against remarriage. A  198 second marriage, even when the character does not instigate it, is (in Merope's words) treason to the first husband and to the son of that marriage (eds. Vrooman and Godden; I, iii). In this area, Voltaire reiterates the laws and social norms of his period. Abensour's summary of the legal position of women in eighteenth-century France sets forth the ambivalent status of the widow: Est-ce a dire, cependant, que la femme veuve soit, meme au point de vue de ses droits familiaux ou de la liberte de disposer de ses biens, absolument l'egale de l'homme? Pas tout a fait. [31] Many aspects of Abensour's analysis of French social structure, such as the strict standards of modesty in dress and behaviour, the legal difficulties placed in the way of remarriage, and the insistance on patrilineal inheritance accompanied by the transfer of authority from father to son re-appear as social norms in the tragedies under study, echoing rather than challenging the status quo. Sexual desire after childbirth and/or bereavement is characteristic of the "evil" wives and mothers such as Eriphile and Semiramis. Statira, in direct contrast, provides the example of correct behaviour by withdrawing to the seclusion of a temple from which she is only drawn by the revelation of her daughter's impending marriage to a former rebel against Alexander (Olympie II, ii). Not even fifteen years of external and internal conflict can persuade  199 Merope to hold the throne in her own right. There is no example of a concluded second marriage in the plays under examination, not even between Clytemnestre and Egisthe, however implausible its delay may seem. Where one does occur, in Oedipe, it is the catalyst to the tragedy itself. Female sexuality, largely controlled by obedience to male authority, is shown to find its fulfillment in maternity, and particularly in the birth of a son. Good mothers channel their energies and ambitions into him, and sacrifice for him. The character of Merope is a notable example, as is that of Idame.^There are few legitimate outlets for a woman's ambition and ability beyond sublimation in her children. Ambitious women are failed and guilty mothers within these plays: Clytemnestre, Eriphile, Semiramis. They suffer remorse and punishment because they dare to pursue their own desires, and allow anything, even a divine prophecy, to interfere with the mother/son bond. Jocaste, years after the event, speaks of her regret that she obeyed the oracle in sacrificing Oedipe; it is a source of her lack of faith in the gods, that they could ask such a thing (Oedipe IV, i).^Merope, even after fifteen years of civil war, and despite her status as the daughter and the widow of kings, refuses to seek the throne for herself, which would seem an excellent way to ensure political stability. Throughout the play she is solely concerned with finding her son and seeing him crowned (I, i; II, i; III,  200 ii). This coronation is her last act and the last scene in which she speaks in the play (V, vii). The relationship between mother and son is paramount for the mother. After failing to reconcile Electre to the status quo, Clytemnestre renounces all ties to her (in II, v), but welcomes Oreste's return though it means her death, even risking herself to protect him from Egisthe (III, vi and IV, viii). She promises, "j'obtiendrai sa grace, en dusse-je perir" (V, iii; V: 147), a promise kept to the letter. This revived mother-love gives her the courage to defy Egisthe her master: C'est trop braver peut-étre Et la veuve et le sang du roi qui fut ton maitre.  (ibid) But Clytemnestre, who did not appear troubled by her acrimonious break with Electre nor Egisthe's treating Electre as a slave, confronts an irreconcilable conflict in those duties imposed on mothers and wives, which leads to her death in intervening between Oreste and Egisthe: Je suis epouse et mere, et je veux a la fois, Si je puis etre digne, en remplir tous les droits. (V, v; V: 150) Among these failed mothers, Semiramis is portrayed, somewhat against the historical record, as turning to government not out of ambition but as a consolation for her unhappy marriage and the sacrifice of her son to another  201 oracle (III, i). Suppressed mother-love is at the heart of her attraction for Arzace, and her final lines are "Je te pardonne tout .... 0 mon fils, mon cher fils...C'en est fait." (V, viii; IV: 566-67) Apparently this forgiveness spares Arzace Oreste's punishment for matricide. When Arzace accedes to the throne, legitimate authority is passed from father to son. The same pattern was set in the precursor to Semiramis, Eriphile. Mdrope, the good mother, preserves that channel in keeping the usurper from the throne; Semiramis or Eriphile, in taking power for herself, becomes the usurper and must herself be removed. Such authority as a mother, rather than father, has over her child is most clearly seen in Olympie, where the father is dead and Statira is "veuve d'un demi-dieu, fille de Darius." (II, ii; VI: 115) As an agent of male authority, she can command Olympie and expect compliance, regardless of the fifteen-year separation and of Olympie's feelings for the man who has saved her from slavery. Olympie hardly dares protest: "Permettez...que je vous fasse entendre une timide voix." (III, vi; VI: 137) Statira's intention is to put her daughter under what she considers legitimate male control, that of Antigone: Si vous la protegez, Si vous vengez son pbre, (emphasis mine) Je ne vois plus en vous que mon dieu tutelaire, (III, vi; VI: 135)  202 which is surely an unexpected thing for a high priestess to say to any man. Statira, in a classic controlling gesture, uses even her own death to tighten Antigone's claim: Elle m'aime, et Vordonne Que, pour venger sa mere, elle epouse Antigone, (IV, viii; VI: 151) an order which he undertakes to see obeyed (V, ii). Olympie, faced with Zaire's dilemma between duty to parents and love, makes Zaire's choice: Cet hymen si cher etait un crime horrible.... Rendez-moi digne du grand nom qui vous reste: Le devoir qu'il prescrit est mon unique espoir. (II, iv and vi; VI: 122-24) Confronted by this complex system of strict but overlapping and conflicting duties to various male authority figures within the family structure, these female characters inevitably fail. They take blame, they are tormented by guilt, and they are punished. We will not see the same degree of suffering and punishment inflicted on male characters, even those who commit matricide. The key relationship for a male is that to his father, not his mother or wife, and we see in many texts an emphasis placed on reconciliation between father and son, as the son avenges the father's death and /or inherits the father's values and status: Oedipe, Eriphile, Mahomet, Alzire, Oreste,  Semiramis, les Guebres. Both Mason and Spica, in  203 the articles already mentioned, find in this recurrent theme an autobiographical element. They do not suggest, nor do I, that the ambivalence and even hostility with which the wife and mother roles are portrayed have a similar source. In view of the open-ended and dialogic writing process we have established for these texts we hesitate to make any unequivocal statements for an autobiographical source. For our purposes, blame can be defined as the taking or accepting of responsibility for another character's actions. When Jocaste refuses to take blame: "J'ai fait rougir les dieux qui m'ont forcee au crime" (Oedipe V, vi; II: 111), she is acting exceptionally. Acceptance, if not self-attribution, of blame is far more common for female characters, and requires no logical connection with the crime. The social principle may be seen even in Voltaire's  drame bourgeois Nanine, where the count attributes to women the responsibility for maintaining the moral character of men (I, i; V: 16).^Mariamne is blamed for her own death at Herode's hands, an action characteristic of the abusive husband: "Elle a voulu sa perte; elle a su m'y forcer."  (Herode et Mariamne V, vii; II: 215) Adelaide, in the same way, is informed by VendOme that while he ordered Nemours' death, she bears the responsibility for it: "Vous avez dicte sa sentence mortelle." (ed. Cartwright, Adelaide, V, iii, 133; in the 1734, 1751 and 1765 versions) VendOme also blames Nemours' death on Coucy, the officer to whom he gave  204 the order (V, v, 143 in 1734; V, iv, 125 in 1751; V, iv, 134 in 1765), an action strongly contrasting with the enthusiasm shown by female characters in taking blame. Zaire, who falls in love with Orosmane not knowing her origin, is informed by her loving Christian brother Nerestan: "Vous demandez la mort, et vous la meritez....tu vivras fidele, ou periras martyre." (Jacobs; III, iv, 132 and 182) Furthermore, she does not challenge his judgment; indeed she asks for death as a solution, wishing only that Orosmane would close her eyes (IV, i, 35-38), a wish which is literally granted. Alzire marries Gusman in obedience to her father's order and only after news of Zamore's death. Nonetheless, she considers that marriage her conscious crime against Zamore, again meriting death (Alzire; III, iv). Artemire in the same way condemns herself for having been forced into marriage to Cassandre (II, ii; II: 135). In Alzire, Montbze's order serves only "pour affaiblir mon crime." (III, iv) But Alzire's love for Zamore becomes her crime against Gusman (V, v); she cannot help betraying them both. In Rome sauvee, Aurelie, who was unaware of Catalina's conspiracy, rushes to attach blame to herself and declares herself his guilty accomplice as soon as she learns of the plot (Rome sauvee III, ii). Palmire, as unconscious as Aurdlie of the intent to murder her father, on discovering the truth takes all responsibility for Zopire's  205 murder by Seide, exculpating even Mahomet (IV, v). Olympie makes herself guilty of Cassandre's crimes against her family committed before she met him: "Votre fille en l'aimant devenait sa complice." (III, iii; VI: 132) Arzame, who did not know of it, takes the blame for inciting Arzemon's attack on Iradan (Les Gu6bres III, iv). Thus we see that within these texts blame is largely subjective and even retroactive, expressing a feeling of responsibility separate from any legal culpability, or even guilty knowledge. Even when the female character is portrayed in a sympathetic way, the patriarchal system under which she lives does not permit her to be innocent; she is automatically complicit with the man under whose authority she lives, whether husband or father. Male figures, on the other hand, are responsible only for themselves, and as we have seen, freely delegate blame to their agents or victims. Furthermore, their violent behaviour is often mitigated and excused. Vend6me's fratricidal jealousy is forgiven because of the depth of his love for Adelalde (V, v, 157-58 in 1734; V, iv, 135-36 in 1751; V, iv, 143-44 in 1765). Arzace is prevented from killing himself for having killed his mother; he must rule after her on his father's throne (V, viii). Iradan and Cesene, who sack the city in which their wives and children are living, are at the end of Les Guebres rewarded by the return of those children and the emperor's favour.  206 Guilt, like blame, is internal rather than external; it stems from conscious violation of one's duties. While remorse and repentance are sympathetic characteristics, they do not constitute extenuating factors. There is no abused wife defence in this system. Artemire and Mariamne are scrupulous in their obedience to duty; their inability to love the husband who threatens to kill them rather than the lover who offers them refuge is the crime for which they feel they must be punished (Artemire III, i; Herode et Mariamne V, ii). Eriphile at sixteen was a passive accomplice to her husband's murder, and has since succeeded in keeping his murderer from the throne. Nonetheless, she is from the beginning of the play consumed by remorse. Repentance is no mitigation : "Vous voyez la mere, helas, la plus coupable, / La mere la plus tendre et la plus miserable." (Eriphile V, i; II: 497) In dying she persists in trying to explain her offence, but never complains of the retribution: Un moment de faiblesse, et meme involontaire A fait tous mes malheurs, a fait perir ton pere. (V, vii; II: 504) Clytemnestre, likewise, from her first appearance on stage expresses grief and regret at Agamemnon's murder, and makes no effort to protect herself from Oreste, whom she knows to be the agent of divine retribution against her. ^Oreste's actions, while punished in the play, were seen as  207 plausible and justified by Voltaire's contemporaries: "Un fils force a hair sa mere la hait bien plus fortement que si la nature n'avait pas mis de lien entre eux. " (Corresp. litt. IV: 440) There is among these plays no parallel situation of a child forced to hate a father, or to be the agent of divine retribution against that father, regardless of the father's behaviour. Even Semiramis, the ruthless and capable Assyrian queen, is painted as devoured by remorse and no longer able to govern. She is not even active within the plot. After a brief appearance in act I scenes vi and vii, she does not re-enter until act II scene vii, in which she informs her accomplice and regent Assur that "les remords, a vos yeux meprisables, / Sont la seule vertu qui reste 'a des coupables." (IV: 532) When she discovers that Arzace is her son, she seems eager for expiation, and calls on him to fulfill the prophecy : "Remplis ta destinde, / Punis cette coupable et cette infortunee." (IV, iv; IV: 555) We do not  see the same kind of persistent and devouring remorse in male characters, even those who commit grievous crimes such as Alcmeon or Arzace or VendOme or Gusman, to say nothing of the unrepentant male villains like Assur, Hermogide, Polifonte and Mahomet. Only when the classical myth requires it do we see the gods' wrath fall on delinquent sons, in Oedipe and Oreste.^Crimes against the father, however unconscious, do result in the son's death as well as  208 in that of Seide. The death which comes as punishment is a welcome release for female characters. Eriphile and Semiramis ask for death at their sons' hands, in fulfillment of the prophecy and as the only expiation for their crimes for which they have been seen to suffer. Olympie, having no such convenient agent of retribution, promises to punish herself for loving Cassandre (III, vi) and keeps her word by throwing herself onto her mother's pyre (V, vii). Zulime's suicide is explicitly made one of expiation for her crimes against her father: "Enfin j'ai rempli mon devoir." (V, iii; IV: 65) To die for one's sins is as much a deliberate choice by the erring woman as a punishment inflicted by some external justice. These heroines know and accept that they deserve their end. In none of his tragedies, not even those with happy resolutions such as Adelaide, Alzire or L'Orphelin de la  Chine, can we see a female character's free choice result in happiness, although there are many instances of male characters achieving happiness through their own actions, as they are elevated to the status of husband and ruler: Alcmdon, Arzace, Egiste. These irresoluble conflicts for women within the authority structure make death an attractive solution for female characters; many express the view that it is the only choice or decision they can make. There is associated with this view of death as a  209 solution a strong tendency to self-sacrifice, to consider one's own death a just price for some desired end. Jocaste on several occasions offers to die if it would save Thebes from the plague (Oedipe II, ii; III, i and v). Mariamne offers herself to stop the civil war being waged on her behalf: "Epuisez tout sur moi." (V, iii; II: 213) Idamd first offers to die in place of the orphan prince so that he and her own son will be saved from the Mongols (L'Orphelin  de la Chine II, iii) and later to avoid Gengis's adulterous advances (V, v). In les GObres Arzbme twice offers to allow the priests to kill her in order to save Iradan and the rest of her family (II, iii and V, ii). Male characters are more likely to act in order to attain a desired resolution: Alcmdon's duel with Hermogide, Oreste's with Egisthe, Arzdmon's attack on Iradan, Soheme's uprising against Herode. Only in Oedipe do we see the king offer to sacrifice himself, as there is no enemy to attack (I, iii). But death seems most often to operate as the resolution to the heroine's own situation of conflict, and as such is described as a gentle or happy outcome. Mariamne considers death the only honourable way out of her marriage: Frappez, le coup m'en sera doux. (IV, iv; II: 210. See also Artómire„ I, i and ii) This resolution would also save her from any implication of adultery: Ii est honteux pour moi de vous devoir la vie.  210 L'honneur mien fait un crime, il le faut expier.  (Herode et Marianne V, ii; II: 211). Alzire, torn between her love for Zamore and her duty to Gusman, exclaims with a similar evocation of Thanatos: Qui me dëlivra par un trepas heureux, de la nêcessita de vous trahir vous deux? (ed. Braun; III, vl 213-14; emphasis mine) Palmire chooses death as her only refuge from tyranny  (Mahompt V, iv). MOrope prefers death to marrying Polyphonte; apparently the idea of seizing power herself is even less attractive (Mêrope III, vii; IV, iii). In all these cases death solves, or is considered capable of solving, the otherwise unresolvable problems of the heroine's situation; it becomes the only way for her to reconcile conflicting duties or to escape being forced into further offenses. It is an interesting contrast that the male accomplices to these crimes are not invariably executed, at least within the action of the play; Hermogide in Briphile and Assur in $emirapis are only taken prisoner. We also notice in Oedipe that while Jocaste dies at her own hand, Oedipe is exiled and his self-blinding takes place offstage (V, vi). The situation of incest provides a particularly intense focus on the limits of authority and sexuality within the family structure. We can find in the plays all three situations of incest: female-dominant, sibling, and  211 male-dominant. The first two types result in punishment for those involved, and are spoken of with great loathing. Jocaste and Oedipe, Eriphile's attraction for Alcmdon, and that of Semiramis for Arzace are all of the female-dominant or mother/son pattern; Palmire and Selde, Arzeme and Arzemon illustrate the sibling. In Palmire's words, such a relationship is "un amour plein d'horreurs." (Mahomet V, iv; IV: 161) The third pattern, possibly because of the similar power held by husbands and fathers over their female dependents, is not seen to be punished. Mahomet expresses no hesitation and receives no criticism for his practice of converting foster daughters like Palmire into wives. Iradan in 1,,ps Guebres, while happy to see Arzeme with Arzemon once he knows they are cousins, feels no guilt over his attempt to marry her. The mere revelation, however, of her unconsummated love for her brother revolts him (II, iii). ArzOme, however, turns not to her natural father but to Iradan for help, and calls him "bienfaiteur" (in II, iii and III, ii; VI: 521 and 537).^One might mention in passing that in Voltaire's Nanine, the count d'Olban disguises his love for Nanine by saying he is acting as a father to her (I, vii; V: 29) There are no cases of incest between a biological father and daughter, only between father-figures and dependent females. It is tempting to see a link between this recurrent father-figure and Voltaire, whose preference for playing wise father roles on the amateur stage as in  212 real life is well-known. Desvignes acknowledges a parallel between the authorial and paternal roles in referring to Voltaire as "'Dere et defenseur de toutes ces victimes," (p. 550) that is, of his tragic heroines, an epithet which is by no means completely favourable in view of the analysis we have undertaken. The texts are scrupulous in indicating that the incest in all cases is unconscious. Jocaste is married to Oedipe for purely political reasons: after his slaying of the Sphinx, to confirm him in the kingship (II, ii). Eriphile is drawn to Alcmeon by something more pure and tender than love, and by her reasonable distrust of Hermogide (II, iv). She believes her own son dead when she proposes the marriage, which is immediately followed by omens of disaster (III, iii).^Arzace feels no attraction to Semiramis, only respect; she insists, as did Eriphile, "Ce n'est point l'amour qui m'entraine vers lui." (III, i; IV: 534) Palmire is drawn to Seide by "un instinct charmant" (Mahomet III, i; IV: 132), Iradan to Arzame by a sentiment stronger than the priests' power (Les Guëbres I, vi; VI: 517). These are instinctive affections which effortlessly become natural familial love on discovery of the blood relationship. Walter Rex argues that in Merope the implicit incest, which was overt in its precursor Eriphile, is so wrapped in idealized mother-love that the passion disappears, and the play's morality appears above reproach [32]. There is no  213 parallel to Racine's Phedre within Voltaire's tragic sphere. Yet despite this careful purity the retaliation is as severe as on those who have actually transgressed. A woman involved in an incestuous affection suffers; the men, with the exception of Oedipe, do not. The exception, one can argue, was forced on Voltaire by the familiarity of the myth, just as Oreste's punishment was. Arzace and Alcmeon kill their unfaithful and potentially incestuous mothers under the same circumstances of divinely caused misunderstanding as Oreste, without pursuit from the Furies, and are even rewarded with accession to their father's throne. While Arzace and Arzemon are discovered to be cousins, and thus are permitted to marry, Palmire and Sdide are siblings, and are made to realize the horror of their relationship before their deaths (IV, v). Thus there is some evidence, in view of the differing degrees of blame, guilt and punishment we have noted, to suggest that the system of justice set forth in these plays has bias, favouring men over women, and father figures over children. It is not new to conclude that the heroines of Voltaire's tragedies are put into insoluble situations, that the standards to which they are held are impossibly high, the punishments they suffer often out of proportion to the crimes (if any) committed. Desvignes indicated the clear pattern of martyrdom, but argued that it was a deliberate exaggeration of the female condition of eighteenth-century  214 France, by which Voltaire intended to evoke sympathy and possibly encourage social change [33]. One should note at this point that contemporary critics recognized and deplored Voltaire's penchant for pathos and remorse in female characterization as leading to implausibility and even derisiveness: Il eat mieux fait de nous donner Semiramis comme nous la represente l'histoire, intrepide guerriere que de nous la peindre comme une femmelette qui craint les  revenants. [34] His heroines, rather than provocative or even normative role models, became frequent and popular candidates for parody in their unnecessary suffering and unlikely silences, as in that of Tancrede called Ouand parlera-t-elle? , written by Riccoboni and performed by the Théâtre italien, and the numerous parodies of Herode et Mariamne listed in the preface to the Moland edition, II: 169. The present analysis goes further than that of Desvignes, and indicates what I consider the significant difference in how these plays critique aspects of French society. When criticizing other elements of contemporary society such as government and religion, Voltaire's texts offered positive models as well as negative ones; we have examined those of what a king, a father and a priest should be. But the texts do not seem to offer alternatives to the social subservience of women. There appear to be no  215 heroines in these plays who succeed in escaping that captivity to duty without suffering retribution. There is not even a consistent reward for female virtue within the tragic universe of the texts: under similar circumstances Zaire is killed but Alzire lives, Mariamne dies and Artemire survives. Incest, which violates the codes of maternal or filial duty, is the most extreme state of non-conformity to the constraints of society on female sexuality. Even when Voltaire's text makes clear that the incest is neither conscious nor consummated, the mere attraction draws the strongest condemnation on the woman involved. Men appear largely to escape retribution for sexual misbehaviour; the association between paternal concern and sexual attraction in Nanine and Les Guébres does not even draw comment. And so we return to our original hypothesis, that within the paternalistic authority structures set up within these plays female independence and especially sexuality are dangerous forces which society must restrain. Denied legitimate power and subject to inconsistent justice in the microcosmic world of these plays, a woman appears best able to keep her various obligations, to be safe within and to society by her retreat into death. While Voltaire's female characters offered excellent roles to eighteenth- century actresses, in the eloquent pathos of their helpless suffering, their attractiveness as role models, then or now, is dubious. As Mariamne expresses the texts' recurrent  216  lesson, a woman's glory lies in knowing how to suffer  (Hdrode et Mariamne V, iii). To sum up our observations on the ideological system or systems at work in these plays, we must first conclude that we found little which seriously challenged social structures or mores of eighteenth-century Prance except the reiterated argument for State control over the Church. In this last regard, the tragic texts show strong links with Voltaire's work as a whole, and propose a fundamental social reform [35]. However, the monarchy, with certain improvements, remains the ideal form of government, given the ineptitude of senates and the inconstancy of the masses. Privilege, power, and even retribution are attributed according to a hierarchy based on birth and gender. These ideal states portrayed - stable, just and merciful - are in modern terms constitutional monarchies, where kings are limited only by universal laws of reason and equity, where duty to god and family is subordinated to one's duty as a citizen. Little if any criticism of the class structure can be detected. The themes of just rulership, of the restoration of order by legitimate male accession, of the inevitable decline of regimes based on violence or usurpation are taken up repeatedly and treated in similar ways within these plays. The prevailing metaphor for just leadership at all levels of society is the father, implying a hierarchy based first on gender.  217 Religion as well takes its place within the monarchical hierarchy, imposing duties that are necessary but subordinate to those imposed by the State and which must be within the bounds of reason and compassion. The substance of belief is itself less contentious than the proposed absolute subordination of the Church to the State, and bears considerable resemblance to that which would be acceptable in French society of the eighteenth century. True skepticism, doubt of the existence of a god rather than of His mercy or power, appears much less often in those plays written after the end of the Regency, as it became less socially acceptable and as Voltaire the deist resisted the rise of the atheist materialists [36]. This hesitation doubtless stems from the considerable powers of the Catholic Church in French society, demonstrated by the banning of  Mahomet and Les Guebres to give only examples from the plays under study. The Church was a target of persistent criticism, but it was also a formidable opponent to the free circulation of criticism, and the texts bear witness to the constraints imposed by that opposition in their print and performance histories. Therefore we conclude that these tragedies, although inherently ideological in content and function, operate within largely non-controversial patriarchal norms of social and familial structure. Indeed, it can be argued that such conformity was necessary to their popular reception, in view  218 of the power of censorship and the reluctance of audiences to accept drastic innovation. The "good father", a favourite theme of Voltaire's, is at best a benevolent despot in a hierarchical social system where not even freedom of belief is consistently extended to women, let alone freedom of action or choice. They are objects of male desire or victims of man-made laws, consistently rendered passive in the plots. The clearest example of this passivity is Adelaide, who can be eliminated from the text withour changing the plot (in Les Frerps ennemis, 1751). This process of limitation leads inevitably to the resolution of "female" problems in death, the common denouement in Voltaire's tragedies from Jocaste's suicidal despair to Irene's immolation.  219 NOTES [1] A preliminary and incomplete version of the research underlying this chapter was presented as a paper at the 1991 conference of the Canadian Society for Eighteenth-Century Studies at the University of Calgary, and is being published in volume 12 of the journal of the Society, Man and Nature, under the title "Voltaire et «Maman>>: Female Sexuality, Maternity and Incest in the Tragedies." [2] Michael Cartwright, introduction to Adelaide du  Guesclin, Complete Works of Voltaire (Oxford UP, 1985) volume 10 and Jack R. Vrooman and Janet Godden, introduction to Merope, Complete Works of Voltaire (Oxford UP, 1991) volume 17, pp. 91 - 387. Thomas J. Carr Jr. (op. cit., 1975) associates certain plot structures with ideological content, but bases his argument on three overtly ideological plays: Brutup, Alzire, Mahomet. As in previous chapters, all reference to Voltaire's works will be to the Moland edition, volume and page number, except where the Oxford edition is indicated. All references to plays in the Oxford edition after the first will be made in text and include the editor's name. All references to the Correspondence will be to the Besterman edition (Oxford, 1968), and will include the Besterman number and the date, with the recipient or writer's name when relevant.  220 [3] On pp. 265 - 67; see also Jean Duvignaud, La  Sociologie du thdatre (Paris: PUF, 1965), p. 36. [4] David H. Jory, "The Role of Greek Tragedy in the Search for Legitimate Authority under the Ancien Rógime," in  Eighteenth - Century Fren0 Theatre: Studies Presented to E. J. H. Greene, eds. Magdy Gabriel Badir and David J. Langdon (University of Alberta Department of Romance Languages and Comparative Literature, 1986), p. 6. [5] Jean-Marie Apostolides, Le Prince sacrifid:  Théâtre et politique au temps de Louis XIV (Paris: Minuit, 1985), p. 28-29. [6] Ubersfeld, p. 283. [7] Ubersfeld, pp. 287 and 296-97. [8] As H. C. Lancaster observes in French Tragedy in  the time of Louis XV and Voltaire 1715 - 1774 (Baltimore: Johns Hopkins UP, 1950), I: 216 and 612.  [9] Zaire, ed. Eva Jacobs, in Complete Works of Voltaire (Oxford UP, 1988), volume 7, pp. 273-569: I, i, 101-27. [10] Jean Sareil, Voltaire et les grands (Genéve: Droz, 1978), p. 144. [11] English Showalter Jr., "Writing Off the Stage: Women Authors and Eighteenth-Century Theater," Yale French  Studies 75 (1988), p. 103. See also Barbara G. Mittman, "Women and the Theater Arts," in French Women and the Age of  Enlightenment, ed. Samia I. Spencer (Bloomington:  221 Indiana UP, 1984), pp. 155 - 169, whose more positive conclusion Showalter Jr. challenges.  [12] Alzire, ed. T. E. D. Braun, in the Complete Works of Voltaire (Oxford: Voltaire Foundation, 1989), volume 14: I, ii, 177. [13] "Fathers good and bad in Voltaire's Mahomet," in  Myth and its Making in the French Theatre (Cambridge UP, 1988), pp. 121 - 135. [14] Act V, scene ii; II: 523. This passage occurs in the 1779 Eriphile and in its manuscript variants; the term "'Dere" is not in the established text. I follow Robert Niklaus in adopting Voltaire's own spelling of the title and character as Eriphile, as argued in "Eriphile de Voltaire et le theatre d'Eschyle," Le Siècle de Voltaire: Hommage  Rend Pommeau, eds. Christiane Mervaud and Sylvain Menant (Oxford UP, 1987), II: 707 - 718. [15] Rend Pomeau, La Religion de Voltaire (Paris: Nizet, 1956), pp. 7 - 10. [16] Pomeau, pp. 12 - 14; see also Peter Gay,  Voltaire's Politics (Princeton UP, 1959), P. 249; William H. Trapnell, Christ and his 'Associates' in Voltairian  Polemic (Stanford French and Italian Studies XXVI, 1982), pp. 214 - 15. [17] Pomeau, pp. 265 and 456. [18] Rend Vaillot, Avec Mme du Chatelet, volume 2 in  222 the series Voltaire en son temps edited by René Pomeau (Oxford UP, 1988), p. 156. [19] Mason (1988), p. 133. [20] Gay, op. cit., p. 259; see also Pomeau, p. 400. [21] Pomeau, op. cit., P. 44. [22] Magdy Gabriel Badir, "L'Anatomie d'un coup d'Etat ou la prise du pouvoir dans la tragedie Mahomet de Voltaire," in Eighteenth - Century French Theatre: Studies Presented to E. J. H. Greene, eds. Magdy Gabriel Badir and David J. Langdon, pp. 102 - 104. [23] eds. Vrooman and Gooden; I, i, 133-34 and IV, i, 42. [24] Alison Scott-Prelorentzos, "Religious Tolerance in Eighteenth-Century Drama: Three Examples" in Eighteenth Century French Theatre: Studies Presented to E. J. H. Greene, eds. Magdy Gabriel Badir and David J. Langdon, p. 89. [25] Scott-Prelorentzos, p. 87. [26] Mason (1988), p. 121. [27] Lucette Desvignes, "Le Théâtre de Voltaire et la femme victime," in volume XLIV (1977): 535 - 51. See also Madeleine Rousseau Raaphorst, "Voltaire et feminisme: un examen du theatre et des contes," SVEC LXXXIX (1972): 1325 - 1335 and for a more recent survey of the subject, see Gloria M. Russo, "Voltaire and Women" in French Women and the Age of Enlightenment, ed. Samia I. Spencer (1984), pp.  223 285 - 295. [28] Maite Albistur and Daniel Armogathe, Histoire du feminisme frangais du moyen age a nos jours (Paris: Editions de femmes, 1978), pp. 191-92. [29] Desvignes, pp. 550-51. [30] See Mason (1988) and Jacques Spica, "Le Fils substitue ou les Menechmes de Voltaire," in Le Siècle de Voltairp: Homage a Rend Pomeau, eds. Christiane Mervaud and Sylvain Menant (Oxford UP, 1987), II: 867 - 880. [31] Leon Abensour, La Femme et le feminisme avant la revolution (Paris: Ernest Leroux, 1928), p. 23. See also Adrienne Rogers, "Women and the Law" in French Women and the Age of Enlightenment, ed. Samia I. Spencer (1984), pp. 33 48. [32] Walter J. Rex, "On Voltaire's Merope," in The Attraction of the Contrary (Cambridge UP, 1987), p. 92. [33] Desvignes, pp. 550-51. [34] Charles Cone, Journal et memoires, ed. Honoré Bonhomme (Paris: Firmin Didot, 1868), III: 236. [35] Pomeau, p. 306 and Gay, op. cit., pp. 32, 309 and 337. [36] Pomeau, p. 387.  224 Conclusion The examination of linguistic changes undertaken in chapter I demonstrated that the pursuit of "noble" language entailed for the playwright a conception of theatre as literature which co-existed in perpetual conflict with the demands of theatre as performance. Tragedy as a genre continued to afford the same prestige to its successful exponents as it had in the seventeenth century, but it also continued to operate under largely the same set of literary rules and audience expectations. Such innovation in language and versification as has been noted was limited in scope, and even this was hotly contested by audiences and critics of the time. The motivations for linguistic change were found to be various. The purity demanded by the "noble" ideal played an important role in textual revisions, but alterations were also provoked by the desire for psychological plausibility in the characters or their actions, by changes in character conceptualization due to reader reaction or to casting, and finally by exterior forces such as critics or censors. It is significant that often the reassertion of authorial control appears to have been the inspiration for the alteration, insofar as it eliminated or concealed a performance change to the text, as in the late revisions to Zulime and AddlaIde.  225 Considerable evidence was found to suggest Voltaire's ambivalence towards outside criticism of the text, in his asking for assessments from his friends even as he largely dismissed the contributions of the actors, and for the struggle within the text between the collaborations between author and reader, actor and spectator on the one hand and sole authorial control on the other, particularly as the text moved from the domain of performance into that of printed literature. Those most capable of affecting the text at this level were a group of male friends and teachers. Women and performers (of both genders) were by their lack of classical education largely barred from direct impact on the tragic text, as they were considered incapable of manipulating such a demanding literary genre. Tragedy, by its adherence to a "noble" form, remained a medium capable of reflecting considerable prestige on the author, but to which access was restricted by gender and class because of the limited access to classically based education. The limited degree to which performance and performers were allowed to imfluence the language of these texts, i.e. to make choices about grammar, syntax and vocabulary, is perpetuated in the struggle for textual control in the domain of performance per se. Performance was clearly essential to the theatrical praxis underlying these texts; a successful run at the  226  Comedie was the goal and the proving ground for each play, as well as a source of prestige for Voltaire. The casting and the quality of the performance were key to the development and reception of the play. Voltaire participated as fully as he could in the choice and preparation of actors. Extensive changes in text and even in the basic conceptualization of a character were made to suit a particular casting, such as that of Mlle Clairon in Ttome sauvee. Successful performance increased the circulation of the printed text. Voltaire deplored the prevalence of versions transcribed from performance rather than produced under his supervision, as much for the loss of control as for the loss of revenue. Audience reaction and expectations could affect the form of a play in the expansion to five acts of  ahomet, or the  characterization in the way Mariamne was made more virtuous and Aurelie more strong, or the action in the removal of Mariamne's onstage poisoning from Herode et  Mariamne(1725) and of the cannon shot from Adelaide (1752), or the staging in the use of the tomb and the ghost in Semiramis to the pyre in Olympie. But the pursuit of authorial control over the text continued to operate powerfully on the text during its performance history. Audience judgment in conflict with Voltaire's own literary judgment was deprecated and usually ignored. We saw considerable resistance to the  227 incorporation of performance edits and revisions into the printed text, i.e. in Zulime, Adelaide and Olympie. There was little recognition from Voltaire or from literary critics that the 1765 revival of Adelaide owed not only its popularity but also much of its textual form to Lekain, who compiled the performance text without consulting Voltaire. ^However, this drive for textual control made Voltaire not merely a playwright but truly a director of his plays in the modern sense of the word. He supervised rehearsals, coached performers in person or by mail, advised on costume and decor. He went further than supervision by elaborately staging his own performances whether at Cirey, at Potsdam or at les Delices, in which he himself acted. This new interventionist style of playwriting was acknowledged as a major contribution to French theatre. These plays made a tremendous change in contemporary standards of stage presentation: increased realism in costume and decor, more exotic locales, and more elaborate effects (ghosts, trapdoors, pyres) which made full use of the clear stage available after 1759. The impact on acting style must also be considered. Both Lekain and Clairon, the leading Comedie francaise performers from the 1750's to the Revolution, built much of their reputations on their roles in Voltaire's plays, and acknowledged his influence on their acting styles. When we move from the domain of performance to the  228 wider implications of the texts as reflections and models of a society, the ideological systems operating within these texts elaborate the gender and class hierarchy previously observed, and demonstrate a relative consistency, particularly in view of the great variations in their textual expression and the length of time covered by their textual and performance histories. The plays' social systems are dominated by the recurrent association between authority and paternal status; in government as in the family, power resides in the hands of one enlightened, reasonable, benevolent but self-determining male figure. The legitimacy of power is based on the legitimacy of (male) birth and the maintenance of a hierarchical class structure. Attacks are made on the parasitical court rather than on the aristocracy Der se, and would be welcomed by a Paris audience habitually differing in its judgments on theatre as on other issues from those of Fontainebleau and Versailles. While female rulers are portrayed in the plays as essentially incapable, male rulers only become unjust when they cease to govern by reason. Order in both cases is restored by another, more enlightened, male ruler. Republican government is painted as ineffective, theocratic government as inescapably unjust. Religion is within the social hierarchy systematically subordinated to the authority of the paternal ruler. The texts over time become increasingly favourable to belief as such in their portrayals of  229  apolitical tolerant priests and reasonable, law-abiding believers, in contrast with the cynicism and skepticism of the earlies texts. But any church above or even independent of the state is unacceptable within these texts, as is any individual faith not limited by reason and the duties of citizenship. Religion is restricted to its social function of promoting national unity and socially beneficial behaviour. In the societies depicted by these texts, women are largely excluded from freedom of belief and autonomy of action within the family as they are from the effective exercise of political power. Belief and behaviour are or should be controlled by the father. There are indications of a differential justice at work in the punishments meted out, based on the gender of the victim and the offender. The child's primary duty is to the father, not the mother, and the wife's is to the husband. A man's failure to return reasonable treatment for obedience does not entitle the child or wife to retaliate in any form, not even such glaring injustices as Agamemnon's sacrifice of Ipheganie or Hdrode's murder of Mariamne's father. Nor does female virtue bring a consistent reward. The prevailing fate of women in the societies within these plays is suffering leading to death. In all these texts there continues the dynamic struggle between control and collaboration, between one voice and many, between a static text and one open to  230  perpetual transformation. This tension of opposing forces is essential to Voltaire's theatrical praxis. Beneath the surface of each text we find an underlying network of outside influences and authorial responses, imperfectly incorporated or concealed or even suppressed in the effort to create an impression of sole authorship and authority. However, the ideological system within the texts is in broad terms coherent, held together particularly by the recurrence of father symbolism. The extent to which class and gender affect one's access to the tragic text, one's ability to have opinions or decisions reflected in print rather than merely in the evanescence of performance, is a part of this dominant ideology based on a patrilineal hierarchy. The rigidity Of this social structure within the text stands in sharp contrast to the freedom of performance and performers, perpetually and infuriatingly escaping from Voltaire's control. Voltaire's choice of tragedy itself as his preferred theatrical genre, in the way it was defined in eighteenth-century France, has in itself ideological implications. Tragedy was an essentially privileged medium, difficult of access, highly rule-bound and tied by its very premises (the requirement for "noble" characters and situations) to a hierarchical social system. The variations between the social structures within these  231 plays and those of the society in which they were produced can be grouped almost completely under the heading of improvements to the status quo. The models in the texts encourage the incorporation of reason, equity and tolerance into the established figures of authority: the king, the priest and the father; they do not present a radical rethinking of that familiar hierarchy, except in their insistence on the subordination of Church to State. In this regard they dared to question the accepted order, and provoked strong censorship, from Mahomet to Les Gu6bres and Les Lois de Minos. This paternalism and conservatism are not, however, surprising in these texts. The intended performance audience was a relatively broad spectrum of French society in Paris and the provinces, to which can be added the large readership within and beyond France's borders. These plays also had to pass an alert and rarely well-disposed system of censors. Goodlad's study of popular drama in the twentieth century came to the same conclusion, that conservatism was necessary to popular success [1]. Innovation at all levels of tragic construction was a difficult and slow process, whether it was the expansion of accepted tragic vocabulary, or the increase in realism and spectacle on stage, or the presentation of a monarchy based on reason, tolerance and fairness rather than on prejudice and patronage. While Gay's conclusion that "in his age and in his regime,  232 Voltaire was a radical," [2] may not apply to his tragedies considered separately, the plays do present societies more just and more compassionate than eighteenth-century France could be.^The plays do establish, however, significant limitations to what are considered the basic tenets of Voltairean philosophy: "toleration, the rule of law, freedom of opinion" [3] in that these fundamental rights do not extend consistently to women, and for women remain subordinate to the always legitimate authority of fathers and husbands. However disappointed the modern reader might be in these texts, however rigid and biased these idealized societies now seem, one need only contrast the decay of the ancien  rdgime into the Revolution with the military and mercantile expansion of Britain (with its greater religious tolerance and its constitutional monarchy) during the same period to realize how considerable an alteration of history might have been realized if France had become a society like those envisioned in these plays.  233 NOTES  [1] J. S. R. Goodlad, A Sociology of Popular Drama (London: Heinemann, 1971), p. 28. [2] Gay, p. 337. [3] Gay, p. 32.  234 BIBLIOGRAPHY PRIMARY SOURCES Voltaire, Francois Marie Arouet de. Oeuvrescomplbtes, ed. Louis Moland, volumes 2, 3, 4, 5, 6 and 7 containing his theatre. Paris: Gamier frbres, 1877. ***********************************• The Complete Works of Voltaire. General ed. W. H. Barber. Oxford: The Voltaire Foundation. 1731 - 1732 Ed. Eva Jacobs, 8 (1988): pp. 273 526. Adelaide du Guesclin. Ed. Michael Cartwright. 10 (1985). Alzire, ou les Americains. Ed. T. E. D. Braun. 14 (1989): 1 - 210. Merope, pp. 91 - 388. Eds. Jack R. Vrooman and Janet Godden. 17 (1991): 91 - 388. **********************************. Correspondence. Ed. Theodore Besterman. Oxford: The Voltaire Foundation, 1968. WORKS CONSULTED - BEFORE 1800 L'Annee litteraire. Ed. Elie Catherine Freron. Amsterdam [Paris]: Lambert, 1754 - 1776. 183 vols. Irregular publication. Continued after Freron's death by Freron fils, the abbe Grosier and J.-L. Geoffroy to 1790. d'Aubignac, Le Pratique du theatre, ed. P. Martino. Paris: Carbonel, 1927.  235 Bachaumont. Louis Petit de. Mdmoires secrets pour servir l'histoire de la republique des lettres en France depuis MDCCLXII jusqu'a nos jours. 36 vols. London: John Adamson, 1777. Beaumarchais, Pierre-Augustin Caron de. Essai sur le genre dramatique serieux in Oeuvres. Ed. Pierre Larthomas with Jacqueline Larthomas. Paris: Gallimard, 1988. Boileau-Despreaux, Nicolas. Art poetique in Oeuvres. Paris: Gamier freres, 1952. Clement, Jean Marie Bernard and Joseph de la Porte. Anecdotes dramatiques. Geneve: Slatkine, 1971; reproduction of the Paris 1775 edition. Clement, Pierre. Les Cinq annees litteraires. Geneve: Slatkine, 1967. Reproduction of the Berlin 1755 edition. Colle, Charles. Journal et mdmoires. Ed. Honoré Bonhomme. 3 vols. Paris: Firmin Didot, 1868. Corneille, Pierre. Théâtre complet. 2 vols.^Paris: Gallimard, 1957. Correspondance littOraire, philosophique et critique par Grimm, Diderot, Raynal, Meister et al, including the Nouvelles litteraires. Ed. Maurice Tourneux. 16 vols. Paris: Gamier freres, 1877 - 82. Diderot, Denis. Oeuvres. Paris: Editions de la Pleiade, 1951.  236  LeMercure de France. Eds. Fuzelier, Raynal, Marmontel et al. Paris: 1672 - 1820. Racine, Jean Baptiste. Oeuvres completes. 2 vols. Paris: Gallimard, 1950, 1952.  WORKS CONSULTED - POST 1800  Apostolides, Jean-Marie. Le Prince sacrifid: Théâtre et  Dolitique au temps de Louis XIV. Paris: Minuit, 1985. Bachum, Albert. Censorship in France from 1715 - 1750:  Voltaire's opposition. New York: Lenox Hill, 1971. Conlon, Pierre M. Voltaire's literary career from 1728 to  1750.^SVEC (1961) XIV. Deloffre, Frederic. Marivaux et le marivaudage: une  próciositá nouvelle, 2e ed. Paris: Armand Colin, 1967. Descotes, Maurice. Le Public du thdatre et son histoire. Paris: PUF, 1963. Fontaine, Leon. Le Theátre et la philosophie au XVIIIe  siècle.^Versailles: Cerf, 1879. Gaiffe, Felix. Le Drame en France au XVIIIe siècle. Paris, 1910. Gay, Peter. Voltaire's Politics. Princeton UP: 1959. Gooden, Angela. Actio and Persuasion: Dramatic Performance jjL Eighteenth-Century France. Oxford, 1986.  237 Goulemot, J-M et M. Launay. Le Siècle des lumieres. ^Paris: Editions du seuil, 1968. Lagrave, Henri. Le Theatre et le Public a Paris de 1715 1750. Paris, 1972. La Harpe, Jean Francois de. Oeuvres. Geneve: Slatkine, 1968. Reproduction of the 1820 edition. Lancaster, H. C. French Tragedy in the time of Louis XV and Voltaire 1715 - 1774. Baltimore, 1950. Lion, Henri. Les Tragedies et les theories dramatiques de Voltaire. Geneve, Slatkine, 1970; reproduction of the Paris 1895 edition. Lough, John. Paris Theatre Audiences in the Seventeenth and Eighteenth Centuries. Oxford UP, 1957. Monod-Cassidy, Helene. Un Voyageur-philosophe au XVIIIe siècle: ^l'abbe Jean-Bernard Le Blanc. (Cambridge: Harvard UP,^1941). Niklaus, Robert. A Literary history of France - the 18th century. New York: Barnes & Noble, 1970. Parker, David. The Making of French Absolutism. London: Edward Arnold, 1983. Pomeau, René. La Religion de Voltaire. Paris: Nizet, 1956. Rex, Walter J. The Attraction of the Contrary. Cambridge UP, 1987. Ridgway, Ronald S. La propagande philosophique dans les tragedies de Voltaire. Studies in Voltaire and  238  the Eighteenth Century (1961) XV. Russell, Trusten Wheeler. Voltaire, Dryden and Heroic  Tragedy. New York: AMS, 1966. Sareil, Jean. Voltaire et les grands. Gen6ve: Droz, 1978. Scherer, J. La Dramaturgie classique. Paris: Nizet, [1950]. Torrey, Norman L.^Voltaire and the English Deists. New Haven,^1929. Trapnell, William H. Christ and his 'Associates' in  Voltairian Polemic. Stanford French and Italian Studies XXVI, 1982. Vaillot, René. Avec Mme du Chatelet. Voltaire en son temps 2.^Ed. René Pomeau. Oxford UP, 1988. Vier, J. Histoire de la litterature frangaise du XVIIIe  siècle,^2 vols. Paris: Armand Colin, 1970. Vrooman, Jack Rochford. Voltaire's theatre: the cycle from  "Oedipe" to "Merope". Studies in Voltaire and the Eighteenth Century (1970) LXXV. Wade, Ira 0. Voltaire and Mme du Chatelet. New York: Octagon, 1967. Wellington, Marie. The Art of Voltaire's theatre:  Exploration of Possibility. New York, Bern, Frankfurt, Paris: Lang, 1987. Willens, Lillian. Voltaire's comic theatre: composition,  conflict and critics. Studies in Voltaire and the Eighteenth Century (1973) CXXXVI.  239 ARTICLES AND DISSERTATIONS CONSULTED Abrate, Jayne Halsne. Play and Playscript: Dramatic Structure in Six Voltairean Tragedies. PhD thesis, Purdue, 1983. Badir, Magdy Gabriel. "L'Anatomie d'un coup d'Etat ou la prise du pouvoir dans la tragódie Mahomet de Voltaire," in Eighteenth- Century French Theatre: Studies Presented to E. J. H. Greene, pp. 99 - 106. Eds. Magdy Gabriel Badir and David J. Langdon. University of Alberta Department of Romance Languages and Comparative Literature, 1986. Berubd, Georges. Le Personnage, instrument d'analyse du theatre de Voltaire. PhD thesis, Laval, 1983.  ***************. "Voltaire et La Mort de Cesar" Man and Nature,^eds Roger L. Emerson, Gilles Girard, Roseann Runte, VI: 15 - 20. Carr, Jr., Thomas M. A Structural Study of Voltaire's Theatre of Ideas. PhD thesis, Wisconsin, 1972.  ********************.  "Dramatic structure and philosophy," in  Studies in Voltaire and the Eighteenth Century^CXLIII (1975): 7-48. Desvignes, Lucette. "Le theatre de Voltaire et la femme victime," Revue des sciences humaines XLII (1977) 537-51. Jory, David H. "The Role of Greek Tragedy in the Search for  ^  240 Legitimate Authority under the Ancien Rdgime," in Eighteenth- Century French Theatre: Studies Presented to E. J. H. Greene, pp. 6 - 16. Eds. Magdy Gabriel Badir and David J. Langdon. University of Alberta ^Department of Romance Languages and Comparative Literature, 1986. Lagrave, Henri. "Voltaire, Diderot et le 'tableau scdnique': le malentendu de Tancréde," in Le Siècle  de  Voltaire: Hommage a ^Rend Pomeau, II: 569 - 575.^Eds. Christiane Mervaud and Sylvain Menant. Oxford: Voltaire Foundation, 1987. Mason, Haydn. "Diderot critic of Voltaire's theatre," in Le Siècle de Voltaire: Homage a^Rend Pomeau, II: 633 ^ IMP  642.^Eds. Christiane Mervaud and Sylvain Menant. Oxford: Voltaire Foundation, 1987.  * ************. "Fathers good and bad in Voltaire's Mahomet," in Myth and its Making in the French Theatre: Studies Presented to W. jD. Howarth, pp. 121 - 35. Eds. E. Freeman, H. Mason, M. O'Regan and S. W. Taylor. Cambridge UP, 1988. Niderst, Alain. "Tragique Voltaire" in Le Siècle de Voltaire: Hommage a Rend Pomeau, II: 701 - 706. Eds. Christiane Mervaud and Sylvain Menant. Oxford: Voltaire Foundation, 1987. Niklaus, Robert. "Eriphile de Voltaire et le thdatre d'Eschyle"^in Le Sidcle de Voltaire: Homage a  241 René Pomeau, II:^707 - 718. Eds. Christiane Mervaud and Sylvain Menant. Oxford: Voltaire Foundation, 1987. ***************. "Eriphyle et Semiramis" in Essays on the Enlightment in honour of Ira O. Wade, pp. 247-54. Geneve:^Droz, 1977.  ****** ********* . "La Propagande philosophique au theatre au siècle des lumiéres," Studies in Voltaire and the Eighteenth Century XXVI (1963): 1219-60. Raaphorst, Madeleine Rousseau. "Voltaire et feminisme: un examen du theatre et des contes," Studies in Voltaire and the Eighteenth Century LXXXIX (1972): 1325 - 1335. Sanderson, Anne. "Voltaire and the problem of dramatic structure: the evolution of the form of Eriphyle" and "In the playwright's workshop: Voltaire's corrections to Irene," Studies in Voltaire and the Eighteenth Century CCXXVIII (1984): 97-170. Sareil, Jean. "Les quatres premieres Lettres philosophiques,"^Romanic Review LXXVI (1985): 277-286. Scott-Prelorentzos, Alison. "Religious Tolerance in Eighteenth-Century Drama: Three Examples," in Eighteenth- Century French Theatre: Studies Presented to E. J. H. Greene, pp. 85 - 98. Eds. Magdy Gabriel Badir and David J. Langdon. University of Alberta Department of Romance Languages and Comparative  242 Literature, 1986. Showalter Jr., English. "Writing Off the Stage: Women Authors and Eighteenth-Century Theater," Yale French  Studies 75 (1988): 95 - 111. Spica, Jacques. "Le Fils substitue ou les Menechmes de Voltaire," in Le Siècle de Voltaire: Homage a.  René Pomeau, II: 867 - 880. Eds. Christiane Mervaud and Sylvain Menant. Oxford: Voltaire Foundation, 1987.  MODERN CRITICAL WORKS Burton, Deirdre. Dialogue and Discourse: a sociolinguistic  approach to modern drama dialogue and naturally occurring conversation. London: Routledge & Kegan Paul, 1980. Demarcy, Richard. Elements d'une sociologie du spectacle. Paris, 1973. Duvignaud, Jean. La Sociologie du thdatre. Paris: PUF, 1965. Elam, Keir. The Semiotics of Theatre and Drama. London and New York: Methuen, 1980. Goodlad, J. S. R. A Sociology of Popular Drama. London: Heinemann, 1971. Helbo, André, ed. Sftiologie de la representation. PUF, 1975. Larthomas, Pierre. Le Langage dramatique. Paris: Armand Colin, 1972.  243 Morgan, Robin. The Demon Lover: On the Sexuality of Terrorism.^New York, London: W. W. Norton, 1989. Pavis, Patrice. Languages of the Stage: Essays in the Semiology of the Theater. New York: Performing Arts Journal Publications, 1982. Penelope, Julia. Speaking Freely: Unlearning the Lies of the Fathers' Tongues. In the Athene series. New York: Pergamon Press, 1990. Reboul, Anne and Jacques Moeschler. Discours theatral et analyse conversationnelle. Cahiers de linguistique frangaise #6, 1985. Searle, J. R. Speech Acts. Cambridge UP, 1969. Sherman, Carol. Reading Voltaire's contes: a semiotics of philosophical narration. North Carolina UP, 1985. Spencer, Samia I., ed. French Women and the Age of Enlightenment. Bloomington: Indiana UP, 1984 Ubersfeld, Anne. lire le theatre. Paris: Editions sociales, 1978. ***************. Le Roi et le bouffon. Paris: Corti, 1974. OTHER WORKS CITED Atwood, Margaret. The Handmaid's Tale. Toronto: Seal Books,^1986. Dezon-Jones, Elyane, ed. Ecritures feminines. Paris: Editions Magnard, 1983.  244 APPENDIX  Chronological Tables of Textual Changes in the Plays under Examination References in the Correspondence to revisions without specific textual location (at a minimum, act and scene) do not appear here. Revisions for which exact texts are not available are not included in the totals for variant lines. ^When the revision involves a reduction in the number of lines, the lines changed column will have a higher number than the total lines. Revisions affecting the length of a specific règligue will when relevant be indicated by the original lenghth of the passaged given in parentheses next to the total length of the scene. Reference to editions are to the Noland edition and its cited sources unless otherwise indicated. Reference to correspondence is to the Oxford edition (1968) edited by Theodore Besterman and will be made by Besterman number. The gender of a character will be indicated only once.^Strictly grammatical changes, such as the order of pronouns, the tense of verbs, the substitution of synonyms, and the -ois/-ais shift, are not noted in the tables. Abbreviations: corresp = Correspondence de Voltaire, ed. Besterman (Oxford, 1968) perf =^performance (at the Comedie frangaise unless indicated) m =^male f =^female det =^detached, referring to lines which cannot be placed exactly No table is given for Artemire, as no complete text exists, only the role of Artemire herself.  Herode et Mariamne, debut 6 March 1724, revised and restaged 10 April 1725 and 7 September 1763  date  ^  location^source^character^content^function lines changed total lines ^(gender)  Oct 1723 II,^ iii-iv  corresp^Mariamne (f)^feelings^characterD 171^ for Varus^ization  Mar 1724- III,^ perf^Nerode(m),^politics^subplot Apr^25 iii-iv^ Varus(m),^ reduced 228 lines^ Mazael(m) V, iv^pert^Mariamne^death on^loss of stage^action The text has been lost; the recit which replaces it has 31 lines. 1757^V, i^ edition^Mariamne^grief^subplot 4.5 lines^ reduced 1757^III,i  ^  edition^Narbas(m)^rephrasing^style  During July and August 1762, the character of Varus, a Roman consul, was replaced by Sohême, Asmonean prince. This change required the the following sections of text to be replaced, all of which can be found in the Moland edition notes:  245 Wade et Nariamne ^ ^ I, i edition^Salome(f) Herode in^exposition ^ ^ 142 lines Mazael Rome I, ii^ 22 lines^  Varus^Varus'arrival^exposition Mazael  I, iii^ 91 lines^  Varus^feelings^exposition Albin(m)^for Mariamne  II, i^ 63 lines^  Salome^Varus' feelings exposition Mazael  II, ii^ Salome^references to^exposition 6 lines 49 lines^Mazael^Varus Mariamne II, iii^ Mariamne^Mariamne's^exposition 14 lines 24.5 lines^Elise(f)^feelings for Varus Narbas II, iv^ Nariamne^Mariamne's^exposition 5 lines 56 lines^Elise^feelings for Varus II, v^ Mariamne^references to^subplot 14 lines 110 lines^Elise^Rome^change Varus III,iii^ Varus^Mariamne's^fore15 lines 44 lines^Idamas (s)^danger^shadowing III, iv -v^ Hêrode^references^subplot 13 lines 121 lines^Mazael^to Varus^change new scene division^Idamas^references to^subplot lltrode^Varus^change vi^ 3 lines 87 lines^Salome IV, i^ Salome^references to^subplot 4 lines 30 lines^Mazael^Varus^change IV, ii^ flêrode^references to subplot 18 lines 68 lines^Salome^Rome^change Mazael V, v^  Htrode^Varus' death^subplot Idamas^ change  Variant lines: 680^Final text: 1307 lines  Eriphile debut 7 March 1732  The text as performed involved a high priest, which role has survived in the 1779 version based on a manuscript owned by Lekain, and offering certain variations in another manuscript. The Noland text is based on the subsequent revision removing the high priest, see the letter to Formant, 25 June 1732, D 497. In anticipation of the resolution of these textual problems in the forthcoming new edition by Robert Niklaus, we will, in the absence of clear dating for the revisions, use the Noland version as our final text, and cite the 1779 and its variant as earlier, abbreviated as 1779 and 1779v. date  ^  location^source^character^content^function lines changed total lines^(gender)  As stated above, Voltaire deleted the high priest soon after the play's debut. This change entailed the following changes, first from the 1779 Lekain text and its variants. 1779^I, i^edition^ThAandre(m)^politics^exposition 86 lost^96.5^ high priest (m) I, ii^edition^Theandre^remorse^character 38 lost^13.5^ high priest Eriphile (f) I, iii^edition^Eriphile^remarriage^plot 18 lost^93 I, iv^edition^Eriphile^remarriage^plot 12 lost^38.5^ Polemon (m) 8 added^38.5^edition^  character of plot Hermogide  II,i^edition^Theandre^AlcmAcin's^exposition Alcmeon (m) 38^76.5^ II, ii 1.5^5.5^edition^PolAmon^stage direction removed II, iii^edition^Eriphile^rephrasing^style 2^66  III,i^edition^Hermogide (m) ambition^exposition Euphorbe (0 28^76.5^ (moved to I, i in place of exposition by high priest) IV,i^edition^Alcseon^threat from plot 56.5^71^ Theandre^Hermogide IV, iii^edition^Alcmeon^staging of 3^9^ Eriphile^ghost IV, iv^edition^Alcmeon^recognition^plot 6^42.5^ Eriphile (this exchange alone expanded from to 22 lines)  247 Eriphile  I V, v  33  ^  ^  edition^Alcmton^foreshadowing removal of high priest^Eriphile's^character 105^ death  Act V in the 1779 text is organized so differently that no comparisions within a given are possible. We will furnish scene lengths, characters and content for the 1779 text and the Noland text to show the extent of the variation V,^i 1779  Noland  Alcaton Th6andre Poltson  duel proposed with Hermogide  Eriphile Zelonide ffl  civil war  Alcston Thtandre Polémon  duel accepted  29  Eriphile Mlonide Theandre  recit of Alaton's proposal  62  41  V,^i 47  1779  Moland  1779  V,^ii  V,^ii  V,^iii 38  Moland  V,^iii 5  1779  V,^iv 12  Noland  V,^iv  Alcorton Hermogide  duel  Eriphile AlcOon Hermogide Thkandre  Alcmton and Hermogide enter the temple  high priest Thtandre  Alcmton stabs Eriphile offstage  Thtandre  anticipation of disaster  Alcmton  remorse over Eriphile's stabbing  Théandre Polemon  recit of Eriphile's death  15.5 1779  V, v 17  Noland  V,^v 16  249 Eriphile 1779  ^ 28  ^  Noland^V, vi - vii^ 54^  Alcmton^Eriphile's dying high priest^forgiveness Eriphile Theandre Polèmon  AlcmOon^Hermogide's capture and Hermogide^Eriphile's dying Théandre^forgiveness Polèmon Eriphile ZOlonide  Act total: 165 in 1779, 199 in Noland The term "manuscript' indicates a manuscript source cited in the Noland edition. date^location^source^character^content^function lines changed total lines ^(gender)  1779v^I,i^manuscript^ThAandre^expansion on^exposition high priest^marriage 26^(12) 86^ I, iii^manuscript^Eriphile^political^exposition ZAlonide^situation 111^94^ I, iv 16^38.5^manuscript^Eriphile^marriage^plot Polemon  II,iii^manuscript^Alcmeon^Hermogide's^exposition 4 (lost)^11.5 II, iv^manuscript^Eriphile^rephrasing^style 26^51^ IAlonide II, v^manuscript^Hermogide^rephrasing^style 4^87 II,vi^manuscript^Hermogide^rephrasing^style 6^18  III,i^manuscript^Hermogide^character of^plot 29^38^ Euphorbe^Alcmeon  249 Eriphile III,^ii 17^68  manuscript  Eriphile Hermogide  supposed death plot of her son  III,iii 4^32  manuscript  Eriphile  marriage to^plot  IV,v 19.5^105  manuscript  Alcmeon Eriphile  rephrasing and^style editing down  Act V - variants compared to the 1779 text; 1779v has the same scene organization. V,^i  41  manuscript  Alcmton Polhon  rephrasing  style  V,^iii 3.5 38  manuscript  Alcieon Hermogide  Amphiaraas' sword  plot  manuscript  Th6andre high priest  Eriphile's death  expansion  manuscript  Alcmeon  mistaken  plot  16  18 9  V,^iv V,^v  V,^vi 3.5  12 17 28  victim  manuscript  high priest  benediction  reduction  Total variant lines, 1779 and 1779v: 683.5^Total lines, Noland text: 1513  ^  250  Zaire, debut 13 august 1732. From the Oxford; Voltaire Foundation edition by Eva Jacobs, 1988. Letters attached to the dates of editions indicate specific editions and manuscripts as established by Jacobs' edition. As many of the changes in Zaire's text do not occur in strict chronology, they are ordered by their first occurrence, with those versions which maintain the variation listed in parentheses below. date  1732  ^  location^source^character^content^function lines changed total lines ^(gender)  ^  11  I, i^prompt copy^Zaire (f)^rephrasing^style 156^ Fatime(f)  ^  1732^I, ii^prompt^  stage direction added  1732^I, iii^prompt^  stage direction added  1732^I, iii - iv^prompt^  location of scene break changed I, iv^prompt^Orosmane (m)^rephrasing^style 4^47.5  1732^I, iv - v^prompt^  location of scene break  1732^I, v^prompt^  stage direction added  1732^II, i^prompt^Nerestan (m)^rephrasing^style 3^160^ Ch2tillon (m) 1732^II, iii^prompt^  stage directions added  1732^II, iv^prompt^Corasmin (m)^rephrasing^style 1^9 1732^III, i^prompt^  stage direction added  III, iv^prompt^Nerestan^rephrasing^style 5^(2)133.5^Zaire 1732,^.5^prompt^ rephrasing (1733, 1736) 1732^III, vi^prompt^  stage direction added  1732^III, vii^prompt^Orosmane^rephrasing^style 33^64^ and editing 1732^III, vii^prompt^Orosmane^expansion^style 4^64 (all subsequent)  251 Zaire 1732  IV,^i  prompt  Zaire  IV,^ii  1732  prompt  stage directions added stage direction added  IV,^iii  1732  omission^style  81  4  IV,^iii  prompt  10  1732  IV,^iv - v  1732  Orosmane  24 prompt prompt  IV,^v 5 5  stage direction added and scene division moved Orosmane  76.5 76.5  1732  IV,vi 13  faith in^style Zaire - expansion  rephrasing^style expansion^style  prompt  Orosmane Zaire  rephrasing^style  Meledor^(m)  name given to slave role  64.5  1732 (1736)  V  prompt  1732  V,ii  prompt  stage direction added  1732  V,^iii  prompt  stage direction added  1732  V,^iii 8  prompt  Fatime Zaire  rephrasing^style  Zaire  Francais for^ideology chretien  48.5  1732  V,^v .5^4  prompt  1732  V,^vi^- vii  prompt  stage directions added, scene division moved  1732  prompt  Orosmane  rephrasing^style  prompt  Orosmane  rephrasing^style  1  prompt  Orosmane  rephrasing^style  4  prompt  Orosmane  rephrasing^style  Orosmane  stage directions, scene division moved recognition^plot  1  1732  V,^vii V,^viii  (1733, 1736)^1 (1733)  5 32  prompt  V,^ix  1732 4  16  prompt  Nerestan  252 Zaire 3^(lost)  (1733)  2  prompt  1732  Fatime  Zaire's love for Orosmane  character development  Nerestan  rephrasing  style  V,^ix^- x  1733  scene division moved  l,i  edition  Zaire  omission  style  edition  Orosmane  rephrasing  style  I, iv  edition  Orosmane  change to ifrancais'  ideology  II,iii  edition  4  156 III,^vii  10  64  I736A (all subsequent) 1736A  1736A  III,^iv  III,^vi 1  I736A  III,vii  IV,iii  2 1 I736A  1738 (1740,^42) 1738 (1740 - 46)  style  edition  Zaire  rephrasing  style  edition  Orosmane  rephrasing  style  edition  Orosmane  rephrasing  style  edition  Corasmin  sympathy removed rephrasing  character development development  24 V.^viii 32  Orosmane V,^ix  1738  rephrasing  64  1 1736A  Nerestan  54  7 1736A  edition 6.5  1 1736A  stage direction added  edition  I,^i 1  156  1  156  III,^vii  64  stage directions added  edition  Fatime  rephrasing  style  edition  Zaire  rephrasing  style  edition  Orosmane  rephrasing  253 Zaire 1740  1  ,^  i  1  edition  1742  I,^i 4  I,^i 1 IV,^i  IV,^iii  1151 .5  I, i  edition  Fatime  rephrasing^style  edition  Zaire  rephrasing^style  edition  Orosmane  edition  Fatime  rephrasing^style  edition  Nerestan  change to^ideology 'frangais'  edition  Zaire  rephrasing^style  edition  Zaire  rephrasing^style  rephrasing^style  156  II,ii 1  44  I,^i 1  156 V,^iii  1  replacing^style lines removed  24  1  Kehl  Zaire  81  1  Kehl  edition  156  1751  1757  rephrasing^style  156  1751  1752  Fatime  156  48.5  Variant lines: 164.5 ^Total lines:^1650  ^ ^  254 Adelaide du Guesclin, debut 18 January 1734, revived under this title 9 September 1765 Les Freres ennemis, at Potsdam 1751; Amtlie ou le Dm de Foix, debut 17 August 1752 From the Oxford: Voltaire Foundation edition by Michael Cartwright (1985) For the purposes of this table, we confine ourselves to the performed texts, and omit the manuscript only version called Alamire, which was not printed until the Cartwright edition, in accordance with Cartwright's conclusion (introduction, p. 59) that there was no intention  that is should be published or performed."  We begin with the 1734 version as our base text, and group the variants by the revision ( as distinct from individual edition) to which they belong, i.e. Adelaide du euesclin or Le Duc de Foix. The abbreviations for sources are those of the Cartwright edition. Letters attached to the dates of sources indicate specific editions and manuscripts as established by Cartwright's edition. Variants within the 1734 text date  ^  unclear  location^source^character^content^function lines changed total lines^(gender) II,vii 2 (added) 114^manuscripts^Coucy (m)^love vs.^character (MS2-NS4)^honour^development 4^114^manuscripts^Vendame(m)^reproof^character (1152 - 1157)^ development  III,i^manuscripts^Nemours (m)^rephrasing^style 10^54^(1151 - 1157) ^4 (added) 54^manuscripts^Nemours^anxiety for^plot Adtlaide III, ii^manuscripts^Nemours^rephrasing^character 50^80^(NM - 1157)^Adtlaide (f)^development III,iii^manuscripts^VendOme^rephrasing^style 3.5^100^(MS1 - 11S7)  IV,iv^manuscripts^Vendame^rephrasing^style 4^8^(MS1 - 11S7) 4 (added^8^ expansion^emphasis IV, v^manuscripts^Vendame^rephrasing^style 7^77^(MS1 - 1157) IV, vii^manuscripts^Adelaide^expansion^emphasis 13 (added) 41^(MS1 - 11S7) 33^41^manuscripts^Adelaide^rephrasing^style (MS1 - 11S7)^Vencrame  Adelaide du 6uesclin  ^  IV, ix 4 (added) 26 V,v - vi  ^  255  manuscripts^Adelaide^Adtlaide's^staging (MS 1 - MS7)^Taise (f)^collapse manuscripts ^no scene division made (M52 - MS7)  Variant lines (within the 1734 text) 138.5 ^Total lines (1734): 1610  Les Freres ennemis (Potsdam, 1751) This is a thorough re-working, with Adelaide replaced by her brother on stage, and the five acts reduced to three. However, considerable portions of the text are taken from the 1734; we will indicate the number and location of the new lines. I,^i 39  89  Dangeste (m)^politics^exposition Coucy (e)  I,^ii 18  18  I,^iii 47.5  47.5  d'Alengon^appeal for^plot Coucy^truce  I,^iv 5.5  5.5  d'Alengon^English^plot Coucy^attack  II, i^ 84  d'Alengon^rtcit of^plot Coucy^battle  II,^ii  d'Alengon^despair^character  40  12  12 II,^iii  34  Nemours^rivalry with^plot Dangeste^d'Alengon  126  Nemours^recognition^plot Dangeste d'Alengon  II,^v 12  12  II,^vi 98  development  34 II,^iv  102  d'Alencon (m) proposed ^plot Dangeste^marriage Coucy  100  Nemours^renewed^plot Dangeste^attack d'Alencon Coucy Nemours^attempted^plot Coucy^reconciliation  Les Freres ennemis  ^  (Act^III drawn^from  Acts IV and V of  2,56  Adelaide)  III,^i  24  31  III,^ii  proposed^plot marriage  20  23  d'Alengon Nemours Dangeste  arrest^plot  III,^iii 50  80  d'Alengon Coucy  order^for^plot execution  III,^iv 16  d'Alengon 40.5  remorse^character  III,^v 5  22  d'Alengon an officer  execution^plot reported  III,^vi 8  30.5  d'Alengon Coucy  execution^plot denied  d'Alengon Coucy Dangeste Nemours  reconciliation plot  III,^vii 10  New lines:  Nemours Dangeste  470  42  Total^lines:  746  257 Amelie ou le duc de Foix,  Comedie frangaise debut 17 August 1752  This revision involves the change of names and setting to avoid the problems with audience reaction experienced by the first Adelaide.  date  ^  location^source^character^content^function new lines^total lines^(gender)  I, i 1752^70^144^edition^Amelie (f)^politics^exposition  Lisois (m) I, ii^edition^Asklie^rivalry^plot 58^65^ Taise^for her I, iii^edition^de Foix^love for^plot Arlene 30^30^ I, iv^edition^de Foix^politics^plot 86^86^ Lisois (Act II draws from act III of Adelaide) II,i^edition^de Foix^jealousy^character development 9^9^ II, ii^edition^de Foix^her refusal^plot 74^105^ Amelie II, iii^edition^de Foix^rage^character 14^14^ development II, iv^edition^de Foix^truce^plot 76^130^ Lisois^rejected II, v^edition^de Foix^attack^plot 10^10^Lisois officer (Act III draws from act II of Adelaide) III, i^edition^de Foix^victory^plot 34^56^ Lisois III, ii^edition^Lisois^recognition^plot 20^20^ Vamir III, iii^edition^Vamir^doubts about^plot 46^46^ Amtlie III, iv^edition^Vamir^recognition^plot 30^56^ de Foix  259 Le Duc de Foix III,^v 81  158  edition  III,^vi 10  10  III,vii 50  50  edition  Vamir A.élie  proposed flight  plot  edition  Vamir de Foix Amtlie Vamir de Foix  arrest  plot  de Foix' vengeance  character development  edition  56 IV,^ii  35  38 edition  IV,^iii 8  8 IV,^iv  IV,v 104  V, i  reproach  character development  edition  de Foix Lisois  execution order  plot  edition  de Foix  remorse  character development  edition  de Foix officer  execution report  plot  edition  de Foix Amtlie  reproach  plot  edition  de Foix Amelie Lisois de Foix Amtlie Lisois Vamir  execution denied  plot  55.5 V,^ii  3  23.5 V,^iii  4  42.5 V,^iv 31.5  8  edition  V,^v 46  plot  Vaair de Foix Lisois  104  26  plot  edition 19  19  14  attempted^plot reconciliation  edition  IV,i 49  Amelie's refusal^of de Foix renewed attack  Vamir de Foix Aselie Vamir de Foix Amelie Lisois Vamir Lisois  New lines:^976^Total lines 1400  reconciliation plot  259 Adelaide du 6uesclin, revived by Lekain, 9 September 1765.  Variants to the 1734 text location new^lines^total^lines  source  character gender  content  function  edition  Coucy Adelaide  politics  exposition  edition  Adelaide Taise  affection for Nemours  plot  I,^iii 0^69  edition  VendOme Adelaide  love for Adelaide  plot  I,^iv 4^23  edition  Vendame Adelaide Coucy  proposed attack  plot  I,v 2^4  edition  Adelaide  fear for Nemours  plot  Vendame  Teta of battle  plot  I,^i 37.5^170.5 .5^67.5  II,i  0^54  edition  Coucy  II,^ii 10^56  edition  Vendame Coucy Nemours  recognition by VendOse  plot  II,^iii 0^10  edition  recognition by Adelaide  plot  II,^iv " 1^LJ  VendOme Coucy Nemours Adelaide  Edition  Adelaide  fear for Nemours  plot  II,^v 38^99  edition  Adelaide Vendame  declaration  plot  II,^vi 14^14  edition  Vendame  despair  character development  II,vii 35^130  edition  Vendame Coucy  loyalties  character development  III, i 4^58  edition  Nemours Dangeste  doubts of Adelaide  plot  III,^ii  edition  Nemours  reconciliation plot  0^76  Taise  Adelaide  260 Adelaide du Guesclin (the new lines in the following scenes are drawn largely from les Freres ennemis) III,^iii 91  III,^iv  10  10  III, v 50  50  44 35  IV,i IV,^ii  edition  Nemours Adelaide VendSme  jealousy^plot  edition  Nemours Adelaide Vendame Coucy  renewed^plot attack  edition  Nemours Coucy  attempted^plot reconciliation  edition  Adelaide Nemours  proposed^plot flight  edition  Adelaide Nemours Vend ame  Nemours'^plot arrest  164  79 38  (the new lines in the following scenes are drawn largely from le Duc de Foix) IV, iii^edition^Adelaide^vengeance^plot 4^8^Vendame IV, iv^edition^VendOme^reproach^plot 19^19^Adelaide (the new lines in the following scene are drawn largely from les Freres ennemis) 90  IV,v  V,i  16  V,ii  0  V,^iii  5  V,^iv 8  108  61.5 21.5 46.5 30.5  edition  Vendame Coucy  execution order  plot  edition  Vent:lame officer  repentance  plot  edition  VendOme officer  execution report  plot  edition  Venclame Adelaide  plea for mercy  plot  edition  Vendame Adelaide Coucy  execution denied  plot  LU1  Adelaide du Guesclin ^ reconciliation plot V, v^edition^Vendame 9^50^ Adelaide Coucy Nemours New lines (compared to the 1734 edition): ^487^Total^lines:^1542 Variant lines within the Adelaide^of 1765 source date^location lines changed^total^lines  65PA (65PB,^W701.)  65PA (65P13) 65PA (6511) 65PA (65P8)  character^content (gender)  function  I, iii 4^69  edition  VendOme^rephrasing  style  II,ii 4^(7)56  edition  VendOme^rephrasing  style  Vendgme^rephrasing  style style  2^56 II,^v 1.5^99  edition  Vendame^rephrasing  II,vii 21^130  edition  exposition Coucy^politics Vendase Coucy^correction to above edition  813015  1765 130 iii,^i 4^57.5  edition  Dangeste^rephrasing Nemours  style  III,ii 4 (added)^76.5 2  edition  Adelaide^expansion  emphasis  65PA (65P8) 65PA (65P8)  III,^iii 9^164  65PA (65PB)  IV,^ii 2^38  65PA (65P8)  IV,^ii^-^iii  no scene division at line 117,^consequent renumbering through the act  65PA (65P8)  IV,^v  no stage direction see D 12964 for its addition  65PA (65PB)^2  IV,^v  W701_  rephrasing  style  edition  Vendame^rephrasing Adelaide  style  edition  Adelaide^rephrasing Nemours  style  edition 108  Coucy^rephrasing  style  262 Adelaide du Guesclin 8^(4)^  Vendale^expansion^emphasis  7^  Coucy^rephrasing^style Vend3me  ^ ^ edition^scene i divided at line 23, consequent 65PA V, i ^ renumbering through the act (6511, T67)  85PA^V, ii^edition^VendOme^staging the cannon shot (65P8)^3 (lost)^21.5 3^21.5^ Vendgme^rephrasing^style Variant lines (within the 1765 edition): 84.5^Total lines:^1542  263 Zulime, debut 8 June 1740, revival 29 December 1761. The base text is that of 1763:  date^location SOUTCC lines changed^total^lines  character (gender)  content  function  Feb 1740^V,^iii 9^130.5  corresp D 2164  rephrasing  style  Apr^1740^V,^iii 1^130.5  corresp D 2200  Atide^(f) Zulime^(f) Ramire (m) Zulime  rephrasing  style  V,^iii May 1760^.5^130.5  corresp D 8903  Zulime  rephrasing  style  Spring 1761^the edition disavowed by Voltaire tie in D 9854) offers considerable variation in text and a different denouement. date  character (gender)  content  function  edition  Hohadir^Cm) Zulime  prior events  exposition  edition  Zulime Atide  love for Ramire  character development  edition  Zulime Atide Ramire  proposed escape  plot  edition  Ramire Atide  Ramire's dilemma  plot  edition  Ramire Menodore(m)  proposed escape  plot  edition  Zulime Ramire  obstacles to marriage  plot  edition  Zulime Benassar^Cm)  refusal to reconcile  plot  edition  Zulime  doubt  character development  source location lines changed^total^lines I,^i  1761 38  114 I,^ii  47  66 I,^iii  21  51  I, v 11  70  II,i 55  93 II,^ii  31  91 II,^iv 98  III,^i 13  48  (The 1763 text adds a new scene ii, 19 lines, and a scene vii; making allowance for the renumbering, 1761 iii - v parallels 1763 iv - vi)  264 Zulime III,^ii 49  III,^iii 19  26.5  III,^iv 89.5  77  III,v 30.5  33  edition  Zulime Atide Ramire  rift^between^plot lovers  edition  Atide Ramire  renunciation^plot  edition  Ramire Btnassar  truce^plot  edition  Ramire Atide  reconciliation plot  (Here again there is considerable divergence between the texts, with only 1761 v- viii corresponding to 1763 iv - vii.) IV,i 4^edition^Ramire^anxiety^character development IV, ii^edition^Ramire^combat^plot 30^ Menodore IV, iii^edition^Ramire^departure^plot 6^ Zulime IV, iv^edition^Zulime^despair^character 4^ development IV, v^edition^Zulime^rtcit of^plot 42^46.5^Benassar^battle IV, vi^edition^Zulime^realization^plot 11^43.5^ of betrayal IV, vii^edition^Zulime^jealousy^plot 27^74.5^Atide IV,viii^edition^Zulime^death^plot 9^17^ Atide^sentence Strame (f) V, i^edition^Btnassar^plea for^plot 17^64^ Mohadir^mercy V, ii^edition^Btnassar^remorse^character 45^46^ Mohadir^ development Zulime  265 Zulime V, iii^edition^Bhassar^Atide's^plot 70^135^ Mohadir^death Zulime Atide Ramire Note:^in the 1763 established text,^Zulime dies. Variant lines:^674^Total lines (1763):^1205 Changes subsequent to the 1761 edition character (gender)  content  function  corresp D 9945  Zulime  rephrasing  style  Jan^1762^V,^iii 4^125  le Mercure II:^202  Zulime  rephrasing  Dec 1766^V,^iii 2^125  corresp D 13727  Zulime  rephrasing  source date^location lines changed^total^lines  Aug 1761^V,^iii .J^135  style  Note: the above letter corrects in text the passage given in the Mercure as the closing lines of the performance; this version is followed until the Kehl edition.  266 Mahomet: Comedie francaise debut 29 August 1742  date  ^  location^source^character^content^function  lines changed total lines^(gender)  Nov 1739^IV, iv-^corresp^/opire(m)^paternal^recognition D 2106^love^scene Jan 1740^IV, v^corresp^Zopire^paternal^recognition 2131^Ride fm)^love^scene Feb 1740^IV, v^corresp^Zopire^paternal^recognition 02160^Ride^love^scene Jun 1740^V^corresp^Mahomet^deaths of^plot revision 02218^Palmire^Ride and^in response Ride^Palmire^to criticism Jul 1740^V^corresp^deaths of^plot revision 02266^ Ride and D 2267^ Palmire Jan 1741^IV, v^corresp^Zopire^paternal^recognition 5^D 2408^Ride^love^scene V, iv^D 2408^Mahomet(m)^remorse^character 5^ Omar (m)^development Jul 1741^II, iv^corresp^Mahomet^motivation^character 02515 II, v^02515^Ride^rephrasing^style .5 V^D 2515^Mahomet^motivation^character Palmire (f)^for movement development  Aug 1741^II, vi^corresp^Omar^description^character 9^D 2525^ of Side^development In August 1741 Voltaire reported to the marquis d'Argenson (0 2523) that he had changed or added 200 lines to the text subsequent to the version the marquis had in manuscript, but gave no specific lines.  267 Mahomet 1748^  edition  Moire^politics^ideology 0^130^ Omar 1748^II, i^edition^Sêide^rephrasing^style .5 II,vi^  Omar^rephrasing^style  .J  III,i^ Palmire^rephrasing^style 20^45^ Sêide III, viii —^ ix^ III,x^ 28^26  Zopire^transposition scene Side^of lines^division change Zopire^paternal love foreshadowing  Omar^rephrasing^style IV,i^ 12.5^42^ Mahomet  1751^  edition  I, i^ 1^ III, ii^ 1 word  1756^  Zopire^rephrasing^style Phanor Palmire^rephrasing  edition  I, i^ .5  Zopire^rephrasing^style  II,i^ 2 words  Seide^rephrasing  1775^  edition  IV, vi^ Zopire^Zopire's dying removal of 5^53^ Phanor^plea^revenge motif Variant lines : 91^Final text: 1448 lines Act V, 190 lines long, was an extension of the original conception of the text, which had ended at Zopire's death. Voltaire added the deaths of Sêide and Palmire, Mahomet's final triumph, and his suppressed remorse in reaction to criticism.  268  Semiramis, debut 28 August 1748, revived (to greater success) 10 April 1749 In subject and in stage Effects (the ghost), Semiramis can be considered a development of Eriphile, and is therefore included.  date  ^ lines  location^source^character^content^function  changed total lines^(gender)  Jun 1748^II, vii^corresp^Assur (m)^critique of^ideology 2^88^D 3677^ religion (lines reinstated after Crebillon had censored them) Aug 1748^I,  ii^perf^  cries from the tomb eliminated  1749^II, i^edition^AMr^attack on^plot (to 56)^14^(18)100^ Arzace 1749^II, iv^edition^Assur/^expansion^emphasis (to 56)^2.5^(7.5)82^ Cedar^ (shift in  speaker) Variant lines: 19.5^Total lines 1649 There are several references in contemporary sources (such as P. Clement) to extensive revisions between 1748 and 1749, but without exact texts.  269 Oreste, debut 12 January 1750 date  ^  location^source^character^content^function lines changed total lines^(gender)  1750^I, i^edition 4(lost)^47.5^Pammtne(m)^attack on^ideology the court 6^47.5^ Iphise (f)^ shift in speaker I, ii^edition 6 (lost) 116.5^ Iphise^compression^style Electre (f) I, v^edition^Electre^rephrasing^style 3^89^ Clytemnestre (f) II,i^edition^Pylades(m)^compression^style 38^34^ Oreste (m) II, v^edition 3^(2) 80^ Pamméne^compression^style 10^(2) SO^  Egisthe (m)^the gods'^ideology silence  II,vii^edition^Iphise^rephrasing^style 2^72 (the correspondance refers to editing in this scene, without citing specific lines; 84095, 4098, 4099, 4104 to Clairon) Act III has a completely different organization of scenes in the 1750 edition; scene i (83 lines) includes what would become scenes i - iii (81 lines). III,i^edition^Nuke 74^83 (in this text)^Oreste^vengeance^plot Pylades III, ii^edition^Electre^recognition^plot 38^ Iphise Oreste Pylades (picking up the established text at scene iv, line 3) ^ ^ edition III,vi Oreste^mistaken^plot ^ 15.5 (19.5)117 Clytemnestre^recognition Egisthe ^ IV,i edition^Oreste^rephrasing^style 2^32  270 Oreste IV,^iii  edition  Electre Iphise  compression  style  edition  Electre  expansion  emphasis  edition  Electre Oreste  compression  emphasis  edition  Egisthe  rephrasing  style  perf  Clytemnestre  death of  staging  edition  Pasméne  recit of death  staging  perf  Oreste  compression  style  edition  Oreste  rephrasing  style  16^(1°) 59 IV,^iv  10  15  IV,v 22^(17) 51.5 V, iv 2  4.5  V,^viii 11^(9) 32.5 V,^viii 7  32.5  V,^ix 15^(11) 28 V,^ix 28  Variant lines: 290.5 ^Total lines: 1524  ^  271 Rome sauvt,e, datut 24 February l752 date^location stucco lines changed^total^lines  charabtar  Sept^1751  I,^iii 17^87  corresp 04577  III,^iii 49.5 III,^vi 1^20 IV,^vi  content  function  Catilina^(m) Aurelie^(f)  rephrasing  style  Aurklie Catilina  rephrasing  style  Aurtlie  rephrasing  style  Aurelie  rephrasing  style  Catilina  rephrasing  style  Cickron^(m)  rephrasing  style  Aurelie  rephrasing  style  (gender)  75 Jan 1752  II,^i 2.5^67  corresp 04774  5 (unprinted) IV,^vi corresp Feb 1752 ^2^75^04787  The following variants come from a manuscript cited by Kehl, called detached and unpublished by Beuchot, date unclear. Act I, v to II, i: there are changes in scene order and scene division which do not greatly affect content. I, v 8 (added) 118^manuscript^Catilina I, vi manuscript^Catilina I, vii 40^60^manuscript^Ciceron^rephrasing^style Caton (m) II, i^manuscript^Ciceron^rephrasing^style 47^67^ Céthegus The following variants come from the Berlin edition, 1752.  1752^II, i^edition^Catilina^expansion^emphasis Cethegus  ^62^68^  ^  272 Rome sauvee II,^ii 44^52  edition  Catilina Cethegus Lentulus  expansion^emphasis  II,^vi 2^77  edition  Catilina  rephrasing^style  III,^i 7^(3)^51  edition  Catilina  compression^style  edition  Catilina  rephrasing^style  Catilina Aurelie  compression^style  Aurtlie  rephrasing^style  III,^ii 8^99 57^(25)  3 III,ii^-^iii  scene division moved  IV,i 4^28  edition  Cèthegus Lentulus  rephrasing^style  V,^i^- ii 37^108  edition  Cictron Lentulus Ctthegus  change in time of entrances  The following variants come from the Dresden 1753 edition 1753^I, i^edition^Catilina^rephrasing^style 10^24 edition^Catilina^motivation^character 22^52^ for rebellion development I, iii 8^87 8 (lost)  10  edition^Catilina^rephrasing^style Catilina^compression^style Aurelie Catilina^stage business removed  I, iii - iv^edition^  scene division moved  I, v^edition^Cictron^rephrasing^style 8^110  II,i^edition^Catilina^rephrasing^style 8^67  273 Rome sauvee II, ^edition^Cesar^expansion^emphasis 4^(8) 97 II,iv^edition^Catilina^compression^style 10^(4) 5 III,ii^edition^Catilina^expansion^emphasis 35^103^ Aurelie (final text expands this to 2 scenes) III, vi^edition^Catilina^compression^style 5^(2) 20 IV, iv^edition^Catilina^compression^style 32^(17)^119.5 IV, v - vi^edition^Aurêlie^change in scene division 1753 - 140 lines^ Catilina^and compression final text - 94 lines^CicOron Cethegus Caton  Variant lines: 658.5^  Total lines: 1512  274 elyppie, Comedie frangaise debut 17 March 1764 date^location source lines changed^total^lines  character (gender)  content  function  Jan 1762^V,^i 10 (det)  corresp 810243  Hermas  Cassandre  44  character development  Feb 1762^IV, v 1  corresp D 10333  Olympie(f)  order to flee  action  87  Mar^1762^IV,^iii 1  corresp D 10388  high priest  rephrasing  style  37  Jul^1762^III,^v 2.5(det)  corresp D 10594  Statira^(f)  rephrasing  style  69.5  Oct^1762^III,^vi 2  corresp D 10770  Olympie  rephrasing  style  73  rephrasing  style  Jan 1763^IV, iii high priest corresp 2 37 810932 (This variant attaches to the 1774 edition text.)  The following variants come from the 1763 Frankfurt edition which preceded performance at the Comedie frangaise. 1763^I, iv^edition^Cassandre (m) rephrasing ^style 39 III,vi^edition^Statira^rephrasing^style .5^73 IV,ii^edition^Antigone (1)^rephrasing^style 2^52.5 IV, v^edition^Cassandre^rephrasing^style 3^87 V, iii^edition^Olympie^rephrasing^style 2^92  The following variants come from performance texts. 1764  ^  ^ ^ I, ii pert Cassandre^history^exposition ^ 12 (lost) 142 Antigone 1 (added)  ^  Antigone^history^exposition  ^  275 Olympic I,^iii 6^L7  pelf  Hermas/ Antigone  shift^in speaker  staging  I, iv 6^(lost)^39 2 (added)  pert  Cassandre  proposed marriage  plot  II,iv 9 (lost)^50  pert  Olympic Statira high priest  recognition  plot  III,i 13^(lost)^50  pert  Cassandre  remorse  character development  IV,ii 2^(lost)^51  pert  Antigone  marriage  plot  IV,^iii 6 (lost)^37  perf  high priest  marriage  plot  IV,^iv 4 (lost)^42  pert  Sostene (0  description of Olympic  character develoment  IV,^v 4^(lost)^79  pert  Olympie  despair  character development  IV,vii 8(lost)^18  perf  Olympic  despair  character development  V, vii 7^(lost)^36  perf  Cassandre  suicide  plot  The following variants come from the 1774 edition, which perpetuated many of the performance edits, indicated by an asterisk. 1774^II, iv * 4 (lost)^50  edition  Statira Olympic  recognition  plot  III,i 4 (moved) 50 * 13 (lost)  edition  Cassandre  rephrasing  style  Cassandre  remorse  character development  IV,iii 2^37  edition  Cassandre high priest  rephrasing  style  high priest  Statira's death  plot  * 6 (lost) 6 (added)  ^  276 Olympie IV,^iv *4^(lost)^42  edition  Sosténe  description of Olympie  character development  IV,^v *4^(lost)^79  edition  Olympie  despair  character dEvelopment  IV,^vii *8 (lost)^18  edition  Olympie  despair  character development  IV,viii 6^24  edition  Olympie  Statira's curse  plot  V, v 4 (added)^22  edition  Cassandre Olympie  plea for fidelity  plot  edition  Cassandre Antigone  shift^in speaker  staging  edition  Cassandre  suicide  plot  V,^vi 45 V,^vii *7 (lost)^36 3 (added)  Textual variants after the 1774 edition are largely grammatical and therefore omitted; however, the Kehl edition offers one more substantial variant: 1784^II, ii^edition^high priest^rephrasing^style 2.5^127.5  Variant lines: 136.5 ^Total lines : 1510  277  Le Triumvirat, debut 5  july 1764  The variants available in the Moland edition come from a undated "early" manuscript, likely 1763, and give an excellent indication of the extent to which text was reworked before  performance. date  ^  location^source^character^content^function lines changed total lines^(gender)  1763?^I, iii^manuscript^Octave (m)^politics^exposition  106^138^Antoine (m) relocated in manuscript to II, i and replaced by 64^manuscript^Fulvie (f)^pardon for^plot Antoine^Porn* Act II 33^0^manuscript^Octave  (scene removed)^Fulvie  reproach for^plot abandon  III, iii, v^manuscript 63  change in entrances and characters on stage  III,vi^manuscript^Octave 61^(45) 117.5^Julie (f)  compression^style  Act IV scenes i - iii are interrupted by new scene ii and iii, and a consequent renumbering of scenes in the manuscript. The scene numbers of the printed text are given in parentheses IV,ii 59^0  manuscript  Julie Fulvie  despair  character development  IV,^iii 56^0  manuscript  Julie  proposed  plot  Fulvie Pool*  flight  IV,^iv 3^(ii)^28  manuscript  Fulvie Pomp&  rephrasing  style  manuscript  Fulvie Pompee Aufide (0  assassination  plot  Octave Antoine  recit of attempt  plot  Julie Fulvie  recit of attempt  plot  IV,vi  12^(iii)(10)76 V,i 77^(lost) replaced by^42.5  manuscript  proposed  Le Triumvirat V, iii^manuscript^Julie^reproach^plot 19^(11) 32.5 Variant lines: 437^Total lines:^1364  279  Les Gugres, unperformed, first printing 1763 date^location source lines changed^total^lines  character (gender)  content  function  Dec^1768^I,^i 4^120  corresp D 15379  Iradan^(m)  rephrasing  style  Sept 1769^I, i 2^120  corresp  Iradan  rephrasing  style  edition  Iradan high priest  religion and law  ideology  II,vi 7^(3) 19  edition  Iradan Uséne (m)  attack on priesthood  ideology  II,vii 7^(3) 52  edition  Iradan y Arzémon ()  attack on priesthood  ideology  III,i 7 7  edition  y Arzemon Mtgatise (m)  rephrasing  style  83 (3)  expansion  emphasis  III,ii 3.5  81  Taking the 1769 text as our base: y Arz6mon = young Arz6mon, o Arzhion = old Arzkon 1771  I, iii 4^(lost)  50  IV,i  Arzame (f) y Arzemon  rephrasing  style  edition  o Arzémon^(m)^rephrasing  style  edition  Ceséne Arzame  rephrasing  style  edition  o Arzemon  rephrasing  style  79  1 IV,^iv 4 1778  edition  20 IV,^v 16  Variant lines:  ^  40.5^Total lines: 1593  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics

Country Views Downloads
United States 35 1
Germany 33 125
China 26 4
Russia 15 0
Japan 8 0
Romania 5 0
Switzerland 4 0
Sweden 3 0
France 3 0
United Kingdom 2 0
Canada 2 0
Singapore 2 0
Chile 1 0
City Views Downloads
Unknown 53 132
Shenzhen 13 4
Saint Petersburg 11 0
Ashburn 11 0
Tokyo 8 0
Putian 6 0
Fremont 5 0
Buffalo 5 0
Beijing 5 0
Wilmington 2 0
Stockholm 2 0
Mountain View 2 0
Frankfurt am Main 2 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items