Although interest in the problem of reference dates back to the Middle Ages (de Libera 1987; Rosier 1987), intensive research on this subject did not begin until the end of the nineteenth century, notably with the works of Frege and Meinong. Most critics dealing with the prominent position occupied by reference in contemporary scholarship tend to distinguish between two different types of inquiry: the linguistic (European, or Continental) school of thought, and the Anglo-Saxon (or analytical) philosophical tradition. While linguists conceive of reference as a process and an act, philosophers view reference as a relationship between an expression and its referent. The linguistic tradition, initially hampered by its roots in Saussure's theories, which accord no place to the referent, has made noteworthy contributions to the study of intratextual reference, focussing on certain factors (modalizers, deixis, anaphora, and cataphora) which determine and affect reference. The philosophical inquiry, largely conducted within the framework of logic and modal logic (as in the writings of Frege, Russell, Ryle, Quine, and the early Strawson), began with ontological questions concerning the logical status of the referent and the truth value of the referring expression (see Linsky 1967). In what could be considered as a second stage of the analytical debate, theorists such as Austin, Searle, and the later Strawson have been less preoccupied with questions of existence, choosing instead to emphasize semantics, speech acts, and other pragmatic issues, especially the role of contextuality and conventions in reference.
In contrast to these well-travelled "routes of reference" (Goodman 1981) of the philosophical tradition, the question of reference in literary texts has remained, until quite recently, a largely unexplored territory. Several different reasons have been suggested for the reluctance on the part of scholars to tackle the thorny problem of referring in literature. In his study of literary reference in terms of reception theory, Fitch (1991:x) notes both the predominance of textual approaches inspired by Saussure's linguistic model and the consequent refusal to admit any relationship between fictional referentiality and the "real," extratextual context. In addition, philosophical studies of the ontological status of the referent did not seem overly relevant to literary critics, who were not particularly interested in the truth value or the "real" existence of referents in literature. Indeed, the path leading from analytical philosophy to the study of referring in literature was by no means a direct route, especially given the lack of interest among these philosophers for literature per se, as Cometti observes: "Si la philosophie analytique n'est pas restée étrangère à la fiction, elle est bel et bien demeurée sourde à la littérature" (1991:18). Whereas certain literary theoreticians (Pratt 1977; Hewitt 1987) have attempted to adapt speech act theories to the analysis of literature, analytical philosophers preoccupied with literature have been more interested in imposing "philosophical standards on literary texts" (Avni 1990:177) than in exploring possible relationships between literature and philosophy.
Despite the lack of attention accorded to reference in literature in the past, the increasing number of publications that have appeared on this subject in recent years, authored almost exclusively by literary theoreticians, attests to the pertinence, relevance, and necessity of this area of inquiry. Not only, as Fitch claims (1991:4), does our sustained interest in story-telling throughout the ages indicate a connection (however direct or indirect it may be) between literature and life, but the analysis of the meanings of a text inevitably entails the issue of reference, as Whiteside explains: "though the signs of literary discourse constitute a text's linguistic dimension, its particular ideological significance lies beyond, and is determined by, the referent" (1987a:176). Indeed, recent studies demonstrate without doubt that the study ofreference forms an essential part of literary interpretation and that literary texts provide a privileged and complex setting for the exploration of the referential functions of language. Two of the latest additions to the ever-expanding list of publications in this field include Avni's The Resistance of Reference and the June 1994 issue of the Québec periodical Tangence, entitled La référence littéraire.
Readers expecting a comprehensive theory of the process of referring in literature will be disappointed by Avni's book, and would best be advised to consult Whiteside's excellent publications (1987a, 1987b, 1990, 1991, 1994) on this subject. What Avni does provide, however, is a survey of various theories, mainly of the analytical vein, relating to the overall functioning of language, in which reference plays an integral role, and an analysis of selected literary texts in light of these theories. The "resistance" of the title has, to my mind, several possible referents: the resistance of literary texts to the direct application of theories in general, and of philosophical and linguistic theories in particular (literary texts produce "excesses" and "agrammaticalities" usually dismissed by such theories); the resistance of certain theoreticians (linguists in general, and Saussure in particular) to include reference within the scope of their investigations; the resistance of analytical philosophers when it comes to dealing with fictional texts and non-existent referents; and finally, the resilient nature of reference as a concept, its durability in the face of efforts to ignore or banish it. While Avni might well have had other forms of resistance in mind while ruminating on her title, I would like to propose yet another: Avni's own resistance to reference, first, in terms of her evident reluctance to discuss reference explicitly and thoroughly, in and of itself; and secondly, in terms of her lack of reference to the expanding body of criticism currently (and previously) published specifically on referring in literature.
Advocating a "two-way relationship" between philosophy and literature, Avni expresses the hope that "philosophy will provide us with new insights into the workings of the literary text and, conversely, the examination and reevaluation of these theories by literature" (178). This dialogue will take place in the theatre of language, hence Avni's decision to "write mostly about thinkers who have enormously influenced the way we think about language today" (16) and to emphasize language's "ability to program and produce irregularities" (16). Avni thus sets the stage, in the book's first and introductory chapter, by addressing the two theoretical traditions mentioned above: European linguistics, based on the work of Saussure, and Anglo-American philosophy. These two traditions correspond, in turn, to that which Avni terms two coexisting "aims" of language: the "semantic aim," which refers to the horizontal functioning of language, its capacity to combine units syntagmatically, producing meaning that is not simply the sum of its constituent parts; and the "deictic aim," whereby words aim at objects by means of monstration, or verbal, vertical gestures. As her detailed discussion of several theoreticians and her intricate readings of four different literary texts indicate, these two aims alone do not suffice to account for the complexities of language and the "excesses" of literature. Confronted with this dilemma, Avni briefly introduces (without satisfactorily explaining) what she will develop in the sixth and final chapter as her solution to the problem: in addition to the two "aims" described above, language also functions by means of the "semiotic" or contextual values of signs, as well as the "cognitive articulation" (13) between the two aims.
Before enlarging upon the somewhat vague concept of "semiotic values," Avni analyzes the complexities and presuppositions of the two aims of language, as represented by two thinkers considered as founders of the two opposed traditions: Saussure (the "semantic aim") and Frege (the "deictic aim"). Given that Saussure's abolition of the referent is a well-known tenet of his thought (see Cometti 1991:3; Coquet 1991:23), it seems rather excessive to devote such a lengthy (sixty-page) chapter to his theories in a book supposedly dealing with reference. Much space is, however, allotted to a careful and critical explanation of Saussure's "resistance" to reference, his tireless efforts to establish that language is not a nomenclature, that words do not name objects or concepts, and that meaning results from the oppositions and relations between the terms of the linguistic system. Although Avni reiterates descriptions of Saussure's major concepts (the langue/parole distinction, the synchronic approach to linguistics, the dual nature of the sign, the sign's arbitrary nature, and linguistic value), which will already be familiar to most readers, this chapter does contain some lesser-known insights. What Avni adds to our understanding of the Cours de linguistique générale is her impressive knowledge of Saussure's three different courses, gleaned from extensive consultation of manuscripts, students' notebooks, and Rudolf Engler's critical edition of the Cours. Avni is thus able to stipulate the specific modifications that Saussure effected each time that he taught the course, the changes in his pedagogical and narrative strategies, and the decisions (and omissions) entailed in Bally and Sechehaye's selection of material for the published version of the course. Having taught numerous courses on Saussure's course, I found this information to be valuable and interesting in its own right, but I could not refrain from questioning its relevance for a theory of reference, be it "resisted" or fully developed. Other issues raised in this same chapter could also qualify for the extraneous category, as far as reference is concerned. Avni reviews various criticisms of Bally and Sechehaye's editorial decisions (especially their choice of langue as a starting point for the Cours), Ogden and Richards' criticisms of Saussure's theories, as well as Culler's presentation of Saussure's thought to the English-speaking public. Not only does the question of reference slip quietly into the background during a good deal of this discussion, but the order in which certain ideas are presented seems rather odd and arbitrary. Avni jumps from the content of the three courses to that of the published Cours, then to the criticism of the actual form of the Cours itself, to various criticisms of Saussure's concepts themselves (which had not as yet been explained), and finally, to a thorough consideration of the major features of Saussure's theory of language. So it is that we read a summary of Culler's summary of Saussure's ideas before Avni actually presents the latter to us in any detail.
Despite these drawbacks, there is much to learn from this chapter, whether it be how Ogden and Richards' misreading and rejection of Saussure's theories delayed the reception of the latter in Anglo-American circles; how and why Saussure insisted on maintaining both the concept of value and that of the dual nature of the sign, which, Avni claims, is a contradiction illustrating "the incompatibility between the exigency for a fixed meaning of words [. . . ] and the contamination of meaning due to the interaction of each unit with others around it" (72); how Saussure himself was one of the most astute critics of his own work; and how the epistemological and methodological issues at the heart of his system (especially his difficulty in determining an obvious starting point or point of origin for the delineation of his theories) reflect and determine the results of the theories themselves. More problematic is Avni's presentation of Benveniste, introduced in the context of the quarrel over the arbitrariness of the sign. Benveniste's "correction" of Saussure, that is, his proposal of a relation of necessity between signifier and signified and the existence of a third term (not included in Saussure's system) - the thing itself, "la réalité" (Benveniste 1966:50) - has had, according to Avni, "meagre" results (57). It is surprising that Benveniste's introduction of the referent into the European linguistic tradition should be so hastily downplayed, if not dismissed, particularly in light of Avni's criticism of Saussure's unique concentration on the "semantic aim" of language. Not only did Benveniste demonstrate that the relation of necessity between signifier and signified is itself a necessity, in order for communication to take place, but this adjustment to Saussure's doctrine presents referring as one of the essential purposes of language: "le langage porte référence au monde des objets" (Benveniste 1966:128). In keeping with his thoughts on the functioning of language later outlined in "Sémiologie de la langue" in terms of the semiotic and semantic modes of language, Benveniste (1974:43-66) modifies Saussure's conception of arbitrariness precisely to account for what Avni has already termed the semantic and deictic aims of language.
But there is more. In her discussion of Benveniste, Avni fails to mention that his notion of the referent is, in fact, related to the cornerstone of his thought, i. e. , his investigation into énonciation (a dimension, as is well known, neglected by Saussure in his quest to isolate langue as a system) and subjectivity in language. As Benveniste contends, "la référence est partie intégrante de l'énonciation" (1974:82), since to use language is to speak about something or someone (Benveniste 1974:62). Furthermore, in Benveniste's view, every énonciation entails a speaker (or writer), a listener (or reader), a context, conventions, and other pragmatic considerations which are all currently deemed essential to the study of reference. Avni's disparagement of Benveniste's contributions to the theories of reference is quite puzzling, if not inaccurate. It should also be noted that she refers solely to Elizabeth Meek's 1971 English translation of Benveniste's Problèmes de linguistique générale, which is limited to the first volume of the Problèmes and therefore does not include crucial (and as yet untranslated) articles such as "L'appareil formel de l'énonciation" and "Sémiologie de la langue," both of which were published in the second volume of Problèmes (1974) and constitute essential references on this subject.
Having exhaustively investigated the "semantic aim" of language as it manifests itself in Saussure's ideas, Avni turns her attention, in the following (much shorter) chapter, to the "deictic aim," as represented by Frege's philosophy. Once again, Avni points out the inconsistencies and paradoxes that plague most theoretical undertakings: in Frege's case, his fervent desire to avoid ambiguity, to rid language of subjectivity, led to his attempts to discover a logical, "formula" language (a Begriffsschrift), but this endeavour, in turn, forced him to use and focus on the very natural language that he wished to abolish. This chapter also features the expected review of Frege's famous distinction between sense and reference, but Avni's discussion is more complete than most, as she includes the oft-neglected concepts of the object (Gegenstand) and the idea (Vorstellung). Avni takes pains to clarify the ambiguous status of the object, which is not merely a thing or "existential entity," but it serves as a link between the linguistic and non-linguistic domains. Even though the object "escapes the grasp of language, its formal delineation and definition [. . . ] presuppose and require a linguistic backdrop" (85). Equally intriguing is Avni's consideration of Vorstellung, which Frege devised as a necessary repository for any subjectivity that might have lingered within the confines of Sinn (sense). Once Sinn was defined as the "minimal semantic charge that all speakers of a given language recognize as being carried by an expression" (93), Frege conveniently attributed all additional semantic charges of a personal, subjective nature to Vorstellung, allowing him to invest Sinn with a strictly objective character. Avni's coherent critique of Frege's thought, while not concentrating explicitly on reference as such, clearly illustrates the pitfalls of any theory of language that denies the subjective, individual element which is "a substantial component of linguistic competence" (109). As in the previous chapters, Avni does not simply refer to the most obvious and familiar source (in this case, Frege's renowned essay, "On Sense and Reference"). Instead, she draws on a number of additional texts, written by Frege and his critics. Continuity with the previous chapter is achieved notably by an illuminating comparison of certain features of Saussure's and Frege's theories; here, Avni observes their common commitment to a synchronic approach to language and to the immutability of signs (although the individual reasons behind these shared allegiances differ).
Having devoted the first 112 pages of her book to thinkers who excluded either reference or literature (or both) from their theories, Avni finally embarks upon an actual study of literary texts in the fourth chapter, entitled "The First Person." This chapter was inspired by a course that Avni taught on "The First Person in Philosophy," the title of which presages the type of theories to be reviewed here, as linguistic theories of the "I" are largely neglected. The pronoun "I" has been chosen as a complex instance of referring in language, for its referent changes with each use, and "its semantic charge is not only totally different but also too complex and private to be fully communicated" (114). Avni appears to vacillate concerning the special referential status of the "I," noting at first how this pronoun constitutes a "puzzle for most theories of language" (114-115), and then attenuating the unique nature of the "I" with this contradictory statement: "My contention is that the I does not differ essentially from other words (at least not from a theoretical viewpoint)" (115). This unexpected declaration, seemingly at odds with the objectives of the chapter, is then countered by the claim that the "I" does everything that other words do, and more (115). To my mind, this rather confusing introduction to chapter four could have been greatly clarified by specific reference to any number of linguistic or textual studies on deixis in general, and on the peculiar nature of the first-person pronoun in particular, where the referential versus semantic debate concerning the "I" has been waged for some time (see Jespersen 1922:123; Benveniste 1974:68; Ricoeur 1975:98; Kerbrat-Orecchioni 1980: 36-37). Avni does, however, correctly point out the importance of the role and function of the context in which the "I" is uttered, in order to determine its "referential, epistemological, and ontological values" (114). The context chosen here is the literary one, not to confer any privileged status to literary language, but to shun "the creation of simple contexts fabricated by the theoretician to demonstrate a point" (116): for Avni, a literary narrative is a complex context in which language is free of the "necessarily biased pedagogical or explanatory purpose" (116) of the theoretician. The particular narrative selected for analysis in this case is Kafka's short story "A Report to an Academy," in which an ape, using the first-person pronoun, makes an invited autobiographical speech concerning his "former" ape self to an audience of humans. What interests Avni is the rhetorical effect of the ape's speech, the irony inherent in the contrast between the ape's consciousness of himself (he insists that he is human, despite bodily characteristics to the contrary) and the public's perception of him as an ape. The procedure Avni adopts is to first consider theories concerning the "I" as developed by five theoreticians (four philosophers and one linguist) and then to attempt, after each theoretical discussion, to apply the theory in question to the functioning of the "I" of the ape-narrator.
Avni begins with a brief review of Frege's view on the "I" as a formal marker of the discursive origin of a specific utterance, devoid of subjectivity. I remain sceptical as to the necessity of including Frege in this chapter, for he did not formulate an actual theory of the first person, as Avni herself admits (127), and his emphasis on "signs, language and truth as objective entities" (128) makes application of his remarks to literary texts problematic. Any comments pertaining to the relevance of his ideas for this chapter could well have been relegated to a footnote, and the space could thereby have more profitably been used to explore other theories in which the "I" occupies a more central place. As it is, Avni does not even bother to apply Frege's ideas to Kafka's text, but turns instead to an examination of Russell's, Strawson's, and Anscombe's theories.
Russell's theories provide a sharp contrast with Frege's insistence on objectivity, given the former's endeavour to "reinstate man [sic] as a subject" (128) and his preoccupation with indexical expressions. The priority that Russell accords to the "I-now" over the "I-then," to an absolute present of the subject, would seem to suit the ape's purposes, according to Avni (in what I would judge as a rather literal application of theory to text), since he could forget his former ape self in an unequivocal "I-now" existence. Strawson's distinction between an expression, its utterance, and its use is evoked, but this proves to be less useful when applied to the ape's tale, as Strawson's theory stipulates that the utterer of an "I" must in fact be a person. It follows, Avni reasons, that either the ape, not being human, cannot speak, and his story therefore cannot exist, or that the ape, since he uses the first-person form, must be considered a person, but in this case, the irony of the text is lost. As for Gertrude Anscombe, a philosopher noted for her translation and studies of Wittgenstein, her claim that the "I" represents "the idea that the speaker has of the activities in which she is engaged" (155, Avni's emphasis) avoids the question of reference and, in this way, Avni maintains, would accommodate the ape's propensity to circumvent propositions relating to his present identity (157-158).
Benveniste's theories concerning the first-person pronoun are also found inadequate to the task of analyzing the Kafka story, because his conception of "I" as a discursive instance lacking corporeal predicates, free of truth conditions, would deny the text's essential ironic dimension. Despite Avni's careful consideration of this aspect of Benveniste's thought, I found some of her arguments irritatingly reductive. Her characterization of Benveniste's theory of language as being "capable of accounting conceptually rather than pragmatically for all linguistic manifestations" (146, my emphasis) is bizarre, to say the least, if we recall that Benveniste's writings on énonciation are known to be an integral part, if not the foundation, of recent research in pragmatics. As Armengaud states, pragmatics constitutes a "prolongement de la linguistique de l'énonciation" and an attempt to unify "le courant analytique d'Oxford et le courant formaliste" (1985:8, 18). I believe that Avni's problems in interpreting Benveniste result not simply from her use of the English translation of Problèmes (which causes such errors as the translation of énonciation as utterance [énoncé]) (147), or from her neglect of Benvenistes's theory of énonciation (the context in which he formulates his observations concerning the pronoun "I"), which goes unmentioned in Avni's account, but also from her reduction of the referent of the first-person pronoun. In Avni's view, Benveniste's "I" refers solely to the instance of discourse in which it is uttered. Consequently, the "I" has no referent, whereas "he," "she," and "it" are thus endowed. This line of reasoning leads Avni to conclude that in the utterance "I'm bleeding," it is the instance of discourse that is bleeding, not the subject or the speaker (151)! While it is certainly true that Benveniste, as Avni argues, is not interested in the speaker's "corporeal qualities" and "states of consciousness" (150), his description of the referent of "I" and other deictic expressions is considerably richer than Avni would have us imagine. The pronoun "I," far from being bereft of a referent, is a doubly referring expression, as it points simultaneously to the speaker, "la personne qui énonce la présente instance de discours contenant je" (Benveniste 1966:252) and to its specific instance of discourse (and hence, to the act of énonciation which produced it). The very fact that "I" cannot be completely and adequately defined outside of its use in a particular context, a phenomenon that Milner terms "non autonomie référentielle" (1976:65), demonstrates the necessary existence of a speaker in order for an "I" even to be uttered in the first place.
In the following chapter, "Speech Acts," Avni scrutinizes certain major features of speech act theory and tests their applicability to two literary texts: Mérimée's "La Vénus d'Ille" and Nerval's "Le roi de Bicêtre." Particular concepts of Austin's, Searle's and Fish's theories are astutely analyzed and criticized: intentionality, the distinction between performative and constative utterances, perlocutionary and illocutionary effects, the role of context and conventions in speech acts, and felicity conditions. The choice of Mérimée's text is an appropriate one in this case, in that it raises intriguing questions about "the conditions for belief and disbelief, felicity and infelicity, sense and nonsense" (185), as well as contexts and conventions. Avni's insightful reading of the text, which also explores the text's intertextual dimensions, clearly illustrates how this story "misbehaves," questioning the very conventions on which it depends. Nevertheless, her analysis is based on her interpretation of a key gesture or "incident" (186) as a performative speech act: Alphonse (a bridegroom), annoyed by his wedding ring while playing tennis, places it on the ring finger of a statue of Venus, and is subsequently unable to remove it. Although a performative is defined as a first-person, present-tense utterance which performs the action it describes (as in baptism), Alphonse actually said nothing, making no verbal promise of marriage to the statue whatsoever. Despite Avni's contention that "there is no difference between a verbal and a nonverbal speech act" (190), the ambiguity of the story rests on the fact that Alphonse did not speak to the statue, but did promise marriage to his wife. Is the specificity of spoken or written language to be considered synonymous with that of gestures? An equivocal, silent speech act (on the part of a character quite capable of speech) thus gives rise to interpretive enigmas, errors, and agrammaticalities. Related issues concerning the sometimes unexpected effects of speech acts, especially in terms of the belief in the deictic aim of language, are further highlighted in Avni's discussion of the Nerval text, but in both textual analyses, the notion of reference itself could have been more explicitly explored.
Reference does resurface at the beginning of the final chapter ("The Semiotic Values"), as Avni recalls the failure of the theories examined thus far to adequately account for both the semantic and deictic aims of
language. In light of this impasse, occasioned by the perception of the "orders of things and words as two autonomous planes" and the insistence on reconciling them (235), Avni proposes the simultaneous pursuit of "two contradictory lines of thinking" (235). While bearing in mind the inextricable connection between the orders of things, thoughts, and language, we should accept the fact that the "radical heterogeneity" of these orders "resists" synthesis into one, all-encompassing theory. Attention must be focussed, when interpreting a literary text, on how the narrative affects the circulation of signifying objects, rather than on the state and actions of characters. For Avni, this preoccupation with the contextual and referential circumstances in which objects become signs constitutes a semiotic perspective. At long last, Avni reveals her solution to the dilemma of accounting for the two aims of language, as she defines "semiotic values" as follows: "the manner in which the signifying object evokes the referential circumstances in which it comes to acquire a meaning, and which subsequently govern its circulation" (240). Using and criticizing Mauss' writings onthe gift and Lacan's seminar on Poe's purloined letter, Avni deftly demonstrates how certain objects -- the twelve diamond studs presented to the queen in Dumas' Les Trois Mousquetaires -- acquire various semiotic values along their circuitous route of "transactions and displacements" (256), as they are stolen, counterfeited, given, and returned. While this emphasis on objects-become-signs functions adequately in the context of Les Trois Mousquetaires, (a traditional narrative), and would probably do so in other texts featuring the obvious transfer of objects, what do we do with contemporary (or other) narratives that do not accentuate the movement of objects, or narratives in which the itinerary of the object(s) is not so easily discerned? Are such texts lacking in semiotic values, merely because they are lacking in objects? Furthermore, I am unconvinced as to the originality of the concept of "semiotic values" acquired by an object, as it is not without certain affinities to the "object of value" presented in Greimas' actantial model (Greimas 1983:19-48), where equal emphasis is, admittedly, accorded to the actants' narrative programmes.
Generally speaking, I am left with mixed feelings towards Avni's study, divided between its obvious strengths and weaknesses. First of all, Avni fails to specify what she actually means by the term "reference," and how it relates, explicitly, to the two aims of language and their "cognitive articulation." Are we to assume that reference is only implicated in the deictic aim, where the verbal monstration of objects is evident? Various conceptions and definitions of reference abound (Lyons 1977:I, 174; Ducrot and Todorov 1972:317-324; Cometti 1991:8; Goodman 1981:121), and it would have been to Avni's advantage to situate her own work more clearly in relation to these studies. A similar "resistance" to reference is evident in the surprising lack of consideration given to the publications of literary scholars who deal precisely with referring in literature. Apart from the obvious absence of Whiteside and Issacharoff's On Referring in Literature, and Whiteside's other analyses of this subject, a variety of other publications on or related to reference in literature go unmentioned: the pragmatic theories of fictionality (van Dijk 1976); analyses of fictional existence, fictional worlds (Pavel 1988) and fields of literary reference (Harshaw 1984); as well as numerous other publications (Riffaterre 1990; Genette 1991; Lavis 1971, to name but a few). In addition, Avni's concentration on the Anglo-Saxon analytical tradition, combined with her identification of the semantic aim of language with linguistics as represented by Saussure (and his abolition of the referent), lead her to neglect more recent studies in European linguistics that do deal with questions of reference and deictics, which are, for Avni, associated exclusively with the deictic aim of language. Not only does current work in linguistic pragmatics address both aims of language and their interaction, the field having developed beyond the confines of Saussurian structuralism, but Avni's linking of the "deictic aim" (defined as "the referential function of linguistic signs" ) solely with philosophical (and not linguistic) theories leads to terminological confusion. The term "deictic" has already been used to refer to a particular class of signs (shifters, deictics, indicators, or indexicals) which have been the object of intensive research by linguists as much as by philosophers, especially following in the wake of Benveniste's (1966, 1974) and Jakobson's (1963) seminal studies. Avni's use of the term "deictic" to denote a much larger entity is consequently somewhat confusing, particularly since she sometimes describes the deictic aim as the "semiotic" aim (10, 230), which naturally evokes the supposedly separate concept of "semiotic values." To these criticisms I would add one final comment: certain of Avni's statements seem unnecessarily exaggerated (in order to emphasize her argument), to the point of being inaccurate. I am thinking here of the "bleeding instance of discourse" and of her contention that Saussure's signifier has absolutely no materiality whatsoever (47, 242), despite much consensus to the contrary (Coquet 1991:24, 27).
These weaknesses, however, do not outweigh the many merits of La référence littéraire.. The perceptive and detailed discussions of the theories selected and the minute readings of the literary texts attest to Avni's impressive erudition and meticulous research. Continuity between the various chapters is attained by the constant parallels drawn between the different theories under investigation. Avni wisely avoids the proverbial red herring that haunts much work on reference in literature; namely, the ordinary versus poetic language debate. Judiciously choosing to curtail elaborate discussion of such perennial philosophical questions as the bald king of France (who is relegated to his Russellian kingdom, where he rightfully reigns) (139-141), Avni concentrates on the importance of context and conventions in the study of literary reference and the "agrammaticalities" of the literary text. Furthermore, her insightful comments on the normative rigidity of theory, on our attempts to master the literary text that refuses reduction to a theoretical grid, and on the text's propensity to "excesses" or "remainders" which enable it to wriggle free from the imposition of theory, serve to remind us how the literary text "denounces, enacts, exploits, and thwarts the essential [. . . ] gesture by which a Ôtheory' distinguishes its outside from its inside" (264).
The complex process of referring in literature is also the subject of a recent issue of the Québec periodical Tangence (No. 44, 1994), a learned journal featuring theoretical and thematic debates of a literary nature. As yet relatively unknown outside of Québec, Tangence publishes quality articles that cover the wide range of the heterogeneous spectrum of contemporary literary theory. Past issues have examined topics such as "Ecritures au féminin: le genre marqué" (No. 47, 1995), "Le récit de soi" (No. 42, 1993), and "La fiction postmoderne" (No. 39, 1993). The various contributors to the issue on literary reference situate their analyses within a largely semiotic and linguistic framework. As guest editor Max Roy maintains, reference is understood here as a process distinct from empirical reality; its status can be extratextual, textual, or metatextual (5). In order to account for these diverse dimensions of literary reference, Roy has included a variety of articles in his issue that approach the question of reference from three major points of view. While certain contributors are concerned with constructing a comprehensive theory of literary reference, and others provide analyses of the modes of referring in specific literary texts, the remaining critics focus on the metatextual level of reference as it functions in the discourse of literary criticism.
The articles written by Philippe Hamon ("Texte littéraire et référence" [7-18]) and Anna Whiteside-St. Leger Lucas ("Dynamique de la référence en littérature: l'exemple de l'écriture visuo-verbale d'Apollinaire" [106-124]) provide the most theoretically- oriented articles in the issue, as both scholars take significant steps towards outlining actual theories of literary reference. Hamon cites several criteria essential to the study of referring in literature (8-9). Proposing that three referential components (text, fantasies, and ideology) combine and interact in a process of constant transformation and rewriting, Hamon takes the theory of literary reference beyond Avni's notion of language as "aims." For Hamon, literature demonstrates that reference does more than simply aim at and name concrete things; it also shows us that the "real is a crossroads of relations in permanent reformulation (the author's customs, norms, values)" (10, my translation). Hamon correctly notes the role of both ideology and subjectivity in literary reference: neither the referent nor the act of referring (for author or reader) is neutral. When examining literary reference, we should not be asking "how does literature refer to the real," Hamon insists, but "how does literature make us believe that it refers to the real" (14, my translation and emphasis). Hamon's own reformulation of the relationship between literary text, referent, and "reality" allows him to investigate an entire area of inquiry related to literary referring; that is, referring as an instance of persuasive discourse of a contractual, fiduciary nature. In the literary text, reference is a type of faire croire, a pact whereby the act of believing is "an activity of continual verification and interpretation" (16, my translation) by the reader. Hamon's theory, which immediately brings to mind Greimas' theory of modalities (see Greimas 1983:67-91), might readily be applied to certain texts, but it would require further precision in order to adequately account for certain postmodern texts (such as those analyzed in Lamontagne's article; see below), where the instability of the act of referring itself forms one of the main textual referents.
The conclusion to Hamon's article, in which he evokes the inevitable interplay of textual, intratextual, extratextual, and metatextual referential contexts, provides a fitting point of entry into Whiteside-St. Leger Lucas' study. Extremely well versed in both the Anglo-Saxon and Continental theoretical traditions, Whiteside's work (here and elsewhere) constitutes the most systematic and complete formulation of a general theory of literary reference. Reference is described as an interaction between three basic, intersecting contexts (the utterer's, the receiver's, and the encyclopaedic context) and four textual spheres (which form a secondary set of contexts): the extra-, inter-, intra-, and metatextual contexts. These two sets of contexts had previously been proposed in Whiteside's article "Conclusion: Theories of Reference" (1987:175-204), but the critic refines and elaborates her theory further here than in the earlier publication. First of all, the relationship between énonciation and reference is more clearly delineated, as the act of uttering is shown to give rise to the three basic contexts (111). Secondly, the discussion of the various contexts is more detailed, to account for the complexity and elasticity of each. In addition, the theory of the four contextual spheres is more developed, as Whiteside divides the four spheres into two groups: endophoric and exophoric reference. Whereas endophoric reference consists of intratextual reference (in which anaphoric and cataphoric processes are crucial) and of metatextual reference within the same text (i.e. , when the text refers to itself by metacom-mentary), exophoric reference regroups both intertextual reference and a second type of metatextual reference, in which one text refers to another, different text, as in literary criticism or parody. Whiteside then illustrates the functioning of these contexts and spheres in a study of selected Apollinaire texts (taken from Le bestiaire and Calligrammes), where the intricate interrelations of verbal and visual texts highlights the subordination of the ontological status of the referent.
Three authors contribute analyses of referring in specific literary texts. In "Le livre et le monde: la référence intertextuelle chez Jorge Luis Borges" (19-31), André Lamontagne explores the (re-)textualization of the referent in two short stories by Borges, in which aesthetic value triumphs over any possible truth value that the referent may possess. For Lamontagne, this collapse of the typical referring process heralds the arrival of the postmodern épistémè. In "Comique et référence: le Paon de Jules Renard et le Canard d'Alphonse Allais" (81-93), Daniel Grojnowski investigates the particularities of reference in the comic text, stressing the importance of certain references to extratextual events and circumstances, the absence of which would entail the loss of comic force in the texts by Renard and Allais. The key role of extratextual reference also characterizes Bret Easton Ellis' novel American Psycho, as Frances Fortier notes in her article "L'esthétique hyperréaliste de Bret Easton Ellis" (94-105). The endless lists of articles, clothing, activities, and gadgets in this novel result in a multiplication of reality effects, enhanced by the ubiquity of quotations from newspapers, television, graffiti, signs, and films. While the latter two articles offer pertinent studies of contrasting processes of referring in two very different types of literary texts, the theoretical framework is somewhat implicit, if not subordinated to the analyses of textual details and examples.
Two articles deal with forms of referring in the metatextual discourse of literary criticism. In "Histoire de seconde main ou les auteurs québécois de Maria Chapdelaine" (32-51), Nicole Fortin considers the reception of this classic French Canadian novel written by Louis Hémon, a French author. This is a situation of conflicting origins and of two referential systems (one national, the other foreign), which condition the novel's acceptance into (or possible rejection from) the Québec literary tradition. Fortin chooses five summaries of the novel, written between 1916 and 1980 by various literary critics, in order to examine how the Québec referent of the novel has been reconstructed, i. e. , how the novel, despite the French origin of its author, has been reinscribed and reappropriated by the Québec literary canon. Also working within the context of Québec literary criticism, Elisabeth Nardout-Lafarge analyzes the figure of antonomasia in Berthelot Brunet's 1946 publication, L'histoire de la littérature canadienne française. In "Référence littéraire et cliché critique: l'antonomase du nom d'auteur chez Berthelot Brunet" (56-65), Nardout-Lafarge illustrates how Brunet's use of authors' names reveals a particular, subjective conception of literature, for these proper names are invested with a conceptual content that conveys an "implicit typology in which authors and texts constitute categories or programmes" (58, my translation). In Brunet's case, proper names of authors and titles of texts form the referents for covert value judgments, whereby the French author or text is postulated as the standard or model towards which the French Canadian author should strive. Although both Fortin's and Nardout-Lafarge's articles could have been more directly and explicitly situated with reference to theories of reference, they succeed in delineating the crucial role of reference in the metatextual discourse of literary criticism.
Max Roy's article, "La référence comme effet de lecture" (66-80), straddles the three types of articles contained in the issue, as it deals with the process of referring in two particular literary texts, in the discourse of literary criticism, and it also stipulates specific theoretical considerations relevant to reading and reference. Claiming that the relationship between reading and reference can be considered in terms of the literary tradition to which a specific text belongs, Roy demonstrates how critical judgments of a work determine the interpretation of its referents and, consequently, its reception by other readers. Using Ringuet's Trente arpents and Poulin's Les grandes marées, Roy outlines two "intellectual attitudes" of reading that form part of literary reference: 1) the analysis of the "lexical meaning of a text [which] first obeys a decoding, an identification of semes" (79, my translation), and 2) a transitive attitude, by which the reader seeks the "readability" of the text in the larger, extratextual context of cultural or literary categories, such as genre or linguistic and aesthetic models. Literary reference is thus at once a textual, co-textual, and contextual activity, due to the various passages between the two attitudes of reading.
Whereas Avni's Resistance to Reference concentrates principally on Anglo-Saxon philosophical theories of reference, to the detriment of recent developments in the Continental linguistic tradition (principally written in French), the opposite is true of most of the articles comprising the recent issue of Tangence on La référence littéraire. Indeed, only two texts (those written by Whiteside and Lamontagne) make reference to any of the numerous studies (written in English) that form part of the analytical school of reference. In this sense, the articles in Tangence complement Avni's book; and it is only in reading both publications that we are afforded a complete picture of contemporary research on reference. The importance of context, which has continually been emphasized in almost all discussions of reference, might well be said to determine one's own resistance to reference(s), as the routes of reference travelled by the critic are, at least in part, influenced by his or her own discursive and linguistic context.
Benveniste, Emile (1966, 1974) Problèmes de linguistique générale. Volumes I and II. Paris: Gallimard.
Cometti, Jean-Pierre (1991) "Littérature, langage et référence." Cahiers d'épistémologie 9111 (Publication du Groupe de recherche en Epistémologie comparée, Département de philosophie, Université du Québec à Montréal):1-27.
Coquet, Jean-Claude (1991) "Réalité et principe d'immanence." Langages 103:23-35.
Ducrot, Oswald and Todorov, Tzvetan (1972) Dictionnaire encyclopédique des sciences du langage. Paris: Editions du Seuil.
Fitch, Brian (1991) Reflections in the Mind's Eye: Reference and Its Problematization in Twentieth-Century French Fiction. Toronto: University of Toronto Press.
Genette, Gérard (1991) Fiction et diction. Paris: Editions du Seuil.
Goodman, Nelson (1981) "Routes of Reference." Critical Inquiry 8.1: 121-132.
Greimas, Algirdas Julien (1983) Du Sens II. Paris: Editions du Seuil.
Harshaw (Hrushovshi), Benjamin (1984) "Fictionality and Fields of Reference: Remarks on a Theoretical Framework." Poetics Today 5. 2:227-251.
Hewitt, Leah (1987) "Getting into the (Speech) Act: Autobiography as a Theory of Performance." SubStance 16. 1:32-44.
Jakobson, Roman (1963) Essais de linguistique générale. Paris: Editions de Minuit.
Jespersen, Otto (1922) Language. London: George Allen and Unwin Ltd.
Kerbrat-Orecchioni, Catherine (1980) L'Enonciation. De la subjectivité dans le langage. Paris: Armand Colin.
Lavis, George (1971) "Le texte littéraire, le référent, le réel." Cahiers d'analyse textuelle 13:8-22.
de Libera, Alain (1987) "Suppositio et inclusio dans les théories médiévales de la référence." In La référence. Danon-Boileau, Laurent and Alain de Libera (eds. ). Paris: Editions Ophrys:9-34.
Linsky, Leonard (1967) Referring. London/Atlantic Highlands, New Jersey: Routledge and Kegan Paul/Humanities Press.
Lyons, John (1977) Semantics. Volume I. Cambridge: Cambridge University Press.
Milner, Jean-Claude (1976) "Réflexions sur la référence." Langue française 30:63-73.
Pavel, Thomas (1988) Univers de la fiction. Paris: Editions du Seuil.
Pratt, Mary Louise (1977) Toward a Speech Act Theory of Literary Discourse. Bloomington: Indiana University Press.
Ricoeur, Paul (1975) La Métaphore vive. Paris: Editions du Seuil.
Riffaterre, Michael (1990) Fictional Truth. Baltimore: The Johns Hopkins University Press.
Rosier, Irène (1987) "Discussions médiévales sur la coréférence." In La référence. Danon-Boileau, Laurent and Alain de Libera (eds. ). Paris: Editions Ophrys:35-62.
Van Dijk, Teun A. (ed. ) (1976) Pragmatics of Language and Literature. Amsterdam: North Holland Publishing Company.
Whiteside, Anna (1987a) "Conclusion: Theories of Reference." In On Referring in Literature. Whiteside, Anna and Michael Issacharoff (eds. ). Bloomington: Indiana University Press:175-204.
--- (1987b) "The Double Bind: Self-Referring Poetry." In On Referring in Literature. Whiteside, Anna and Michael Issacharoff (eds.). Bloomington: Indiana University Press:14-32.
Whiteside-St. Leger Lucas, Anna (1990) "Apollinaire's Ideogrammes: Sound, Sense. . . and Visible Signs." Word and Image 6. 2:163-179.
--- (1991) "Framing Fictional Reference." Recherches sémiotiques/Semiotic Inquiry 11. 1:27-46.
--- (1994) "Dynamique de la référence en littérature: l'exemple de l'écriture visuo-verbale d'Apollinaire." Tangence 44:106-124.
Barbara Havercroft is Associate Professor of Literary Studies and Semiotics in the Département d'études littéraires at the Université du Québec à Montréal. The author of numerous articles (published in European, Québec, Canadian and American books and journals) on énonciation and contemporary literary texts, autobiographical discourse, as well as the literary encounter between feminism and postmodernism, she has recently completed a book entitled Oscillation and Subjectivity: Problems of "Enonciation" in Robbe-Grillet, Sarraute, and Johnson (forthcoming). The editor of RS/SI (Recherches sémiotiques/ Semiotic Inquiry), she is currently working on a book dealing with contemporary forms of autobiography.