Richard L. W.
Clarke LITS3304 Notes 03D
ROLAND BARTHES "INTRODUCTION TO THE STRUCTURALIST ANALYSIS OF NARRATIVES" (1966) Barthes, Roland. Introduction to the Structuralist Analysis of Narratives. Image Music - Text. Ed. and trans. Stephen Heath. New York: Hill and Wang, 1977. 79-124. Narrative is transhistorical and transcultural. The goal of narratology is, drawing upon Saussurean linguistics, to describe a "common model . . . to master the infinity of utterances [paroles] by describing the language [langue] of which they are the products and from which they can be generated" (80) precisely because no given example of narrative exists "without reference to an implicit system of units and rules" (81). The enormity of such an undertaking implies that such a process cannot be inductive and must necessarily be deductive--that is, the goal is to "devise a hypothetical model of description . . . and then gradually to work down from this model towards the different narrative species which at once conform to and depart from this model" (81). The model for this undertaking is the science of Linguistics which, however, confines itself to the study of the sentence. There is, however, a "homological relation" (83) between sentence and discourse and, indeed, all semiotic systems precisely because narrative is a long sentence filled with tenses, aspects, moods, persons/subjects/actants which, accordingly, lends itself to grammatical analysis: it is in this sense that the general langue of narrative is the subject of a linguistics of discourse, as it were. In other words, it is precisely because any discourse (note that Barthes is using this term in a way different from Bakhtin) is organised "as a set of sentences" (83) that it accordingly forms "another language" (83), the units, rules and grammar of which become the object of a second linguistics. Moreover, as Saussure proved, language is not merely an expressive vehicle that can be set aside once the content that it contains or expresses is ascertained: "it never ceases to accompany discourse, holding up to it the mirror of its own structure" (85). The focus of Narratology is, consequently, on the structure of narrative rather than its content, precisely because the one is never abstractable from the other. Linguistics reveals and describes the organisation essential to every system of meaning. The description of the sentence encompasses a number of levels (the phonetic, the phonological, the grammatical and the contextual), no level being able to produce meaning on its own. Each level has two types of relations: distributional (consisting of its relations to other units on the same level) and integrational (its relations to other levels). It is precisely because "narrative is not a simple sum of propositions" but is composed, rather, of a large amount of clasifiable "elements" (85) that Structural analysis aims "to distinguish several levels or instances of description and to place these instances within a hierarchical (integrationary) perspective" (86). Narrative, according to Todorov, possesses two levels: story (the "logic of actions and the `syntax' of characters" [87]) and discourse ("tenses, aspects and modes of the narrative" (87)). In his scheme, to "understand a narrative is not merely to follow the unfolding of the story, it is also to recognise" (87) its hierarchical construction, to "project the horizontal concatenation of the narrative `thread' onto an implicitly vertical axis" (87) (and thus to take into account both the syntagmatic and the paradigmatic dimensions of the narrative). Meaning is not to be found "`at the end' of narrative, it runs across it" (87). Barthes proposes to expand this into a three-tiered model consisting of i) functions, ii) actions (characters = actants), and iii) narration (this corresponds to discourse in Todorov). Functions: these are the "smallest narrative units" (87) that can be isolated, the functional criterion of which is their meanings: "it is the functional nature of certain segments of the story that makes them units" (87). Every detail has functional importance as a unit of content, that is, in terms of what rather than how it is said. Note, however, that a particular "constitutive signified" (90) may have any number of
Richard L. W. Clarke LITS3304 Notes 03D
different signifiers" (90). Functions do not necessarily coincide with paragraphs or the other traditional divisions of narrative, nor with psychological or characteriological analyses nor with traditional linguistic units (indeed, a given functional unit may be longer or shorter than a sentence). Each functional unit proceeds along two different axes: the distributional (or the properly functional) (syntagmatic/ metonymic, that is, linked to other units with which it forms a sequence of units) and the integrational (or the indicial) (paradigmatic/ metaphoric, that is, there is a vertical relation of a semantic order pertinent to each unit). Whereas the former is concerned with the sequence of actions, the latter is concerned with what Barthes terms conceptual indices. Each narrative is classified according to whether a `functional' or `indicial' structuration predominates (eg. folktales versus psychological novels respectively). The former, the functions proper as such, can be subdivided into cardinal nuclei (those "hinge-points of the narrative" which inaugurate or close an uncertainty, which are consecutive and consequential and which telescope logic and temporality) and catalysers which "fill in the narrative space separating the hinge functions" (93) and perform a parasitic chronological functionality in order to maintain contact beteen narrator and addressee. All narrative proceeds on th basis of the confusion of consecution and consequence. The latter, the indices, as it were, which perform a "continuous" (95) role extended over an episode or a whole work, may be subdivided into indices proper (these refer to character, feeling, atmosphere, philosophy, etc.) which consist of implicit signifieds that necessitate deciphering and informants that serve to identify and locate in time and space, to "authenticate the reality of the referent, to embed fiction in the real world" (96). Catalysers, indices and informants all expand upon the framework provided by the cardinal nuclei. There is a functional syntax to any narrative: the different units are strung together along the narrative syntagm according to a certain `grammar'. Indices and informants combine freely while catalysers imply the existence of cardinal nuclei. Each nucleus implies the existence of other cardinal nuclei which together provide the "framework of the narrative" (98) and this is the real focus of the narratologist. Narratology seeks to understand the atemporal logic (consequence) behind the temporality of the narrative (consecution): the "order of chronological succession is absorbed in an atemporal matrix structure" (98). The goal is to give a "structural description of the chronological illusion--it is for narrative logic to account for narrative time" because "time . . . only exists functionally, as an element of a semiotic system . . . `true' time being a `realist', referential illusion" (99). The narrative logic of the nuclei is analysable in three ways: in order to, according to Bremond, "reconstitute the syntax of human behaviour utilised in narrative" (100) by retracing the choices that face the characters at each point; in order to, according to Levi-Strauss and Jakobson, "demonstrate paradigmatic oppositions in the functions" (100); and, according to Todorov, to set the analysis at the level of the `actions' performed by the characters. These three axes do not account for all the cardinal nuclei. The nuclei are comprehensible not only by their importance in themselves but syntagmatically, that is, by the nature of their relation to other nuclei in a "sequence" (101), that is, as a "logical succession of nuclei bound together by a relation of solidarity: the sequence opens when one of its terms has no solitary antecedent and closes when another of its terms has no consequent" (101). Every logical succession of actions is graspable by the reader as a "nominal whole" (102), hence, the division of the narrative into a number of named sequences. Moreover, paradigmatically, any point within the sequence offers a whole host of potential alternatives, therefore the sequence itself "constitutes a new unit, ready to function as a simple term in another, more extensive sequence" (102). In other words, there is a syntax both "within" and "between" (103) the sequences to the point where the "terms from several sequences can easily be imbricated in one another" precisely because "sequences move in counterpoint" (103). (In episodic narratives (those where there may be no sequential relation between episodes), sequences may be linked at the level
Richard L. W. Clarke LITS3304 Notes 03D
of the actants, that is, the same character may appear in different episodes.) Actions: For Aristotle, action is more important than character: that is, the action of a play (the genre with which Aristotle is concerned), for example, does not exist to reveal character--the action exists for the interest excited by concatenation of the events displayed (as one event leads directly to another) rather than for psychological revelation. For psychological criticism, however, characters, as the embodiment of "psychological essences" (105), are more important than the action. Structuralism is opposed to the latter approach ever since Vladimir Propp's Morphology of the Folk Tale, and is reluctant to treat character as an essence in this sense. Propp reduced characters to a "simple typology based not on psychology but on the unity of the actions assigned them by the narrative" (105). The notion of character is indispensable to the intelligibility of actions but "`agents' can be neither described nor classified in terms of `persons'" (105). In short, Structuralism reclassifies character in terms of participation in the action rather than some intrinsic essence. According to Bremond, every character is the "hero of his own sequence" (105). According to Todorov, characters are analysable according to the three "base predicates" (106) of relationships in which they are involved: love, communication, help. According to Greimas, characters are classified not according to what they are but what they do. They are "actants" (106) in so far as they participate in `actions' divisible according to three main semantic axes: communication, desire/quest, ordeal. This participation is ordered according to couples: the actant is either the subject or the object of the actions in question, the donor or the receiver, a helper or an opponent. The term `actant' defines a class of characters and thus "can be filled by different actors" (107) per se. According to the "regulated transformations (replacements . . . )" (107) the actants perform the different roles, an `actantial typology of narrative' can be produced. The question also arises as to the hero or subject of narrative. Traditionally, criticism privileged either a single or dual chief protagonist(s). Barthes wants to retain this privileging but by understanding the actant(s) under the category of the grammatical and not the psychological (in other words, actants exist only on paper and do not refer to real human people). Such an approach would draw upon Benveniste's understanding of the pronoun as a category "defined in relation to the instance of discourse, not that of reality" (109). Pronouns, as Benveniste points out, such as `he' or `she' always refer to particular identifiable individuals defined by a particular use (parole) of language (langue) in contrast to proper names such as `James' or `Timothy' which can refer to anyone and is thus more vague. Names and pronouns in literature do not refer, evidently, to concrete individuals: they remain grammatical creations rather than creatures endowed with psychological essences. The characters (or rather actants), as a unit of actional level, find their intelligibility only when integrated in the third level of structural description: narration. Narration: Any narrative is an object of communication between narrator and narratee. Barthes is not interested in the narrator's motives nor their effect upon the reader but rather in the "code by which narrator and reader are signified throughout the narrative itself" (110). Evident bits of information are provided about the narrator (eg "I . . . ") which also necessarily comment on the reader's positionality precisely because there is "no sense in the narrator giving himself a piece of information" (110). The traditional `signs of narration' include the following notions: a) the narrative is considered as the "expression of an I external to it" (110-1); and b) the narrator is considered either as an "omniscient . . . impersonal, consciousness that tells the story from a superior point of view" (111) or the narrator is limited to what his/her characters can observe and know. In Barthes' view, the author is not the same thing as the `paper' narrator whose signs are "immanent to the narrative" (111) itself and which are thus "accessible to semiological analysis" (111). Structuralism does not assume that the author is a "full subject and the narrative the instrumental expression of that fulness" (111). In fact, "who speaks (in the narrative) is not who writes (in real
Richard L. W. Clarke LITS3304 Notes 03D
life)" (112). The signs by which the narrator signifies him/herself may be personal, (often consisting of the first person, and which emphasise the "hinc et nunc of the locutionary act" (112)), mixed, and apersonal (third person the traditional narrative mode). Even in the personal mode, however, the "psychological person . . . bears no relation to the linguistic person, the latter is never defined by state of mind, intentions or traits of character but only by its coded place in the discourse" (14). Today, narrative has been transposed from the constative to the performative plane (that is, the focus is less on what it denotes than how it does so). (According to Speech-Act theory, language does not only `refer to things', rather, language also performs actions.) The prevailing sense, in fact, is that "the meaning of an utterance is the very act by which it is uttered" (114). Here, to use Aristotle's terminology, lexis (the words themselves or the `form') is privileged over dianoia (thought or `content'). Much conetmporary literature is, in this light, transitive rather than descriptive, more interesting in itself than for what it refers to (i.e. something external to the text itself). Barthes argues that particular narrative situations articulate certain appropriate narrational signs. In fact, "every narrative situation is dependent on a `narrative situation', the set of protocols according to which the narrative is consumed" (16): in oral literature, for example, there are specific codes of recitation; in other genres, there are particular modes of authorial intervention (the author is absent in plays, present in (lyric) poetry or mixed (sometimes present and sometimes absent in epic poetry); elsewhere, there are devices that feign `natural' circumstances, there are codes of beginning and ending, and there are different styles of representation (various points of view, etc.). All language consists of syntagmatic and paradigmatic dimensions, what might be called, in the old and obsolete scheme of things, form and meaning, as it were. In the former, there is an "articulation, or segmentation" of the succession of events "which produces units" (117), while in the latter, there is an "integration, which gathers these units into units of a higher rank" (117). Narrative is a highly "synthetic language" in that "each part . . . radiates in several directions at once" (118): it partakes of two dimensions: distortion--the extension of its signs over the entire length of the story (the units of one sequence are separated by the intrusion of another sequence)--and expansion--which inserts unforseeable expansions into these distortions. Two units which are contiguous from a realistic point of view, may be separated by the insertion of other different functional units which leads to the establishment of a logical time out of sorts with real time (eg the creation of suspense). These separations may be filled with catalysers. (Narrative, by its very nature, is elliptical and can be identified even when reduced to its principal actants and functional units. It lends itself to summary and is more paraphraseable than poetry.) The meaning of a given narrative resides in the intersection of the paradigmatic function of integration with its syntagmatic progression: "each level . . . gives its isotopy to the units of the level below" (122). In short, narrative is a "concourse" (122) of syntagmatic (horizontal) and paradigmatic (vertical) dimensions. The freedom of narrative is "literally hemmed in between the powerful code of . . . langue . . . and the powerful code of narrative" (113). There is no ascending order of freedom in that the phonematic level is the freest, the sentence less free and the sequence of actions which constitute the narrative the least free. Narrative analysis does not venture beyond the realm of discourse because "beyond the narrational level begins the world, other systems (social, economic, ideological)" (115) each of which necessitates "another semiotics" (116). Too often, narrative is naturalised, the play of the signifier suppressed because bourgeois society and its culture "demand signs which do not look like signs" (116). In the final analysis, realistic imitation is secondary in importance to the information pertinent to the development of the plot: the "function is not to `represent'" (123). The "origin of a sequence is not the observation of reality, but the need to transcend . . . repetition-`what takes place' in a narrative is from the referential . . . point of view literally nothing; `what happens' is language alone . . . the unceasing celebration of its coming"
Richard L. W. Clarke LITS3304 Notes 03D
(124). Drawing upon Freud, Levi-Strauss, and Lacan's rereading of Freud in the light of Levi-Strauss, Barthes draws a link between the origin of narrative, the origin of the sentence and the acquisition of language during the Oedipal complex at around the age of three.
