The Second Person: A Point of View?
The Function of the Second-Person Pronoun in Narrative Prose Fiction.

1. Framing Person

Before I explore arguments concerning the "second person's" potential to radicalise narrative discourse (and more) in later chapters, it is necessary to look further at how that potential is contained, for certainly, not all texts that employ a "second-person" narrative modality realise any such promise. In the reading of "second-person" narratives - indeed, of course, in reading any narrative fiction, and in reading any text as narrative, as Heider and Simmel's animation demonstrates - conventional practice works powerfully to constrain meaning through very many means. Not the least of these means is the provision of a large number of unexceptional literary and "natural" discourse models that employ the "second-person" pronoun, and the figuring within those models of a limited range of proper or canonical relations between story-tellers, audiences and objects. As Jonathan Culler observes, whenever we read narrative there is a need for "the strange, the formal, the fictional [to be] recuperated or naturalised, brought within our ken, if we do not want to remain gaping before monumental inscriptions" (Culler, 1975: 134). In our habits of reading, then, we will bring to bear on these sometimes baffling, even unnatural texts whatever hermeneutic frames, whatever interpretive keys, come to hand, casting around for ways of making "second-person" utterances in narrative discourse seem familiar. Amongst the most efficacious models for "second-person" discourse one can identify the maxim and aphorism, didactic and hortatory address, the language of the courtroom, confessional and speculative discourse, dramatic monologue, internal dialogue (i.e., talking to yourself), classical apostrophe, the epistle and diary, the guide book, skaz- and yarn-styled texts, the self-help and do-it-yourself manual, the choose-your-own adventure story, each one tending to organise its participants within particular prescribed relationships of knowledge, authority, status and civility, and so on. To that extent, literary critical practices depend for the sense they make of texts on the metaphors of "person" and on the relationships they assume and describe between "the reader," "the narrator" and so on. This aptly describes the work done by the four addressee identity figures proposed in the Chapter 1. Those four models show conventionalising operations at work, facilitating interpretation by framing a "you" utterance as forming a particular mode of address. The inescapably dialogic and allocutionary character of much second-person narrative fiction has ensured that most extended discussions are informed in some measure by an extrapolation of this model and its limited (and limiting) range of proper relations between anthropomorphised speakers and listeners.

Culler would point out, moreover, that such conventionalising operations are for the most part transparent - that is, "naturalised." On the whole, for instance, the reader will quickly recognise the basic mesofictional nature of the relationships of Barnes's Talking it Over (1991) and "The Stowaway" (1989), finding neither text in and of itself particularly confrontational (if the reader notices such instances as "non-canonical" at all). But just who, precisely, is being addressed as "you" in each case?in one by a circle of monologists clearly aware of one another's recriminating orations but each speaking to the addressee as if alone, and in the other by a woodworm. Similarly, the first two chapters of Bill Manhire's book The Brain of Katherine Mansfield (1988) immediately nominate a particular interpretive approach.

Chapter 1.

You are just an ordinary New Zealander. You have strength, intelligence and luck, though you are not particularly good at languages. Your family and friends like you, and there is one special friend who really thinks you're swell. Yours is a well-rounded personality; your horoscope is usually good; your school report says "satisfactory". But somehow you are restless. Your life is missing challenge and excitement. You want to make things happen. Go to 2.

Chapter 2.

On your way home from school one day, you find an old man waiting outside your house. He is holding a leather-bound book. He looks as if he has been expecting you.

"I have been reading your story," he says. "But it seems to have stopped. Something seems to happen when you enter the house."

He goes on to explain that he is eager to know how your life will continue. In fact, he says, your life is essentially an unwritten story. You yourself are the hero of the story.

"Many are the choices you must face, but the outcome of the tale will depend on you alone."

You stare at the stranger speechless, but your heart is beating with excitement. Dimly, as at a distance, you hear him say that you will need more than human help on the adventures which await you: you must choose one of three magic weapons to take with you on your journey. But you will have to come with him to his house. It lies in a distant suburb of the city.

It is getting late. The dark clouds of a winter afternoon swoop down over the familiar hills and houses. You shiver. The time of your first decision is upon you. Do you dare to turn the pages of adventure?

If you decide to accompany the old man, go to 5. If you decide to go home and think it over, go to 11. (Manhire, 1988: n.pag.)

Manhire's book draws on a genre that has found its niche in fiction for young adults: the "choose-your-own-adventure" story. You, Gentle Reader, are to find your own pathway through the text from the alternatives offered at the end of each chapter. The germane feature of this genre is the way its hero or heroine will be typically designated by the "second-person" pronoun. It also draws something from speculative or hypothetical discourse of the type Margolin describes as modelling "alternative life stories" in which "had you done X you would have been Y" (Margolin, 1997/1991: 7). Thirdly, the style of the very short first chapter at least is familiar to us from magazine and television advertisements that directly address the customer in an unsubtly interpellative bid to persuade him or her that s/he is the sort of person who would enjoy or who deserves the nominated product or service. The Brain of Katherine Mansfield becomes a little less baffling because we find that we are familiar with its habits, we have met it before - and better yet, it is a game. Because of our familiarity with the genre and our delight with its ludic qualities we are less resistant to any work being done by the pronoun. This work, first, is to facilitate the reader's conceit in thinking that she or he does choose the path of the adventure, the illusion - the role-play illusion - that he or she is the person upon whose wisdom and fortune rests the final outcome. Secondly, the work of the pronoun is to help the reader identify more closely with the protagonist. The "you" of its opening lines is an instance of the generic-"you" case, a generalised-"you" with which any average, urban New Zealander might identify, if only until the pronoun shifts to a more fully particularised narrative-"you," and perhaps even then. As incontrovertibly as Manhire's "you" does become a discrete protagonist, it retains some of its generalising force: the protagonist remains without distinguishing physical features or marked behavioural characteristics, without vocation, without a singular history, and beyond nationality without (explicit) gender, class or racial identity. In spite of this insistence of the general, our knowledge of genre and the way in which we construe the relationships between the narrator, protagonist and reader has ensured that making sense of the text at least on the denotative or surface level has been a relatively straight forward task. The work of naturalisation, however, is far from exhausted.

I have already proposed that categories of narrative "person" (and the metaphors of "person" and point of view more broadly) so thoroughly inform the social significance and signification of narrative fiction texts as to be inseparable from contemporary reading practices without partially or even wholly relinquishing some degree of their social, thematic, emotional or even ethical meaning. Like the critic David Lodge in his discussion of the broadly discredited (but yet deeply influential) mimetic relation between literary discourse and the world, I am mindful that in applying the broad assumptions of a post-structuralist critique to narrative fiction "I am not seeking to deny or sever [narrative fiction's] connection with 'life,' but merely asking that the crucial role of language . . . be adequately recognised" (Lodge, 1967: 169).

Arguments concerning the constitutive roles of language and the relations between "the world" and "the literary text" might now extend far beyond Lodge's 1967 critique in "Towards a Poetics of Fiction," but his qualified deconstruction of mimesis shares at least one assumption with my own arguments. This is that, put crudely, as social subjects, we readers tend to make meaning of narrative texts"readerly" narrative texts, at the very least - by relating to them as if they are about (fictional) people who more or less resemble ourselves and who inhabit (possible) worlds that resemble the one we ourselves inhabit/experience. It would appear that characters and narrators can and do take the form of virtual people as we read literary texts and we do make judgements about these virtual people in the light of our own experiences. The image of a character's identity that emerges from the sets of attributes accruing around the proper noun or pronoun that designates the character is permitted a continuity within the world of the text quite like the continuity of identity and existence we (guided by Cartesianism) construe for ourselves, the two images of identity simultaneously and mutually supporting, "mirroring," one another. Says one critic:

It is because fictional people, places and events exist in a manner which bears comparison with the independent objective status of real people, places and events, that fictional people, places and events do not cease to exist the moment they are no longer presented in the particular word or phrase or sentence or paragraph that we are actually reading. (Harland, 1995: 93).

Having allowed the character what in Freudian terms might be referred to as "object permanence," we then psychologise, socialise, and otherwise motivate our character. As Richard Harland explains, the way a reader understands Hindley's self-destructive behaviour in Wuthering Heights, for instance, is that he or she "draws upon default assumptions carried over from his/her experience in the real world - assumptions about physical causality and human psychology" (Harland, 1995: 94). Even if one leaves aside the vexed question of precisely in what manner fictional people, places and events bear comparison with real people, places and events (for Harland, this relation of mimesis seems permitted the status of a critical given), what does need to be addressed is the notion of the "independent objective status" of real people. What is at issue at this point is not only the way meaning is made, but the values placed on such meaning - and the transparency of those values. Critics who sponsor mimetic approaches would agree that of course no-one believes these people are actually real in any way, readers merely construe them as such, construe them "as if." As Ross Chambers demonstrates in Story and Situation (1978), however, quite beside the seeming inevitability of construing literary characters as people like ourselves inhabiting worlds like our own, there is nothing per se within literary theory that precludes a self-critical re-interrogation of such an approach. Without interrogation, the metaphor of "person" and its language of word-world correspondence and verbs of representation (reflect, present, mirror, depict, portray, illustrate) naturalise themselves as the proper, common-sensical way of speaking critically about literary character. Speaking about Judith Guest's novel Ordinary People (1976), for instance, Deborah Appleman writes:

Developing an integrated sense of identity is a daunting task, one made even more difficult by our confused and confusing times. That search for self-definition is the focus of Conrad's psychological odyssey as he attempts to derive "integrated" self-definition after the layers of sibling rivalry, parental expectations, and peer-imposed roles have been pulled away.

Conrad's struggle to define himself within the context of his family, his peers, and his personal history, mirror with remarkable accuracy the process of self-finding faced by many adolescents. The fact that both Conrad and the adolescent reader are immersed in the same difficult and troubled search for self will facilitate the students' ability to identify with and respond to the text. (Appleman, 1992: 93-94)

Conrad's search for an "integrated" self-definition, we are told, is a search that adolescent readers of the novel can identify with and share in. The character's search is for a more authentic sense of himself as a person beneath the onion-skin layers "of artifice that have surrounded his life and the life of his family" (Appleman, 1992: 94). For Appleman, it is not only that we might interpret Ordinary People in the light of our own experience - a proper claim to make, I believe - but that the novel "mirrors" adolescent readers' (apparently homogeneous) experience. The Cartesianism underpinning Appleman's discussion - its positing of an ethical individual responsible for his or her own behaviours; indeed, its assumption that any authentic self will be found at the heart of the onion - emerges even more tellingly from the responses to the novel she reports from her students. "[A]sked to write several statements of meaning that arose from their transaction with Ordinary People," they wrote:

"Most answers to problems are personal."

"I have to learn to deal with my feelings and face them better."

"Everybody needs to feel normal, to be accepted."

"You can't hide from life and its problems."

"Everyone has problems, even 'ordinary people.' The only difference is in the way people deal with their problems." (Appleman, 1992: 97)

This ethical, knowing self - a self-identity posited both for the character within the text and for the reader responding outside it - generates questions that presume that the characters' experiences are universally "readable" or transparent to all competent readers within the story's assumed readership. 72 Thus, Ogle Burks Duff will ask the reader in the classroom how your character thinks, acts, and feels as a result of being an African American as well as a human being (Duff, 1992: 209). Transparent and perhaps even inevitable as the question and its responses may appear, both are far from benign. Asked in a particular context - the ideological context of Cartesianism - the question produces answers appropriate to that context. Duff's proposition that white, female (and male) readers, "through vicarious experiences in literature using reader-response theory, . . . will have the opportunity to learn and feel what it is like to be an African American" (Duff, 1992: 208), read again in terms of both postcolonial or race theory, and feminist or gender theory, may be seen to be problematical, and a practice that addresses the pedagogical and ideological - and expressly hegemonic - needs of western classrooms rather than fully critical and self-critical reading.

It seems, therefore, that there is a learnt but perhaps inescapable recourse to "as if." As Jerome Bruner observes, from early childhood we habitually, as behaviour that is socially circumscribed, look for meaningful relations between the worlds we "experience" or "inhabit" through listening to or reading fiction and the experiential world/s we exist within (leaving completely in suspension, if only for a moment, the issues of what constitutes our knowledge or sense of that experience and of "the world," our identity, and so on).

Once young children come to grasp the basic idea of reference necessary for any language - that is, once they can name, can note recurrence, and can register termination of existence - their principle linguistic interest centres on human action and its outcomes, particularly human interactions. . . . People and their actions dominate the child's interest and attention. This is the first requirement of narrative. (Bruner, 1990: 78)

Given dominant reading practice and regimes of meaning-making, I take this situation to be in a certain sense inevitable: (to be invited) to say what a narrative text "is about," in the most ordinary and "pre-critical" way, is (to be invited) to paraphrase a tale in which somebody does something somewhere sometime, and to express opinions about the story-teller's insight, probity, and veracity.

As I submit above, therefore, before I explore arguments concerning the "second person's" potential to radicalise narrative discourse, I should first look at how that potential is contained. I could invoke a number of notions about framing and intertextuality in this respect, among which I would include Bakhtin's conception of the speech-genre as well as issues already explored such as the anthropocentric imperative in narrative, the notion of voice, narratology's geometric imaginary, and Cartesianism itself. I will look at the role of mimesis (and its assumptions about worlds and texts), arguing that one might best think of mimesis as providing reading one of its most potent and robust hermeneutic frames, and that this frame is one enabling condition for - and in turn is powerfully articulated by - the metaphors of "person" and point of view. First, however, I will now look at the more general issue of naturalisation, drawing specifically on Culler's discussion of naturalisation in terms of the vraisemblable.

2. The Vraisemblable and the Uses of Genre

What we speak of as conventions of a genre, writes Culler in On Deconstruction (1989), "are essentially possibilities of meaning," practices through which readers can make the text seem to fit the world naturally - or if not naturally, then at least tolerably (Culler, 1989: 137).

To assimilate or interpret something is to bring it within the modes of order which culture makes available, and this is usually done by talking about it in a mode of discourse which a culture takes as natural. (Culler, 1989: 137)

Whether one calls this process recuperation, naturalisation, motivation, or vraisemblisation, he concludes, "it is one of the basic activities of the mind" (Culler, 1989: 138). Although structuralist criticism has tended to figure naturalisation "as a bad thing," as he points out, it is nonetheless "an inevitable function of reading" (Culler, 1989: 159). To naturalise a text, Culler says, is to do nothing so much as

bring it into relation with a type of discourse or model which is already, in some sense, natural and legible. Some of these models have nothing specifically literary about them but are simply the repository of the vraisemblable, whereas others are special conventions used in naturalisation of literary works. (Culler, 1989: 138)

In speaking about interpretive practice in this way, it has been said, Culler insightfully stresses the positive, constructive force of naturalising strategies, as against the negative connotations of "recuperation" and "naturalisation" as circulated, for example, within much Materialist Marxist literary criticism which stresses respectively "the notion of recovery, of putting to use," and "the fact that the strange or deviant is brought within a discursive order and thus made to seem natural" (Culler, 1989: 137). By construing modes of naturalisation in terms of vraisemblisation, Culler allows us to see them as enabling, even empowering - as practices that provide us with the means to make sense of the narrative texts we come to lest they stand before us as intimidations, as wonders.

If indeed criticism might benefit more by regarding naturalisation as enabling than casting it as pre-eminently recuperative or deterministic, though, the case remains that processes of naturalisation, however conceived, always originate from and return the reader to particular ideological/discursive structures. As Christopher Prendergast puts it, vraisemblance is

a system of conventions and expectations which rests on/reinforces that more general system of "mutual knowledge" produced within a community for the realisation and maintenance of a whole social world. (Prendergast, 1986: 51)

All of the forms of the vraisemblable are (always-)already instituted within social, cultural relations or meaning, such that what vraisemblance describes is the way we fit the inscriptions we read - that is, the way in which we naturalise them - into those given (cultural and social) forms.

The vraisemblance of the fictional text is assured by virtue of the reader and writer engaging in a "contract of mutual recognition," a contract of meaning which meshes into that general network of agreements from which a society demarcates boundaries between sense and nonsense, the typical and the anomalous, the normal and the abnormal. (Hill, 1976: 336)

To offer a disarmingly reductive metaphor, these habits, in a sense, involve a fitting of appropriately shaped pegs to appropriately shaped holes, the pegs and holes given - remembering of course, as any child's play shows, there is tremendous enjoyment and expression to be had as well as frustration in the play of deferral and mismatch and of forcing new syntheses between shapes and spaces. Clearly, some moments of the vraisemblable will be more thoroughly inscribed by a culture's dominant ideologies than others, just as some utterances or modes of behaviour will be. When interpretation (re)turns us toward any of these - whenever, for instance, we read the inscription "I" and see in it Cartesianism's (anthropomorphised, deeply psychologised) grinning face, or read that you are an ordinary, well-rounded New Zealander who has intelligence and luck and usually a good horoscope, though no particular skill in a second language, and so know yourself and your heritage to be European-centred rather than Pacific Islander-centred - then the work of hegemony might be said to have been done. The suggestiveness of Culler's description of vraisemblance lies in its avoidance of the pessimism attending many discussions of naturalisation and of the rigidly deterministic and nominalistic practices of reading that those discussions can imply. Although social/cultural relations are instrumental in limiting the meaning available in a text, far from being repressively deterministic, described in terms of vraisemblance, the way in which we make sense is absolutely resilient. As Culler shows,

we can always make the meaningless meaningful by production of an appropriate context. . . . Certain dislocations in poetic texts can be read as signs of a prophetic or ecstatic state or as indications of a Rimbaldian 'dérèglement de tous les sens'. To place the text in such frameworks is to make it legible and intelligible. When Eliot says that modern poetry must be difficult because of the discontinuities of modern culture, when William Carlos Williams argues that his varied foot is necessary in a post-Einsteinian world where all order is questioned, when Humpty-Dumpty tells Alice that 'slithy' means 'lithe' and 'slimy', all are engaged in recuperation or naturalisation. (Culler, 1975: 138)

Culler divides naturalisation or vraisemblance 73 into five broad forms, "five ways in which a text may be brought into contact with and defined in relation to another text which helps make it intelligible" (Culler, 1975: 140).74 First, there is "the socially given text, that which is taken as the 'real world'" (Culler, 1975: 140). Works that use this first level of vraisemblance - and it is difficult to conceive of a narrative text that would not use it to some degree - are to that extent inherently intelligible, writes Culler. When they deviate from the text of the "real," the reader's tendency is to translate the deviation - which might be construed, for instance, as symbolism or a metaphor or some other form of trope - back into this "natural" way of speaking (Culler, 1975: 141).

The four forms of the addressee identity model, for instance, should be understood as belonging to this first level of the vraisemblable. As such, its terms draw both sense and authority from "the text of the natural attitude of a society (the text of l'habitude), entirely familiar and in this very familiarity diffuse, unknown as text" (Heath, cited in Culler, 1975: 140). On one point in particular, therefore, Culler is unequivocal. This order of vraisemblance "is what we should today call an ideology: 'a body of maxims and prejudices which constitute both a vision of the world and a system of values'" (Culler, 1975: 144). 75 To be defined as "a discourse which requires no justification because it seems to derive directly from the structures of the world" (Culler, 1975: 140), the "text of l'habitude," expresses everywhere as natural, for instance, Cartesianism's particular conception of being and being-in-the-world that underpins the addressee model.

We speak of people as having minds and bodies, as thinking, imagining, remembering, feeling pain, loving and hating, etc., and do not have to justify such discourse by adducing philosophical arguments. It is simply the text of the natural attitude, at least in Western culture, and hence vraisemblable. (Culler, 1975: 140-41)

The second form of vraisemblance, sometimes indistinguishable from the first, "is a range of cultural stereotypes or accepted knowledge which a work may use but which do not enjoy the same privileged status as elements of the first type, in that the culture itself recognises them as generalisations" (Culler, 1975: 141). Culler describes Balzac's description of the Count of Lanty as employing both of these levels. The Count is "small, ugly and pock-marked, as gloomy as a Spaniard and as boring as a banker" (cited in Culler, 1975: 141).

The adjectives are intelligible as qualities which it is quite natural and possible for someone to possess (whereas "he was small, green and demographic" would violate this first-order vraisemblance and require us to construct a very curious world indeed). The two comparisons, however, involve cultural references and stereotypes which are accepted as vraisemblable within the culture ("as gloomy as an Italian" and "as boring as a painter" would be invraisemblable in these terms) but which are still open to question: a banker need not be boring, and we accept that possibility along with the stereotype. (Culler, 1975: 141-42)

The social/cultural texts that constitute these two forms of vraisemblance, then, might also be conceived of as circulating as ideology and as Barthesian "myth," even as "common sense."

Culler's third form, alternatively, is "a specifically literary and artificial vraisemblance" comprised of "a set of literary norms to which texts may be related and by virtue of which they become meaningful and coherent" (Culler, 1975: 145). This artificial vraisemblance, of course, is the province of the conventional notion of genre. It is necessary, however, to extend this beyond the formally literary, because this level of vraisemblance also draws on non-literary and non-fictional models of discourse such as travel writing, letter writing, the cookbook, and so on. I would also include under Culler's third form of vraisemblance what Bakhtin means by speech genres, as ways of speaking that are appropriate to given conditions or to given contexts. The "second-person" modality commonly identified by critics as "courtroom-you," for example, finds its point, as it were, in oral rather than written discourse, and to that extent would escape a purely "literary" conception of this third from of vraisemblance.

The fourth form and fifth forms, on the other hand, are more explicitly literary. The fourth is what Culler calls "the natural attitude to the artificial, where the text explicitly cites and exposes vraisemblance of the third kind so as to reinforce its own authority" (Culler, 1975: 140). This involves the authorial conceit that the text does not follow literary convention. But of course, such claims themselves take literary forms.

The introductions to eighteenth century novels which explain how the diary or manuscript came into the narrator's possession, the use of external narrators who vouch for the truth of the tale told by another, are, of course, conventions in their own right. . . . (Culler, 1975: 148)

The fifth form of naturalisation, Culler says, "may be seen as a local and specialised variant of the fourth" (Culler, 1975: 152), and involves parodic or ironic intertextuality.

In calling something a parody we are specifying how it should be read, freeing ourselves from the demands of poetic seriousness, and making the curious features of the parody intelligible. The amazing alliteration, the thrusting anapaestic rhythm, the absence of content in Swinburne's self-parody, 'Nephelidia,' are immediately recuperated [naturalised] and given significance when we read it as parody: we read them as imitations and exaggerations of features of the original. (Culler, 1975: 152)

Culler's use of the concept of vraisemblance or naturalisation "to account for readers' interpretative strategies when confronting textual or semantic inconsistencies" has, as Fludernik says, a particular aptness to discussions of "second-person" narrative textuality, the "second person" seeming to foreground such processes (Fludernik, 1996: 20). Certainly, most commentators on "second-person" fiction have pointed in one respect or another to the ways in which readers draw heavily on given forms in their bids to make sense of these texts. Whereas my focus has tended to be on issues related to the first of Culler's levels of naturalisation, however, literary criticism for the most part has maintained its disciplinary bias towards the third level, that of generic vraisemblance. One of the points critics make most frequently, though, is that "second-person" fiction's genre models are not always themselves genres of literary fiction. Eric de Haard, for instance, in his discussion of the "you" modality of the first of Tolstoy's Sevastopol stories, notes that:

just as first- and third-person fictions have their counterparts - or can be considered modelled after - non-fictional genres and text types (memoirs, confession; report, historiography), so have second-person texts. For example, it is a device frequently employed in cookery books . . . and travel guides or tourist brochures. (de Haard, 1990: 266)

Likewise, Uri Margolin, in "Narrative 'You' Revisited" (1997/1991), observes that literary narrative "can reproduce or imitate a wide variety" of modalities, which he groups in terms of past, present and future tense. Past-tense "you" narratives can "be modelled on the eulogy or laudatio pattern but also on those of an argument for the prosecution, verdict, reminder or summary of performance" (Margolin, 1997/1991: 6). Present-tense narratives can "follow the model of an on-the-spot reporting of current ongoing activity, [or] use the more static stocktaking or character sketch patterns" (Margolin, 1997/1991: 6). Conditional or subjunctive modes, which he sees as a variety of hypothetical "you," are modelled after "alternative life stories," and future-tense and hypothetical modes, writes Margolin, are modelled after "forecasts, predictions, scenarios, warnings, and promises" (Margolin, 1997/1991: 7). Fludernik, too, provides a systematic discussion of what she sees as some of the principal prototypical discourse models for "second-person" textuality. She identifies eight, dividing them into two broad classes, the first group (conversational storytelling, skaz, letter writing, and dramatic monologue) involving models of discourse that "highlight a prominent address function" and the "second person's" communicative level. The second group (the guide book, courthouse, generalising, and self-addressed forms) involve models, she writes, in which the "second person's" address function has become latent or inactive, so that, instead, they will only "portray a you's experiences" rather than engage the "you" in any form of explicit or implicit communicative act (Fludernik, 1993: 230). Other critics who have explored the ways in which genre practice or framing can limit meanings in "second-person" texts include Morrissette (1965), Bonheim (1983), and Richardson (1991).

3. The Occurrence of Congruence

Within mimesis, it seems, certainly at the most general level, an equation has been drawn between language and its objects that has the reassuring, homely quality of the self-evident. It has made the common-sensical presumption that the mime must have an object. The penetrating authority of this equation and its apparent transparency has made the mimetic approach to analysis and criticism very appealing for its certainties: mimesis depends on a notion of what the world already is.

The authoritarian gesture of mimesis is to imprison us in a world which, by virtue of its familiarity, is closed to analysis and criticism, in which the "prescriptive" and the "normative" (themselves tacit) ensure that the "descriptive" remains at the level of the undiscussed, in the taken-for-grantedness of the familiar: mimesis deals in familiarities ("recognitions"), but the recognitions it supplies are often misrecognitions ("méconnaissances"). (Prendergast, 1986: 6)

Its traditional terms assume a relatively stable, self-sufficient and knowable world and some form of representational or reflective relation between the world and texts that is primary in processes of making meaning. Its terms also continue to hold faith in the human subject as the origin of his or her own thoughts and judgements and in this human subject's transparent access to the world through consciousness, such that the certainties of mimesis have come to owe no small debt to Cartesianism. To that extent, texts, paintings, sculptures, and so on become intermediary as objects inserted between autonomous human subjects and the world known and experienced by them. It is a way of thinking about the relation between language and the world that has reduced what Peirce describes as the essentially triadic nature of human semiosis and what Timothy Reiss refers to as the "Galilean trinity of mind/language/phenomenon" to a sober dichotomy in which

language reveals thought, and in so far as [language] refers to objects it can operate as a perfect stand-in for them. It is not, to be sure, the object itself; but it is conceived of as a sufficiently accurate representation for the purposes of discourse, into whose system it may be inserted. (Reiss, 1973: 9)

The conventionalising and naturalising work done by theories of mimesis in explaining relations between the world, our being in it, and texts is thus one further buttress by which the good standing of the metaphor of "person" is preserved in traditional and pre-critical modes of analysis. One might best think of mimesis (and its assumptions about worlds and texts) as providing a powerfully influential and governing hermeneutic frame, and that this frame is one enabling condition for - and in turn is powerfully articulated by - the metaphors of "person" and point of view. One premise underpinning much of the discussion that follows is that an argument that presents mimesis as in any way productive of meaning - which I necessarily do, given my recourse to narrative "person" as a signifying function - must not become a mere apologia for a mimetic aesthetics and poetics. It must, instead, situate mimesis as a term within contemporary conventional reading practice, and at the very least remove that term from its traditional position of transparent primacy and authority.

The need to retain some sense of mimesis as a category within explications of signifying practice, it appears on the other hand, is broadly accepted. The Yale deconstructionist Paul de Man argues persuasively for the retention of a retheorised notion of mimesis, observing:

It is impossible to conceive of phenomenal experience that would not be mimetic, as it is impossible to conceive of an aesthetic judgement that would not depend on imitation as a constitutive category, also and especially when the judgement . . . is interiorised as the consciousness of a subject. (de Man, 1986: 67)

Derrida himself is unequivocal on the place of mimesis within prevailing systems of meaning, suggesting that "the whole history of the interpretation of the arts of letters has moved and been transformed within diverse logical possibilities opened up by the concept of mimesis" (Derrida, 1981a: 187). Plato's conception of it, he writes, continues to mark the paradigm. Mimesis for Plato is firmly "lined up alongside truth" (Derrida, 1981a: 187), both negatively and positively. It either "hinders the unveiling of the thing itself by substituting a copy or double for what is; or else works in the service of truth through the double's resemblance [homoisosis]" (Derrida, 1981a: 187). Usually, though, mimesis is constituted as the former, as subversive and damaging to society. Derrida glosses this position, though, by arguing that although Platonism describes mimetic representations as disturbing the ideal hierarchy of things and so as "injurious," it also implies that mimesis must also inevitably draw attention to the vast capacity of the human mind for fabricating the systems - for producing discourse whose claim to referentiality can never, finally, shake off the proper residue of scepticism - under which men live (Prendergast, 1986: 12). And indeed, that it is also the very nature of mimesis to disavow this creativity, to naturalise the relation mimesis posits between the mime and the mimed - to disguise, that is, its own status as discourse, as fabrication. Derrida proposes that the whole history of literature and literary interpretation is lodged somewhere "in the strange mirror that reflects but also displaces and distorts one mimesis into the other, as though it were itself destined to mime or mask itself" (Derrida, 1981a: 191). He asks, faced with this prospect, what does "Platonism" decide?"Platonism" here standing "for the whole history of Western philosophy, including the anti-Platonisms that regularly feed into it" (Derrida, 1981a: 191).

What is it that is decided and maintained in ontology or dialectics throughout all the mutations or revolutions that are entailed [in the history of mimesis]? It is precisely the ontological: the presumed possibility of a discourse about what is. . . . It follows, I apologise for repeating this, that the image supervenes upon reality, the representation upon the present in presentation, the imitation upon the thing, the imitator upon the imitated. (Derrida, 1981a: 191)

The need to retain some sense of mimesis as a category is also addressed by Bernard Duyfhuizen. In his article "Mimesis, Authority, and Belief in Narrative Poetics" (1985), Duyfhuizen makes the observation that:

In a sense, there is a mimetic trap in criticism: to be considered "meaning-full," a reading must recount the story of a real reader's engagement with the text, and conventionally we tend to value that story as it clarifies our own mimetic assumptions. The epistemologies Western philosophy has devised to help us understand our world support valorisation of mimetic reading. (Duyfhuizen, 1985: 219)

Constrained by "the epistemologies [of] Western philosophy," and more pointedly by its ontologies, we find ourselves asking, have been taught to ask: where and when does this take place, what moral/ethical values are being thematised and to what ends, who does what to whom, and so on. De Man's reference to a judgement that is "interiorised as the consciousness of a subject," and Duyfhuizen's to the circular valorisation by dominant epistemologies of "our own mimetic assumptions," are crucial in this context, adverting as they do to the ideological work done by mimetic explanations of textual practice. Embedded in de Man's and Duyfhuizen's arguments is the understanding that the incorporation of any mimetic aesthetic or theory of textuality into critical analysis necessitates an acknowledgment of the work of mimesis in the maintenance of prevailing relations of power within social and cultural practices.

Bruner, too, posits mimesis as essential to narrative signification, but is discontent with the traditional Aristotelian notion of mimesis construed in its most crude (and influential) paraphrase as "art imitates life." His own sense of mimesis accords much more with that proposed earlier: that mimesis functions as a hermeneutic frame (with all that "framing" entails 76). Indeed, Bruner himself invokes the notion of framing or schematising. Although at this point he is addressing himself to concepts of narrative function, his argument is also clearly true of a mimetic function. He argues that frames provide for us a means of "constructing" a world, of characterising its flow, of segmenting events within that world, and so on. If we were not able to do such framing, we would be lost in a murk of chaotic experience and probably would not have survived as a species in any case.

The typical form of framing experience (and our memory of it) is in narrative form . . . and what does not get structured narratively suffers loss in memory. Framing pursues experience into memory, where, as we have known since the classic studies of Bartlett, it is systematically altered to conform to our canonical representations of the social world, or if it cannot be so altered, it is either forgotten or highlighted in its exceptionality. (Bruner, 1990: 56)

Bruner suggests that, rather than "art imitating life," Aristotelian mimesis might better be understood as "the capturing of 'life in action,' and elaboration and amelioration of what happened" (Bruner, 1990: 46). He likens this notion of mimesis as "capturing life in action" to Paul Ricoeur's description of mimesis as "a kind of metaphor of reality" (cited Bruner, 1990: 46). The "referentiality" that a concept of mimesis necessarily presumes is not to be taken as directed at "reality" or the world. Rather, mimetic referentiality should be understood as being of a kind that is proper to metaphor. And in so far as this might itself be accused of referring back to "reality," it can only do so, suggests Bruner, in the way that Peirce's interpretant refers to its object (Bruner, 1990: 46). In the notion of mimesis being put forward here, the true concern of any mimetic function, as a function of interpretance, can only be something akin to Peirce's immediate object, the object as the sign represents it, the only cognitive access we can have to the object. As for Ricoeur, mimesis can only produce a discourse (or artifact) that refers itself to reality "not in order to copy it, but in order to give it a new reading" (emphasis added, Bruner, 1990: 46). And as in any reading, as contemporary literary theory will tell us, and no matter our faith in the aptness of the reading or in the procedures by which we came to it, the relation between every act of reading and the object/text that is read can no longer be taken as innocent of ideology.

To clarify this notion of mimesis, it will be useful to look more closely at Duyfhuizen's argument. He proposes that the discipline of narratology, given that its heritage lies in literary formalism, structuralism and semiotics, has moved away from mimetic explanations of textual discourse. Although this move "has done much to improve formal and linguistic analysis of texts," he argues, it has also resulted in a tendency "to reduce texts to abstract formulae of 'deep structure,' which seemingly have little relation to how a text is consciously written or read - the structures discovered have significance but little meaning" (Duyfhuizen, 1985: 219). He then (more approvingly, it seems) identifies an entrenched mimesis within pragmatic and reader-based theories (a point that I will take up at length below), pointing out that the central term of such theories - the reader"relies on either a unifying mimesis or a theoretical abstraction, and therefore cannot be adequately defined" (Duyfhuizen, 1985: 220). He argues that reader-based theories tend to assume

a correspondence between the text and the real world. Fictions can only mean if they reflect, or are found to reflect, conventional experience, and if they solicit the reader's participation in concretising the fictional world and his or her belief in the "reality" depicted. (Duyfhuizen, 1985: 220)

Against the background of a deconstruction of traditional poetics, Duyfhuizen's strategy isn't to propose that critics eliminate concepts of mimesis from explications of the processes of fiction, but that they reconceive it whereby mimesis is to be seen as one element amongst others (and so no longer the primary element) constitutive of "a theory of narrative transmission and textuality" (Duyfhuizen, 1985: 222).

That Duyfhuizen does not discuss or problematise his assumptions about the nature of "conventional experience" or about "how a text is consciously written and read" (my emphasis) is significant here. In spite of his insight into the necessity to qualify and re-frame any theory of mimesis incorporated into contemporary critical analysis, his terms slide towards those in which mimesis itself is most comfortable. He himself enters the mimetic trap - and does so, it seems, willingly. While he argues against the traditional primacy attributed to mimesis in theories of narrative textuality and wants to "mak[e] problematic any poetics of fiction using mimesis as a basic, unquestioned premise" (Duyfhuizen, 1985: 219), Duyfhuizen would still seem to assume some degree of transparency between worlds and texts. His argument is essentially that mimesis is to be problematised because narrative meaning is not primarily constituted by a text's representation of a world and of persons mobile in that world: there is the reader's experience of the world, the reader's experience of the text (over time) and of criticism and/or prior readings of the text, and the text's (mimetic) relation to the world. This is not my own approach - indeed, I have no wish to "rehabilitate" mimesis, so to speak, as Duyfhuizen appears to. Rather, I would describe mimesis as a realm of ideology. This is what I intend by asserting that mimesis should be awarded no other status than that of hermeneutic frame. I would argue for an acknowledgment of the ways in which concepts of mimesis are articulated and played out in current pre-critical, dominant reading practices, and the ways in which the concepts and their assumptions assist in maintaining and reproducing those practices.

To conclude, there is a second, telling point of issue to be taken with Duyfhuizen's approach to the issue of mimesis. Duyfhuizen proposes that narratology has moved itself away from the mimetic assumptions of many other modes of literary analysis. I would suggest that it has not removed itself so very far. 77 As has been noted, narratology remains centrally concerned with explicating relationships between narrative agents. Consider, for instance, the central place in Genette's analysis of categories of mood, focalisation and so on. Indeed, one might justifiably identify narratology's primary shift as being that of moving the focus of analysis - remaking its object - from that of somebody does something somewhere sometime to somebody tells something to someone somewhere sometime. If narrative "content" or "story" becomes less significant beneath the formalism of some narratological explication, it is nonetheless always present by implication in that middle term: telling "something." Moreover, as Margolin has observed, whether explicated in the terms of discourse and story (Chatman), narration and the narrated (Prince), or discours and histoire (Benveniste), the work of narratology seems always to return to the human subject in one guise or another as an anchoring term of reference, and this subject "is either someone who is telling what happened and how things were/are, or he is someone who participates in the states of affairs and actions of which we are being told" (Margolin, 1986-87: 183). "Making sense" of narrative textuality is therefore a matter of identifying, describing and interrelating individuals in respect to "their existence, individual properties, uniqueness and continuity" (Margolin, 1986-87: 183) in the domain of the narration (ie. the narrator/s) and the domain of the narrated (ie. the narrative agents, the characters). That is, narratological exposition, at least at the level of the language used to explore its objects, assumes precisely what traditional and pre-critical analysis assumes: that narrative existents are individuated and unique and that they exhibit a stable, continuous identity. Whereas it might be claimed that the nature and ontological underpinnings of the subject presumed by traditional and narratological methodology respectively is quite different, the nature of the textual subject that both project is remarkable for its similarity. And it would appear that what underpins this similarity is their thorough-going faith in the "truth" of mimesis.

the skinny

top

Index

1

2

3

4

5

6

7

Notes

Bibliography

Links

Slackwire