The Second Person: A Point of View?
The Function of the Second-Person Pronoun in Narrative Prose Fiction.
Chapter 3
Radicalising Models and Functions
|
|
|
|
The Second Person: A Point of View? Chapter 3 Radicalising Models and Functions |
|
|
|
|
In the previous two chapters I have contextualised my discussion of "second-person" narrative as a critique of Cartesianism. In doing so, however, it is not my intention to advocate a sweeping dismissal of Cartesianism's terms. I have proposed that the anthropocentric categories of "person" are powerfully entrenched not only within literary criticism, but also within notions of subjectivity and identity. The issue to be put is the nature of those categories; the issue to be taken with Cartesianism is its naturalisation and institutionalisation of particular conceptions of "person," and the particular social and cultural relations that these conceptions circulate and sustain - relations that privilege the individual over the social. As I will argue more fully later, it is in some measure inevitable - for the time being at least - that we will speak amongst ourselves (and to ourselves) about "persons" and "voice" in narrative and about the behaviour of speaking subjects in texts, for to do otherwise would be to cease to speak about narrative as narrative. As Fludernik argues, narrative is quintessentially anthropocentric. Narrative is about "persons." How we speak about "person" and the speaking subject, then, becomes crucial. In this chapter, therefore, I want to explore ways of describing "second-person" modalities in terms that are explicitly self-conscious of the Cartesianism instilled within the practices they describe. The first is a general descriptive model that conceives of "second-person" modalities within a system of referentiality rather than - as with the paradigmatic model described by figures 1 to 4 - a system of address. I will also explore two key functions of the "second person" that are fundamentally concerned with the anthropocentric nature of "person." Both of these functions are elements of the figural, and so can exert effects on what I have already identified as narrative-"you" as well as narratee/reader addressed texts. The first of these functions is the generalising function, which is attached to the generic mode of "you;" the second is the address or allocutionary function. I will argue that the "second person" has a generic or generalising function that can be overridden (overwritten) by the individuation or particularisation of the "you" as either a reader/addressee figure or a protagonist; and that it has an address function whose allocutive force is diminished by the disengagement of the deictic/allocutionary function of the "second-person" pronoun from the pronoun's necessary grammatical form. I will show, moreover, that the generalising function and the allocutionary function, whilst being independent of one another, are most frequently concurrent and congruent. Another way of characterising "second-person" narrative employs as its benchmark the object of the "you" utterance, so that the pronoun may signify a protagonist (narrative-"you"), a recipient of direct address (there are two: metafictional-"you;" and what I will coin as mesofictional-"you"), or none of these (figural-"you"). The narrative and figural forms have been spoken about earlier, but will be more clearly defined here. The metafictional case is defined by its direct address to a (putative) reader; the mesofictional is defined by its direct address to a narratee. 51 However, unlike address models such as those proposed by Kacandes and Fludernik or within my four-part model above, the object of a "second-person" utterance is not conceived of here so explicitly as an object of address. Rather, it is to be regarded as an object of reference: as (merely) the object or objects referred to deictically by the "second-person" pronoun in any passage of text. [NOTE] Besides the unremarkable intradiegetic "you" (of characters addressing one another), there are four very broad classes of reference. Mapped over the narrator-character-narratee-reader nexus of the addressee model (on line b.), the cases of the reference model (on line a.) are:
Fig. 11. Reference Identity model There are three determining criteria for the reference model. The first concerns the force of the allocutionary function, and specifically the degree to which the "you" pronoun carries the grammatical, deictic function of the vocative case, in such a way that the "you" signals its referent to be an object of direct address. The second criterion concerns the degree to which the referent of the "you" is to be construed as a character/protagonist in a story. The third asks whether or not the object of the "you" utterance is to be understood as designating a reader implicated in processes beyond the fictional realm of the story - that is, as an entity self-conscious not only of his or her own activity as a reader, but also of separation from the text world. By setting these criteria out as a sequence, first through third, I do not mean to imply that they are successive or that one is in any way dependent or prior to another. It is the combination of the three criteria (and their absence as well as their presence) that produces the four broad classes. Three criticisms of such a schema can be made (and then set aside) quickly. The first is that these four terms are too general. Intended to encompass the entire field of the "second-person" pronoun in narrative beyond mere dialogue between characters, this generalness and breadth, however, is exactly the worth of these terms. The second is that this new reference model overlaps the address model, so that one or the other should surely fall redundant. The overlapping of models, however, is not only a consequence of differing approaches to textual analysis and narrative textuality, but also of the fluidity of the "second person" itself. "Second-person" prose narrative does not lend itself to unifying models and typologies. The formal, grammatical properties of the pronoun itself are such that "second-person" textuality will perforce be sensible to readers in terms of address, but an address model alone cannot account for the varieties in form and effect. Indeed, this need to regard the "second person" in the light of address and reference is intimated in David Herman's insights into deictic mechanisms in "second-person" texts. Although he makes no claim for the general applicability of his discussion, and circumspectly confines himself to commenting on Edna O'Brien's A Pagan Place (1984), Herman's observations are indeed useful in the broader exploration of "second-person" textuality (Herman, 1994: n.9 406). He argues that the "you" of A Pagan Place does carry the grammatical features of the "second-person" pronoun. [This] prima facie seems to encode the role of the addressee into narrative discourse, nonetheless . . . the [referential] deictic functions of you are in some instances only partly in agreement with the term's morphosyntactic [or grammatical] features. Functionally speaking you superimposes the deictic role of the audience or overhearer (in this instance the reader) onto the deictic role(s) spatiotemporally anchored in the fictional world elaborated over the course of the narrative. The grammatical profile of you thus drastically undermines its deictic functions; the text projects itself into a range of contexts that cannot be strictly delimited. (Herman, 1994: 390) Or rather, the "you" of A Pagan Place "projects itself into a range of contexts" that cannot be contained by a model constituted primarily by conventional, grammatical properties of address. Thirdly, it can be argued, properly enough, of course, that the terms "object of address" and "object of reference" are simply two terms that designate a single object. Certainly, whether the object of any particular "you" utterance is conceived of as an addressee or as a referent, it is one and the same object/identity. It would be hasty to assume from this, however, that one or the other term is therefore redundant. The terms "object of address" and "object of reference" are only superficially interchangeable. The address functions of the "second-person" clearly rely for their very sense on the referential, deictic functions of the "second-person" pronoun. That the shifter's function is substantially suspended in "second-person" narrative is, I suggest, illustrative of the point. Ordinarily, the shifter function requires that the "you" and "I" renew their referents in each utterance, dialogue being the paradigmatic case; but, as already shown, in "second-person" narrative, the "you" can acquire a fixity so that it comes to denote one protagonist more than any other. This, indeed, is a feature that Herman focuses on in his reconception of deixis as sociocentric rather than egocentric (see page 100 below). It is exactly the augmented referential function of the "you" in "second-person" narrative that stalls the shifter. It would appear, therefore, that a signifying function is a prior condition for any address function. Even an address to "nobody" (or nobody in particular) refers itself, exactly, to "no body" (in particular). For my purposes, the seizure of the deictic function of the "second-person" pronoun underscores an important distinction between the two paradigmatic approaches to characterising "second-person" narrative. The addressee identity approach sees the "second person" in terms of communicative circuits and functions of address, and so necessarily maintains the anthropomorphic metaphors of somebody addressing or not addressing, speaking to or not speaking to, somebody else. Communicative discourse models require entities between which the discourse is travelling: the metaphors themselves are the grounds upon which these models are raised, modelled on assumptions about "natural" circuits of communication. Alternatively, the object-as-referent notion approaches "second-person" narrative through the non-humanist terms of textual functions and structuralist linguistics/semiotics. To an extent, this system circumvents some of the still inevitable call to the metaphors of "person." Powerfully embedded deep within the system, of course, the metaphors do remain. All four classes have an anthropomorphic constituent, three of the classes explicitly so. 52 The fourth class, the figural, too, has recourse to a "person" as a constitutive element, albeit in a negative sense: negatively, in that the "you" referent" is defined by its difference from the other classes, as being not a reader, not an addressee, and not a character. But, if the referents remain anthropomorphic, the capacity to respond, to "speak," within a communicative circuit is denied them here - or rather, is made beside the point. The system remains advantageous beyond its inevitable constitutive anthropomorphism for its disruption of the circuits of communication so often used as definitive within models of "second-person" narrative. In a sense, these four classes borrow something of the functionality conventionally reserved for "third-person" pronominal forms, pronouns that Benveniste describes as non-personal precisely because they do not constitute in and of themselves any communicative circuit between "persons;" they don't merely designate, but actively exclude the object from a circuit of communication by reifying it as object.53 The value of this broad system is that it facilitates an account of the combination and fluidity of "second-person" modality in any passage of text. Its advantage can be illustrated by turning to David Herman's discussion of O'Brien's A Pagan Place, in which he makes a painstaking micro-analysis of the address mechanisisms of narrative you. Herman is at pains to demonstrate that any passage will carry within it a number of "formal features help[ing] us negotiate instances of you that deviate from what we might be inclined to call the default interpretation. . ." (Herman, 1994: 384). Take this passage, for example: "The squeels of each particular pig [being slaughtered by the father] reached you no matter where you hid, no matter where you happened to crouch, and it was heart-rending as if the pig was making a last but futile appeal to someone to save him" (19). Here, arguably, the chain of emphatic identification stretches beyond the diegetic situation of the novel--beyond that virtual "you" who, in O'Brien's fictional world, addresses herself comments regarding her own inability to ignore the pigs' final appeal--and reaches those fragments of our world(s) in which pity for pigs is actually to be found. . . . It seems that in descriptions like [this], textual you functions not (or not only) as discourse particle relaying and linking the various components of a fictional protagonist's self-address, but (also) as a form of address that exceeds the frame of the fiction itself. You designates anyone who has ever been or might conceivably be upset at the slaughter of animals. . . . (Herman, 1994: 385-86) Herman identifies a complex, micro-level shift in modality here, and suggests that the reason we take its meaning is that the discourse model of self-address--the "default interpretation"--is briefly superseded by a second discourse model, that of reader address. As he put it earlier, deixis has shifted transgressively from its anchor in a self-referential narrator in the fictional world to the reader in the (or an) actual world. But Herman's frame--and limit--remains determined by communicative models of address. The value of his observations about the fluidity of the "second person" at the micro level of narrative, I suggest, can be sharpened by the strategy I have been outlining: that of turning away (formally) from discourse models to a reference model. 54 What Herman identifies here can also be characterised as the emergence of a brief case of figural modality--the case I will define below as the highly generalised generic-"you"--within the "default" case of narrative-"you." This narrative-"you," moreover, to return to the conventional discourse models so as to complete Herman's picture, is the form modelled as narrator self-address. You, dogged, uninsatiable, print-oriented bastard, it's you I'm addressing, who else, from inside this monstrous fiction. You've read me so far, then? Even this far? For what discreditable motive? How is it you don't go to a movie, watch TV, stare at a wall, play tennis with a friend, make amorous advances to the person who comes to your mind when I speak of amorous advances? Can nothing surfeit, saturate you, turn you off? Where's your shame? (Barth, 1969: 123) I mean metafictional in a narrower, more exclusive sense than the sense it has taken over the last two decades. My intention in confining the sense of the term metafiction is not to redefine the term or its objects as explored by Patricia Waugh (1988), Robert Siegle (1986) and others, but to mark a special case of "second-person" functionality. Waugh defines metafiction as "a term given to fictional writing which self-consciously and systematically draws attention to its status as an artifact in order to pose questions about the relationship between fiction and reality" (emphasis added, Waugh, 1988: 2). She goes on to argue that: metafiction is not so much a sub-genre of the novel as a tendency within the novel which operates through exaggeration of the tensions and oppositions inherent in all novels: of frame and frame-break, of technique and counter-technique, of construction and deconstruction of illusion. . . . The expression of this tension is present in much contemporary writing but it is the dominant function in the texts defined as metafictional. (Waugh, 1988: 13-14) By Waugh's account, virtually all "second-person" narratives of the types I refer to as character-address and self-address, and a great number of those classed as reader-address, would (properly) be regarded as exhibiting metafictional and reflexive qualities. Indeed, it is the very nature of "second-person" narrative utterances, in their provocative gestures toward readers, to draw attention to their texts' "status as an artifact." In order to retain the usefulness of the term in this scheme, then, I find it necessary to restrict the object to which it refers. Metafictional-"you" involves a "you" that is implicated in processes beyond the fictional realm of the story. It designates an implied reader who knows that s/he is reading a work of fiction. 55 Furthermore, by its very definition, this class of the "second person" is ideally experienced by the actual reader as direct address. Its most typical form is illustrated by O.Henry's story "The Story of the Caliph Who Alleviated His Conscience," in which the narrator intrudes into the narrative to address an implied reader directly. Unlike the mesofictional-"you" of The Catcher in the Rye which disavows its fictitiousness, metafictional-"you" involves a consciousness of the narrative's status as fiction. In O.Henry's story, for instance, the narration breaks off after the first lines so that the narrator may address its readers directly. Thus, by the commonest article of trade, having gained your interest, the action of the story will now be suspended, leaving you grumpily to consider a sort of doll biography beginning fifteen years before. [ . . . ] There now! it's over. Hardly had time to yawn, did you? I've seen biographies that - but let us dissemble. [ . . . ] After all we are all human - Count Tolstoy, R. Fitzsimmons, Peter Pan, and the rest of us. Don't lose heart because the story seems to be degenerating into a sort of moral essay for intellectual readers. (O.Henry, cited in Ejxenbaum, 1978: 255) Recalling the three determining criteria of this model, then, the metafictional case requires the following: the presence of a marked vocative function; the absence of characterisation of the "you" referent; and self-conscious acts of reception and interpretation, the actual reader having decided that the "you" does denote some approximate rendering of him or herself. 56 In the beginning of the last chapter, I inform'd you exactly when I was born;but I did not inform you, how. No . . .besides, Sir, as you and I are in a manner perfect strangers to each other, it would not have been proper to have let you into too many circumstances relating to myself all at once. You must have a little patience. (Sterne, 1940: 10-11) The term mesofictional is coined as an adjunct to the term metafictional: and, like metafictional-"you," it is quintessentially a category of address. Where the metafictional case implicates a self-aware reader who stands on a higher ontological level than the story and the world of the text, the mesofictional case constitutes an addressee figure - whether "reader" or implicit interlocutor - that is credulous, that accepts the world of the text (or more precisely, accepts at least one of the textual worlds available in the reading of the text). This is the reason I identify the "you" of Sterne's Tristram Shandy as belonging to the mesofictional class rather than to the metafictional. Certainly, Tristram Shandy draws attention to itself as metafictional in Waugh's broader sense; but, in the narrower sense, as a special case of the "second person," it does not exhibit the self-consciousness at the level of the utterance required of metafictional-"you." The reader-narratee of the mesofictional text is required to be credulous and respectful of the narrative utterances. It is the case in which the "you" signifies a peripheral but still uncharacterised narratee, a "you" referent who may be installed within the story, if not yet as a participant in narrative action--Virginia Woolf's The Waves (1963); Atwood's "Rape Fantasies" (1978)--or who may be situated outside the story, but nonetheless held embedded in (and believing of) the story's (fictional) world--Charlotte Brontè's Jane Eyre (1966) and Villette (1990). Indeed, The Waves (1963) and Jane Eyre (1966) can stand as the two paradigm cases of the mesofictional class. The "you" of the final section of The Waves represents the interlocutory narratee present upon the telling of the story; and the "you" of Jane Eyre's journal represents the literary narratee, the reader. You see me, sitting at a table opposite you, a rather heavy, elderly man, gray at the temples. You see me take my napkin and unfold it. You see me pour myself out a glass of wine. And you see behind me the door opening, the people passing. But in order to make you understand, to give you my life, I must tell you a story. . . . (Woolf, 1963: 169) You have forgotten little Adèle, have you, reader? I had not. . . . (Brontè, 1966: 475) In neither case does the "you" participate in the tale or intrude markedly on its unfolding. Rather, these two act principally (but not solely) as recipients of acts of "first-person" narration and thereby provide a realist alibi to the story-telling situation. In each case, the mesofictional becomes the "person" employed by a narrative to generate, contingent upon further contextual material, a sense of the situation of narrational enunciation, calling forth or animating the situation in which the narrating takes place. Salinger's opening gambit in Catcher in the Rye (1958), for example, uses a mesofictional-"you" to establish the situational relationship as that of patient/psychiatrist, in which the "you" is constituted as a silent but not disinterested, critical listener. "If you really want to hear about it, the first thing you'll probably want to know is where I was born. . . . I'll just tell you about this madman stuff that happened to me around Christmas . . . " (Salinger, 1958: 5). Atwood's strategy in "Rape Fantasies" of presenting a monologue spoken by a woman to a (never characterised) male stranger across a bar table constitutes a situational relationship similar to Woolf's, but one that is finally much more charged and implicitly menacing. The principal function of the passive and receptive mesofictional-"you" of The Waves is, firstly, to provide a space into which the character Bernard can narrate his tale in a bid to delimit the bounds of his own identity; and, secondly, into which, at another level, the engaged reader might step, in that move constituting him-/herself as a particular type of reader. In "Rape Fantasies," the "you" is present in the narrative not merely (or not even) as an enabling, receptive addressee: "I don't know why I'm telling you all of this, except I think it helps you get to know a person, especially at first, hearing some things they think about" (Atwood, 1978: 103). The "you" also plays a significant role in the implications of the narrative - in the events that follow its last sentences, which by no means constitute an ending. The narrative's true closure is elided, yet to come, the stranger (by implication) following the naive narrator from the bar, perhaps himself a rapist of a quite different disposition to those of the narrator's confessed fantasies. For instance, I'm walking along this dark street at night and this short, ugly fellow comes up and grabs my arm, and not only is he ugly, you know, with a sort of puffy looking face, like those fellows you have to talk to in the bank when your account's overdrawn . . . but he's absolutely covered by pimples. So he gets me pinned against the wall . . . and he starts to undo himself and the zipper gets stuck. I mean, one of the most significant moments in a girl's life, it's almost like getting married. . . , and he sticks the zipper. So I say, kind of disgusted, "Oh for Chrissake," and he starts to cry. (Atwood, 1978: 99) If the male reader has indeed settled into the position of addressee over the course of the story, perhaps adopting an attitude of vague, even wry condescension towards the naive narrator, then this is a confronting position to find oneself in at the close of the story. Given the explicit gendering of the "you" as male, the situation for the female reader may be quite different. She may not have tied herself so closely to the narratee position in the first place, but might instead have come more readily to identify with the authorial element, that is, the ironic voice above the narrator inviting judgement on her expurgated notions of sexual assault. Or, given the nature of the psychoanalytic concept of fantasy, the female reader might identify more readily with the "first-person" narrator herself. Of course, the female and male reader might both finally come to see the narrator's fantasies as bids to maintain some degree of autonomy and control - even if only in the Real of fantasy - within a social environment in which she feels vulnerable and impotent. The degree to which the mesofictional is less generalised and more singular - the degree to which it is subject to more detailed description of specific attributes, spatial and temporal relations, or characterisation - marks the movement towards and across the soft boundary between the mesofictional and narrative classes of "you." Those barely-drawn diegetic figures of The Catcher in the Rye, "Rape Fantasies" and The Waves rest squarely on that cusp between the mesofictional and narrative-"you." When the "you" does become a fully-fledged protagonist, when in the story it stands up from the table and comes to stand behind the "first-person" narrator to rest its hand on the narrator's arm and amiably begins to speak in its turn, then it enters the narrative class. In this class, the referent of the "you" is involved as a protagonist in unique narrative action. Before moving on to discuss the narrative case, then, it will be useful to return to the questions raised by the mesofictional case's three constitutive criteria. The referent of the "second-person" pronoun is not a participant in the unfolding action - not, certainly, at the moments of the utterances in question. 57 The "second-person" pronoun does carry the grammatical function of the vocative. And lastly, the "you" signified by the vocative is a figure embedded in the world/s of the text itself and is not an appropriate cipher for the implied (or actual) reader. You, yesterday, did the usual things, just as any day. You don't know if it's worth remembering. . . . Yes: yesterday you will fly home from Hermonsillo. . . . [E]ntering the Valley of Mexico . . . flames will begin to hiss from the outside right motor; and everyone will shout and only you will remain calm and motionless. . . . The fire suppressors inside the motor will operate, and the plane will land uneventfully, and no one will have noticed that only you, an old man of seventy-one years, that only you maintained composure. You will feel proud of yourself without showing it. You will reflect that you have done so many cowardly things in your life that now courage has become easy. . . . (Fuentes, 1966: 7-8) Narrative-"you" embraces all forms of self-addressed "you" and character-addressed "you" narratives as represented by the first two figures of my addressee model in Chapter 1. Only those involve the "second person" as a subject of narrative action - as an existent involved in particularised and unique narrative action. In the above passage, for instance, the "you" clearly refers to a narrative protagonist who exists, so to speak, in the particular time and space of the text-world. This is made explicit in the text itself in that the passage follows one in which a "first-person" narrator urges himself to reflect back on the day before. The "you" immediately refers itself to the "I" of the previous pages, a patriarch on the verge of death. The metafictional, mesofictional and figural, on the other hand, as modes that appeal to the (implied) reader or narratee or that deploy a host of rhetorical functions and tropes, point above or beside the point of the action of the story and the story's ontological planes. The narrative case requires that the antecedent of the "second-person" pronoun, the deuteragonist, to borrow John Barth's expression, possesses an actantial function. 58 For Monika Fludernik, actants (or existents as she prefers to call them) are prototypically human, by which I take her to mean that the nature we attribute to them is modelled on the nature we auto-affectively attribute to our own experientiality/existentiality. They are textual entities who "can perform acts of physical movement, speech acts, and thought acts, and their acting necessarily revolves around their consciousness, their mental centre of self-awareness, intellection, perception and emotionality . . . " (Fludernik, 1996: 26). Implicit in Fludernik's notion, moreover, is the assumption that the "you"-referent's acts of speech, thought, movement and so on will possess the singularity and uniqueness attributed to lived experience. By way of illustration, Fludernik's description of the actant provides a useful gloss on Morrissette's reason for arguing that Maugham's "The Beast of Burden" (1925), a passage of which follows, is not an example of narrative modality. You see a string of coolies come along, one after the other, each with a pole on his shoulders from the ends of which hang two great bales, and they make an agreeable pattern. . . . You watch their faces as they pass you. They are good-natured faces and frank, you would have said, if it had not been drilled into you that the oriental is inscrutable. . . . You will be thought somewhat absurd if you mention your admiration to the old residents of China. You will be told with a tolerant shrug of the shoulders that the coolies are animals. . . . You can no longer make a pattern of them as they wend their way. (Maugham, 1925: 209-301) According to Morrissette, "The Beast of Burden" does approach narrative modality at some points through the description of specific past action - the speaker's implicit admission of his own admiration for these workers, for instance. For Morrissette, this narrativisation is obstructed because the text "is immediately generalised by the combination of 'you' with the future tense ('you will be . . . ') into something standard or typical and thus rendered nonnarrative" (Morrissette, 1965: 6). Although Fludernik would agree, what the generalisation of the "you" referent achieves for her is the stalling of any movement towards that figure's particularisation in spite of its clearly anthropomorphic nature. And indeed, typical of such pieces, the "you" remains throughout "The Beast of Burden" a detached, if affected, observer. The difference between the "you" of narrative and the "you" of the "second person's" guide book mode is that, in narrative, the "you" is necessarily unique. On the other hand, the "you"indeed the action itself, so far as there is any - of the guide book mode is infinitely repeatable and is always typical rather than unique. As assuredly as the "s/he" of "The Beast of Burden" exists in the world of the text, and as "human" and "moral" as "s/he" is represented to be by the passage's end, the "you" performs no acts, exhibits none of the singularity and uniqueness required of a narrative character, and its moral dimension becomes subsumed under the amorphous, humanist categories of "the universal." About some towns there is a quality that impresses straight away, an intangible quality hard to define, just as you may be lost for an explanation as to why some individual appeals to you. . . . Of course, your most secret thought will be public property before you utter it, and your reputation shredded up and down the main street. (Farwell, 1949: 22-23) One of Morrissette's most significant observations in "Narrative 'You'" is that the "second person's" rhetorical cast "never, in fact, disappears entirely from the mode, even when it becomes unmistakably narrative" (Morrissette, 1965: 10). Rhetorical functions - what I refer to as the figural cases of the "second person"will deeply inflect instances of narrative-"you," mesofictional-"you," and metafictional-"you." Nor are the effects of this affective core restricted to texts in which there is a sustained, systematic use of the "second-person" pronoun. The "second person" may appear in any passage of narrative (or quasi-narrative) text, and indeed, as examples below show, the rhetorical effects themselves rely not only on orthodoxies of literary practice and infractions against orthodoxy, but very much also on everyday usage. They work variously to generalise, moralise, axiomise, coax, goad, question, interrogate, evaluate, judge, reproach, challenge, exhort, mock, accuse, admonish, advise, instruct, lecture, and so on. Two of its most frequently noted rhetorical functions (as discussed in Chapter 1), are its tendencies to universalise/generalise - which in turn can facilitate the reader's identification with the "you" and its didacticism. Indeed, these two rhetorical elements are often drawn into a particular relation that can function to constitute the text or passage itself as a particular form of "knowledge," as something "finally known." It can be said about this knowledge, "of course you know this, remember? this is what you know." And in "remembering," the reader is interpellated or constituted as a particular type of reader. This knowledge, this memory provided from outside, is what Costello refers to when she describes the "second-person" pronoun as reaching out uncannily as a "memory" that needs to be returned or "narrated over," its utterances constituting for the reader a "re-imagined self" (493) a re-imagining that often carries an explicitly moral dimension. This returned knowledge or memory also appears to have the character of what Paul Virilio, in War & Cinema: The Logistics of Perception (1989), calls "paramnesia." Reflecting on popular/mass audience genres such as the Second World War movie which narrate stories that we seem, again uncannily, already to know (for instance: The Holocaust; The Besieged City), Virilio describes "paramnesia" as a type of memory produced (intertextually) in our experience of texts rather than within actual experience, but a memory to which we respond, nonetheless, in the manner of: "Yes, I know this to be true" (Virilio, 1989: 33). Thus, in the film Zentropa, the disembodied narratorial address made to "you" - not necessarily the protagonist, but conceivably to the cinema audience - provides an exemplary "second-person" narration of just such a "paramnesiac" memory. Zentropa's narratorial address, moreover, employs the artifice of a psychoanalytic/hypnotic exhumation of repressed memory (albeit in a self-consciously allegorical context), taking/talking the "you" back to a dream-like, "lost" recollection that is in need of being "narrated over." To return one last time to the determining criteria of the reference model's cases, if the first criterion concerns the degree to which the "you" pronoun carries the function of address, then the response for the figural case has to be heavily qualified. As in the narrative case, the figural can no longer be said to carry the full vocative force expected of the "second-person" pronoun, because the "second-person" pronoun's deictic function has become substantially and even fully disengaged from its grammatical, morphosyntactic form (Herman, 1994: 392). Where the function of address remains implicit in the grammatical form of the figural "second-person" utterance and the pronoun's related verbs, only the most general of addressees is designated. This general addressee, moreover, remains non-particularised, so to answer both the second and third criteria, in its "degree-zero" form (in which "you" = "one"), it can systematically designate neither a character/protagonist in a story, nor a reader or readership. On the one hand, as soon as the text endows the referent of the "you" with the marks of particularity and with "the parameters of a real-life schema of existence . . . situated in a specific time and space frame" (Fludernik, 1996: 30), at that point the utterance begins to be constituted as an instance of narrative-"you." On the other hand, the generality of the figural also precludes its identification as a reader, because, as an element within figures of speech, the figural points to no particular "person" at all - points to "no-body." At the point at which the "you" becomes a less generalised addressee, here it begins to be constituted as an instance of meso- or metafictional-"you." Part 2. The Generalising Function. In parts two and three of this Chapter I want to focus on the two key functions of generalisation and address, both of which are not only pervasive within "second-person" textuality, but, as I will argue in Chapter 7, contribute to the deepest effects of Protean-"you." A powerful function of generalisation is the defining feature of the case of the "second person" I call generic-"you." It tends to make incidents in a passage of text appear typical and probable, presenting an incident as that which is likely to occur to anybody, or a scene as the one likely to be viewed by anybody, in the circumstances given. Uri Margolin describes the generic or impersonal "you" through the aphorism: "You can't take it with you." This "you" is non-specific, "involving everyone or anyone, singular or plural: any reader or hearer of the message, plus others" (Margolin, 1997/91: 3-4). The generic utterance, then, refers to nobody (in particular). This refusal of the utterance to designate a particular addressee is the feature Herman focuses on in his description of what he calls the "pseudo-deictic you," named after Melissa Furrow's example (Herman, 1994: 396; Furrow, 1988: 372). In conventional circumstances, deixis and grammatical form will agree, such that, operating deictically, the pronoun will designate the addressee of any utterance. In two broad classes of "second-person" utterance, however, deixis and address cease to agree. These two classes are those that I have designated narrative-"you" and figural-"you." Herman's analysis of pronoun deixis and grammatical address (discussed further in Part 3 of this chapter), helps further explain the impression of similarity between some figural and narrative utterances. Herman shows that both share the feature of the disengagement of the deictic function from the grammatical form. In narrative-"you," the work of deixis is frozen so that the "you" no longer works as a shifter, but refers dominantly, instead, to a particular actant. In the case of the figural, deixis is not so much frozen as dissipated: "you and your virtually lose their deictic force" (Herman, 1994: 392). Allowing Herman to cite his own points of reference, the uncoupling of the pronoun's deictic function from its grammatical form produces an "impersonal or generalised you" that: often plays a prominent role not only in second-person literary narratives but also in (the language of) proverbs, maxims, recipes, VCR instructions, song lyrics, and, though they might tell you otherwise, astrologers' prognostications. Thus, in a string like When you're hot, you're hot, "the second-person pronouns are impersonal: non-deictic in that their interpretation does not depend directly on any feature of the non-linguistic context of the utterance" [Anderson and Keenan, 1985: 260].59 (Herman, 1994: 397) To that extent, generic-"you" is to be defined inversely against the degree of specification, singularity or uniqueness of place, time, character/personality, activity, speech/thought, and so on attributed to the "you" referent. At some point, coming with greater specificity, it is no longer feasible to regard the "you" referent as just anyone.
Fig. 12. The Generic-Particular curve. The "you" can be made particular in one of two directions: it can be constituted as a narrator's addressee; or it can emerge as an individuated actant (see figure 12). As an addressee it falls into the vocative cases, becoming a meso- or metafictional-"you," and finally a specified, actual reader. Characterised, it crosses the soft boundary from the figural's generic modalities into narrative cases of "you," becoming recognised as a particular narrative protagonist performing particular acts and expressing particular attitudes. It is to be remembered, of course, that these terms are still defined for their referential function, not for the allocutionary function which will, by and large, remain present. Being equally available to narrative cases as to the vocative cases, the allocutionary function cannot be taken as a definitive term of this system, forming a second keystone of the pervasive figural class, standing adjacent to the generalising function. Morrissette himself describes a "family of cases" that make explicit the generalising function, a family of generic text-forms that includes cookbooks, guide books/travelogues, craft and technical manuals, and "a whole group of 'you' modes related to publicity, advertising, and journalese," in which "'you' invites the reader to place [her or] himself in the position of the writer, with the clear implication that anyone who so places [her or] himself will witness the identical scene or perform the same action" (Morrissette, 1965: 2). This generalising force is clearly diffused by any increase of detail attached to the "you." Nonetheless, such is the generalising function's tenacity that some trace will often remain. Morrissette's generic "family of cases," however, clearly involves a complex range of over-determinants, tonalities, and variants, so that even brief instances may exert a medley of rhetorical force and tonality: generic "second person" utterances rarely just generalise. A generic address, moreover, might arise a small number of times and with no regularity in any text, as is the case in each of these three short passages: Overhead, clouds were shredding themselves into rags. You could see the firmament better, the air was so much warmer up there in the mountains than it was down by the coast. . . . (Olshan, 1994: 60) We advised each other on which remote cities were well maintained, which were notable for wild dogs running in packs at night, snipers in the business district at high noon. We told each other where you had to sign a legal document to get a drink, where you couldn't eat meat on Wednesdays and Thursdays, where you had to sidestep a man with a cobra when you left your hotel. (DeLillo, 1982: 7) I got to know the trees pretty well over the years and I never had trouble finding the ripest fruit. You only had to stand there and cup the cumquats gently in your palms and they would fall off and come to you. (McGregor, 1985: 5) No supplementary function is invoked in these mesofictional utterances, certainly not in the way that the generic's generalising function will frequently be combined with others. The guide book-"you" mode, for instance, will typically combine the generalising function with the procedural and propositional functions, as in this passage from Book II of The History of Herodotus, written circa 440 BC. The following is the general character of the region. In the first place, on approaching it by sea, when you are still a day's sail from the land, if you let down a sounding-line you will bring up mud, and find yourself in eleven fathoms' water, which shows that the soil washed down by the stream extends to that distance. (Herodotus, n. p.) The excerpts from Olshan, DeLillo and McGregor quoted above, however, are representative of the more exclusive instance of the generic-"you." But they are not all of a piece. Only the first passage might be considered neutral: each of the others has acquired a tonality, a colour. The nuances of tone are not drawn solely or even primarily from the "you" utterances themselves, but, rather, arise from the passage in which they are embedded. Significantly enough, however, the tonality of each appears to sharpen at the moment of the "you" utterance, so that tonalities or effects do assemble around instances of the generic "you" address. The ethical/evaluative function that generates overtones of censure in some "you" texts, for instance, cannot be said to arise because of the "you" utterances alone, but certainly finds its focus and forcefulness in those utterances. Where the "you" figure of the first passage is quite neutral of tone, the second passage, while certainly generic in character, draws to itself something of the extraordinary: the "you" is particularised to the extent that it signifies an entity familiar with the exciting dangers of travel in exotic places. Similarly, the third passage is inflected both by the vernacular informality of its tone and by a certain emotional/psychological attitude. The two generalised, referentially unspecific "you" figures of the DeLillo and McGregor passages are much more narrowly cast than the every-person "you" of the Olshan. The "you" of the third passage even acquires for itself a shadow of narrative-"you," an effect arising in part from the explicit relation between the narrative-"I" of the first sentence and the rhetorical "you" of the second, 60 and in part from the way in which the image of the palms cupped under the cumquats begins to provide the "you" a sense of embodiment. What these examples show is both the rarity of the neutral case and how readily generic-"you" takes up other tonalities. At this point, I would like to grant three premises with regard to the generic-"you:" first, that the generic-"you" mode is defined by the generalising function (so that it is proper to say that, wherever the force of generalisation asserts influence on any "you" referent, then there is the presence, at least as a trace, of the modality of the generic-"you"); second, that this generalising function will, in most instances, be accompanied by other functions, so that any "you" utterance will rarely serve only to generalise/universalise (though this may certainly be the dominant function); and third, that this generalising function is dissipated by degrees (if rarely fully) through the text's act of making particular or unique the "you" referent's identity and activity, whether this is by (progressively) particularising the "you," or by specifying or naming the addressee/reader. The "second person's" glide away from the generic, for convenience of description, might be broken into five standard or typical cases. These are: a. generic-"you," whose referent is "no-body in particular" b. particular-"you," whose referent is a specific, textually embodied protagonist c. a significantly common variation of particular-"you" in which the referent is transposed from generic-"you" status to particular-"you" status d. inclusive-"you," whose referent is "anybody, in general" e. exclusive-"you," whose referent is "this specific person" Refining the earlier bell diagram, the paradigm cases can be represented thus:
Fig. 13. The Generic-Particular model revised as a field. The five nodes are to be understood solely as indications of the status of the "second person's" generalising function. They do not name categories or types of utterance. Instead, they name loose, if nonetheless definable, clusters of paradigmatic cases of "you" utterance, each of which is associated with a specific function. There are three such functions in this model: the generalising function (for the generic case); the allocutionary function (for the two vocative cases); and the narrative function (for the narrative cases). The increasing particularisation of the addressee or narratee and concomitant waning of the generalising function does not, strictly speaking, trace a linear continuum. The myriad cases of the "second person" are complex and highly fluid, no case, generally speaking, necessarily superseding or excluding another. For instance, the possibility that a "second-person" utterance might project a generalised form and a particularised form simultaneously is already implicit in my arguments that the generalising function is pervasive. The diagram above (fig. 13) is not to be thought of as depicting a spiraling, linear progression from node to node. The model is more appropriately understood as a surface or field across which can be mapped any number of cases of the "second-person's" relationship with the generalising function. The nodes I will discuss stand around the arched periphery of this field. Ranged across its interior are the more ambiguous and hybrid cases. That the periphery of the model is not closed might itself be taken as an iconic metaphor. "Second-person" discourse cannot be so easily contained even by so broad a descriptive model. The first case is the more "pure" generic as it is illustrated in the Olshan, DeLillo and McGregor passages above. From there, as the "second person's" modality moves away down the left wing of the arc into the meso- and metafictional cases, or down the right wing into the narrative cases, the force of the generalising function carried potentially by any "second-person" utterance will typically diminish. On the vocative (or address) wing, the least generalised "you" referent will be a specified narratee or figure of address not involved in narrative action--as Bernard's interlocutor in the last third of Woolf's The Waves (1963), and Phill Donahue in Lee Smith's "Dear Phill Donahue" (1981). On the protagonist (or narrative) wing, the least generalised "you" referent, of course, is the (immediately) particularised actant of narrative-"you"--as Delmont in Butor's La Modification and the unnamed "you" of McInerney's Bright Lights, Big City (1986). If the utterance in question is mesofictional-"you" or metafictional-"you," then the strength/weakness of the generalising function will reside, in the first instance, in the degree of specification of a referent. The status of the referent as mesofictional or metafictional, though, has little immediate bearing on the generalising function itself. What does count is the inclusivity or exclusivity of reference. That is, what counts in this respect is the degree to which the pronoun refers either to anybody or somebody in particular. That the "second person's" sense of address can be divided into these two broad categories has also been suggested by Katherine Passias, who has argued that the "second person" pronoun carries a pair of grammatical functions that, for her, are mutually exclusive of one another. "The two basic linguistic functions of the second person pronoun are the illocutionary function - the act of addressing a receiver of the message directly - and the collective function" (Passias, 1976: 198). The grammatical function that dominates, she writes, depends upon the reader's experience, degree of identification with the "you" being addressed, and whether the reader feels directly (and personally) or generally addressed. Unlike my own position, however, she contends that both of these functions vanish altogether when "a very personal experience" is described, that is, in narrative-"you" (Passias, 1976: 199). In such instances, she states, the "surface" pronoun of "you" reveals itself to have the "deep" structure of an "I," an "I" speaking to him or herself and not addressing a reader. Passias makes one concession: both address functions will be present as a "deep" residue. This concession, however, is made on the basis that narration is always an address to a listener/reader, and that "surface" pronouns don't alter the implied, underlying relationship between author, reader and text - and so it is a concession that applies to all literary narrative. In other words, "second-person" modalities can make no special claim to them. In the inclusive case, the referentiality of the "you" remains at its broadest, and emerges through an addition of a marked allocutionary function to an otherwise generic utterance, the pronoun now necessarily referring to a person (or at the very least a thing, a body), rather than to no-body. The most traditional and conventional form of the inclusive case is the parenthetical "dear reader" variety of general reader address used to such advantage by Charlotte Brontè. And reader, do you think I feared him in his blind ferocity?if you do, you little know me. (Brontè, 1966: 456) Religious reader, you will preach to me a long sermon about what I have just written, and so will you, moralist; and you stern sage; you stoic, will frown; you cynic, sneer; you epicure, laugh. Well, each and all, take it your own way. (Brontè, 1990: 194) Here, the "you" addressed might be anyone at all. Certainly, of course, the passage from Villette appears more specific in its designation of particular addressees, but, although it does provide a list of readers, the passage does not address each one individually so much as hypothesise on the broad composition of the tale's readership. All of these readers - the pietist, the moralist, the sage, the cynic, the epicureare included within Villette's generalised, confessional address, and it is clear, moreover, that they are by no means the sum of the tale's readership. The two readers most conspicuously omitted, of course, are the young woman for whom this didactic moral tale will have some benefit, and the like-minded, sympathetic confessor who will take the story Villette's way. The reader inscribed into the text by Jane Eyre, too, is subject to moments of specification which nonetheless do not significantly affect the generalising function. In the exclusive case, the identity of the referent is more narrowly cast. The "you" no longer refers to just any addressee, but to somebody in particular. Like the movement from the generic to the inclusive, the movement from the inclusive to the exclusive is scalar. This "you" referent can be made particular through the narrative's attribution to the narratee of specific characteristics, experiences, implied specialist or private knowledges, a name, and so on: the greater the density of such specifications, the more exclusive is the case. As this case involves acts of direct communicative address between identifiable parties, another of its indicators is the detailed articulation of the situational context, such as is a feature of most epistolary fiction. The "you" referents constituted through the exchange of letters in Helene Hanff's novel 84 Charing Cross Road (1981), for instance, leaves no ambiguity about who is saying what to whom on what date and in what circumstances. Indeed, the epistolary form, for its typical lack of ambiguity (and the accompanying weakness of any sense of generality), can be given as the paradigmatic case of the exclusive recipient. Take, for instance, the following passage from Margaret Coombs's The Best Man for This Sort of Thing (1990). To Jemima Ayling Chelsea, 25 October 1971 Dear Jemima, Here I go again - but it has to be you because I've lost faith in "Dear — ". At least you're real in a way that "Dr Argyle" and "Dear — " are not - and you won't always be only six months old! I'm writing to tell you what it's like, how lonely I am. . . . Kate has a middle ear infection and you, poor baby, have a cold, and this morning I was close to paralysed with depression. . . . (Coombs, 1990: 313) Conversely, Coombs's The Best Man for This Sort of Thing also provides a number of instances of a rarer, non-specific mode of epistolary address, hinted at in the above passage: "Dear — ." The addressee of these letters is unidentified, but is construed by the letter-writer to be a sympathetic, attentive correspondent. For a period, the atypical openness and non-specificity of the address facilitates the generic mode's function of drawing the epistle's reader into the story, in some measure inviting the reader to take the address as "Dear reader." But this effect is neither strong nor sustained. Helen Ayling's letters are written to no person at all, not even, we realise soon enough, to a Dear Reader. Unlike Celie's letters to God in Alice Walker's A Colour Purple (1983), Coombs's "Dear — " is entirely a fabrication, a nobody, so to speak: "Dear —, God I wish there was somebody I could talk to about this, someone who'd understand how angry I felt. (Not like fucking Don!)" (Coombs, 1990: 249). For a period, then, Ayling imagines her own correspondent, only to reveal it finally as always having been inadequate: "Dear —, I'm reduced to writing to you again. I have to tell someone" (Coombs, 1990: 280). Finally, in a stroke of self-irony, despairing of the value of even this imaginary addressee, she addresses herself to the inanimate medium itself. "Dear Piece of Paper, I am grief-stricken . . . " (Coombs, 1990: 250). The immediately particular case of the narrative pronoun is that in which the protagonist is fully individuated from its first appearance in the text. The protagonist might be introduced within the first lines, as in Michel Butor's La Modification), or far deeper into the work, as in Brian Aldiss's Somewhere East of Life (1994). As Darlene Hantzis puts it, individuation occurs in the specificity of the character (Hantzis, 1992: 94). Particular, unique behaviour and individual characteristics "both affirm and participate in the perception of the presence of a subject" (94). Standing with your left foot on the grooved brass sill, you try in vain with your right shoulder to push the sliding door a little wider open. You edge your way in through the narrow opening, then you lift up your suitcase of bottle-green grained leather, the smallish suitcase of a man used to making long journeys, grasping the sticky handle with fingers that are hot from having carried even so light a weight so far, and you feel the muscles and tendons tense not only in your finger-joints, the palms of your hand, your wrist and your arm, but in your shoulder too, all down one side of your back along your vertebrae from neck to loins. (Butor, 1958: 9) Not all cases of narrative-"you," however, so hastily declare their referents as narrative protagonists. The becoming particular case, as clumsily named as it is, is a variation on the previous case in which a generalised "you" transforms into a more fully narrativised and less generalised "you." The generic offers the reader a more comfortable and more ordinarily stable relationship with the "you" referent because of the generic's established place in conventional rhetoric, and consequently draws the reader into such a relationship with greater ease than would a potentially more confronting and alienating narrative-"you" referent. But, having engaged the reader, having successfully invited him or her to identify with an every-person, the narrative then transforms that "you" into a unique protagonist. The first lines of Don DeLillo's Running Dog (1978) exemplifies this movement. "You won't find ordinary people here. Not after dark, on these streets, under the ancient warehouse canopies. Of course you know this. This is the point. It's why you're here, obviously" (DeLillo, 1978: 3). The opening sentence offers an immediately recognisable generic-"you," one strongly inflected with the vernacular and a tone of irony: the statement is flat and imperative and the "you" might be anyone at all. The next appearance of the "you" very few words later, however, is immediately and unmistakably narrative. The movement from the generic to the particular in Ruth Morse's "A Journey" (1988) is even more gradual, but also much more conflicted. The work seemingly begins with a generic-"you" address to a reader who knows what it is to travel beside a nuisance co-traveller, but then begins to mark a shift from the figural's realm of the general into the narrative's realm of the unique and particular. It is a movement, however, that is not decidable until far into the piece, if not until the very end. You recognise her. You've been her victim, too. She climbs watchfully . . . up the Greyhound steps. . . . Is it peripheral vision that sets the alarms ringing? Indeed, you, too, have been nervously watchful, hoping that the bus will be empty or at least not crowded, that the plagues of nuisances who threaten the fragile and certainly precious privacy of public places will pass you over. (Morse, 1988: 75) For much of the piece, robbed of certainty by the ambiguous character of the two opening statements, only partly persuaded by the figural character of the guide book mode invoked as a genre model by the opening sentences, specificity or typicality is undecidable. The "you" may belong to either the narrative or the vocative arm. It clearly refers to an entity involved in the performance of narrative-like action, but this action appears to resolve into action that is typical and generic, rather than unique. That is, it becomes action more akin to the iterative action that is the staple of travel writing than to the unique action of narrative proper. As the narrator herself reminds us, "There are jokes about this encounter" (Morse, 1988: 77). We come to recognise the tale as one of those pieces in which the narrator typically sets out to say: "I/you/we have all had this experience, haven't we?" This sense of typicality soon appears confirmed, moreover, when the details of the bus trip itself become more fluid and speculative. For instance, the "you" may be seated first, or she may board late; and the family member meeting the unwelcome traveller at the conclusion of the journey is identified alternately as a married daughter and as an unmarried son. The sense of typicality/generality is also maintained through the presence of lists. Rather than stipulating singular and unique acts and elements, the narrative is punctuated by catalogues of options that are merely typical of a circumstance: "Now, even now, though you bury yourself in your newspaper book magazine letter term paper . . . " (Morse, 1988: 75). But simultaneous with this are tokens of specificity: No, no, if she had to make a long haul flight you know where she would be sitting. In fact, coming back from your last conference flight trip journey you thought she was right there. And perhaps she was, but in another language and didn't have a word of English, and your glee was quite unseemly. (Morse, 1988: 78) Catalogues of options like "conference flight trip journey" are accompanied by references to specific incidents or facts. "A Journey," then, begins to rewrite itself as a particular woman's travel story on a particular occasion. Albeit gradually, it begins to become a story, or a vignette, about a Greyhound coach ride that continues a journey begun with an Atlantic crossing (implying a journey of some importance), travelling past trees that are "scarlet memories of the fall you have not seen in twenty years" (Morse, 1988: 77). It is only as the piece progresses that the "you" referent begins to acquire the specifications of a life history, and that the experience of the long-haul bus trip becomes more than merely typical. Indeed, the structural gradualness of the move toward specification itself becomes a thematic element, coming to figure or echo a dramatic/psychological deferment of "coming to the point" in the story - a point about aging and a child's responsibilities toward a parent that also glances at the situation of remaining a single, childless woman. In a story about sitting next to a woman who talks incessantly about aging, family, and grandchildren, not only is the referential specificity of the "you" suppressed for much of the piece, but so are clear enunciations of the particular issues facing the younger travelling woman. It emerges only very slowly and obliquely that the "you"-protagonist has taken a long journey for family reasons, to see a parent, probably her mother, responding to a late-night, cross-Atlantic phone call. More dangerous still, the reason for this journey which you have made suddenly and without warning beyond the often contemplated knowledge that it would come with just this suddenness. . . and when the telephone rang in the middle of the night because this is how it always happens you had to come via New York. . . and then the shuttle is solid and there is a waiting list you decide it will be almost as quick to take the bus anyway you will have the solace of keeping moving and awake the comfort of doing something when there is nothing to be done. For there is old age waiting for you and this above all you do not wish to share with this old woman who must think in the night what you think but more sharply, more presently will it be today for you think you have time but she only hopes that she has another little while as New England fades outside will it be all tubes and flashlights or a human being who resembles the one I remember? Is there time? Lord, if not now, when? (Morse, 1988: 79-80) In spite of the work undertaken by the text to maintain the impression of generality, of typicality, of non-identity, there is throughout the story a specific "you" moving underneath its generalising utterances, an unclear figure that does finally emerge as a unique entity in the last paragraphs, and which, in the concluding lines, does face that which she has been avoiding throughout the piece. What she has been avoiding, in fact, is not only confessing (to the old woman? to herself?) the "dangerous reason for this journey," the matter of her aged, ill mother - a confession which can only work now as a reminder to herself of her own status as aging, single and childless, themes already established in the story. What the closing lines also mark is that the younger woman, a dissembling "I"/"you" narrator, has been avoiding an identification with the older woman, the nuisance traveller, a woman "who must think in the night what you think but more sharply. . . for you think you have time" (Morse, 1988: 80). The presentation of a generalised "you" with which (any)one can identify followed by a shift to a more fully narrativised "you" is a characteristic of "second-person" textuality observed by many critics. Its use in this way is described as a strategy for drawing the reader into the tale before the pronoun is particularised in narrative. Readers coming upon this generalising modality certainly may feel sorely tempted to identify at first, but only so long as their situation overlaps with the protagonist's. . . ; as soon as the protagonist becomes too specific a personality, becomes, that is, a fictional character, the quality of the presumed address to an extradiegetic reader in such texts evaporates. (Fludernik, 1994b: 287) Faced with specificities of "sex, job, husband or wife, address, interests, and so on, . . . the reader has to realise that the 'you' must be an other, a or the protagonist" (Fludernik, 1994c: 452). In many instances, however, the effect of direct reader address may not entirely evaporate after all. As Fludernik notes, there are indeed some texts in which the generalised reading ("you" equals "one"), in the form of a very specific reader role, persists despite the narrowing of reference, and it does so because in these texts the desired effect is precisely to make the reader feel personally responsible, personally caught in the discourse. (Fludernik, 1994c: 452) In Jamaica Kincaid's A Small Place (1988), for instance, writes Fludernik, the "you" initially seems to be the generic "you" of the guide book mode (Fludernik, 1994c: 453). "If you go to Antigua as a tourist, this is what you will see. If you come by aeroplane, you will land at the V.C. Bird International Airport" (Kincaid, 1988: 3). It slowly emerges, however, that the "you" within the opening section of the text refers to a particular tourist whose "background and current experience are sketched in ever more specific terms, thereby signalling that she is no longer just 'anyone' (and therefore no longer the virtual reader in his or her real-world identity) but has turned into a fictional character" (Fludernik, 1994c: 453). This woman is made to shoulder the guilt of Western society towards the colonial subject and implicitly becomes an object of identification with the real reader. . . . Even better than third-person (or first-person) reflector-mode narrative, the "you" [as] a reflector character can induce the hypnotic quality of complete identification by a maximal bid for readerly empathy on the discourse level in terms of the generalised "you"a "you" that initially keeps the communicational level well in view - and it can even make this generic meaning reemerge, turning fiction again into virtual facticity or "applicability." (Fludernik, 1994c: 453) Thus, where the simultaneity (and undecidability) of case in "A Journey" resolves finally into a particularised narrative-"you," in A Small Place the case resolves back into the generic modality that was invoked in its opening lines, but does so with a re-doubling of effect, the narrative discourse achieving the striking duplicity of being experienced by the reader as an inclusive-"you" address and as direct, aggressive address, the reader "asked to feel guilty, to recognise him- or herself in the negative image" presented by Kincaid of the Western tourist and of that tourist's complicity with the economic and cultural degradation of Antigua ("Test Case" 453). Part 3. The Allocutionary Function 10. The Allocution of "You" The allocutionary function concerns the "second person" pronoun's effect of address, which it only rarely gives up entirely. 61 Whereas an utterance such as, "Spot, you go fetch the stick, go on," clearly does cast the referent of the pronoun as being a participant within a speech event, the "you" unequivocally picking out the addressee (albeit that the canine addressee's response is unlikely to be verbal as such), the "you" in an aphorism such as, "You can lead a horse to water but you can't make it drink," refers to no-body. When the context and nature of the utterance is ambiguous, however, as it is in the opening line of DeLillo's Running Dog (1978), the generic-"you" for "one" may still carry an effect of address. "You won't find ordinary people here. Not after dark, on these streets," says Running Dog's extradiegetic narrator (DeLillo, 1978: 3). Whether the address is to the putative reader, a narratee or a protagonist is a closely related, but, nevertheless, subsequent question. Where the generic-"you" still carries an allocutionary function it is provoked through the insistence of the trace of what is taken as the "second-person" pronoun's grammatical, unmarked case: the "you" as quintessentially a pronoun of address.62 As I will argue, moreover, the effect of address can be deepened by the degree of particularisation or individuation of the "you" referent. A further point about the allocutionary function must be made. In that the standard account of address as developed by Bühler, Jakobson, Benveniste and others assumes deictic relations between a dyadic "I"-"you" couple, the address function has become instrumental in binding the reader securely into Cartesianism's imaginary. It does so by explicitly positing the "I"/ego as the centre of communicative practice and as the "subjective" ground or "origo" of deixis - the centre from which "I," "here" and "now," the three classical terms of deixis, are understood. That the address function should lend itself to Cartesianism, however, is not an inevitable circumstance. In the first place, as I will take up further below, it need not be the case that deixis within instances of address be ego-centric, for as William Hanks proposes, other participant domains of deixis are conceivable. For instance, he writes, one can also discern: the Common ground (sociocentric), the ground of the Addressee (altercentric), and of the Other (of the non-participant in the current speech event) (68). In the second place, as Peter Mühlhäusler and Rom Harré show in Pronouns and People (1990), there is no inevitability that Cartesianism's ego-centred, privatised notion of the self and of Being should subtend communicative practice. Through close scrutiny of a large and varied number of social and grammatical systems, they demonstrate that a culture's conception of the self closely and reciprocally mirrors the pronominal system and grammar in which that conception is articulated (Mühlhäusler and Harré, 1990: 16). The implications of such a situation are not negligible. "What sort of beings we take ourselves to be in [the] ontological or metaphysical sense will depend on the grammar of our language" (Mühlhäusler and Harré, 1990: 20). Traditionally, discussions about addresser-addressee relations have, by and large, conceived of these relations as being fairly straight-forward. The systematic distinctions made between participant roles within literary texts, no matter how elaborate they become--as, for instance, Percy Lubbock's typology of narrators in The Craft of Fiction (1957), and Susan Sniader Lanser's more contemporary reworking of point of view theory in The Narrative Act (1981), in which the narratee/addressee is properly elevated to a position beside the narrator/speaker, or Kacandes's continuum of reciprocity--remain essentially dyadic, in that such systems tend to identify utterances in terms of a speaker and an addressee. The two main approaches to speaking about this relation are the (structuralist) metaphor of the conduit, and the (humanist) model of prototypical conversation. The former approach, that of the metaphor of the conduit, is typified by Shannon and Weaver's communication model, originally devised during the Second World War within the Bell Telephone Laboratories to model the efficient transfer of information (see figure 14).
Fig. 14. Shannon and Weaver's communication model. Chatman's classical model of narrative textuality (see figure 7 above, page 44) clearly takes its lead from the conduit metaphor. Roman Jakobson's communication model, too, is similar to Shannon and Weaver's model in several key respects. "A message sent by its addresser," Jakobson argues, "must be perceived by its receiver. Any message is encoded by its sender and is to be decoded by its addressee. The more closely the addressee approximates the code used by the addresser, the higher is the amount of information obtained" (Jakobson, 1971: 130). Certainly, like literary criticism's use of Shannon and Weaver's model, Jakobson's basic model "situates literary discourse firmly within the same communicative context of other speech events" (Lanser, 1981: 66), implicitly representing the agents of those events as both autonomous and intentionalist.
Fig. 15. Jakobson's communication model. No matter how sophisticated, write Mühlhäusler and Harré, such models remain typical of the way linguists treat human communication "as a process comparable to the exchange of telegraphic messages" (Mühlhäusler and Harré, 1990: 11). They point out that such formulations invite analysts "to concern themselves with a well-known social monster, the ideal speaker-hearer" (Mühlhäusler and Harré, 1990: 12). This choice of metaphor tacitly includes the view that speech communities are homogeneous, that the same code is shared by all interactants, that messages can become signals and vice versa without any problems, that there is no limitation to memory, that production and reception are symmetrical, that there are such processes as encoding and decoding and that what language use is principally all about is exchange of bits of information. (Mühlhäusler and Harré, 1990: 12) Lanser's detailed revision of Jakobson's model in her book The Narrative Act (1981) (see figure 16) complicates this scheme somewhat in at least two ways: first, she proposes that the sender and receiver need not employ the same code nor respond to the same knowledges, beliefs, values, needs and so on; and second, she focuses attention on the manifold nature of communicative context, particularly the communicative contexts of the literary text. She proposes that "the reduplication of the sender's role" in the author-narrator pair, and the "similar doubling of the receiver's role" in the narratee-reader pair, "yields a model for written communication that is more complex than that for spoken discourse" (Lanser, 1981: 117).
Fig. 16. Lanser's revision of Jakobson's communication model. In arguing this, and also in arguing that the "textual speaker-listener construct" will at the very least be homologous (if not necessarily identical) to the relationship between the historical author and audience (Lanser, 1981: 118), however, Lanser's greater degree of sophistication might yet be characterised as a conventional nesting of discourse levels, the author-reader relationship standing parenthetically around the narrator-narratee relationship, itself standing parenthetically around protagonist relationships. Moreover, such a formulation continues to imagine the categories, ideally, as individuated "persons" sending and receiving a narration. Lanser's model remains true to the formal dyadic structure of Shannon and Weaver's communication model. As "pernicious" as the conduit metaphor is, say Mühlhäusler and Harré, it is nonetheless "the most common metaphor in the metalanguage of English" (Mühlhäusler and Harré, 1990: 12). It is pernicious, they argue, because "this conception of communication enshrines the Cartesian myth of mental 'theatres' behind the skilled activities of everyday actors, the myth of mental entities more or less imperfectly represented in public symbolic actions" (12). The second approach is illustrated by prototypical conversation. As Stephen Levinson describes it, this approach is a matter of divid[ing] each such category according to presence/absence from the speech event. . . . Thus we would have speakers who speak for themselves versus those that speak for absent others (spokesmen), addressees who are intended recipients, versus those that are vehicles for a message to absent others (messengers), and third parties who are present (audience) versus third parties who are absent (non-participants). . . . (Levinson, 1988: 166) Levinson's criticism of traditional, simple dyadic and structuralist models is that they reduce all of the possible roles within address events to two paradigmatic terms - speaker and listener. Dyadic schemes that offer no finer categories than author, narrator, character, narratee and reader, Levinson argues, cannot capture the breadth of "the kinds of participant roles actually employed . . . " (Levinson, 1988: 166). Thus, in order to describe speech events adequately, he contends "we need some finer-grained conceptual analysis" of participant roles within address utterances than is offered by traditional models (Levinson, 1988: 167). Levinson's own model divides the roles into ten participant and non-participant producer roles, and seven participant and non-participant reception roles, none of which necessarily excludes the others, so that any address event may involve more than just a speaker and a listener, as laid out below. Levinson's participant and non-participant roles are:
While I agree fully with the thrust of Levinson's proliferation of distinctions at this point, the fact that he divides his roles under the two paradigmatic headings of "production" and "reception" is revealing. For even in his model, the speaker-listener dyad that underpins the traditional models remains substantially in place, although now considerably expanded beyond the reductive terms that he rightly criticises. Levinson is quite aware that his own model shares their basic premise, however, and in his own defence argues that "it does seem that there is something natural enough about the grouping of concepts into 'speaker,' 'addressee' and 'other' to make the classical three-person system recur in most natural languages. . . ." (Levinson, 1988: 183). Herman clearly shares this view, arguing that: it seems that all languages encode through their personal-pronoun systems reference to (at least) the speaker, the hearer or addressee, and the non-speaker or non-hearer. Typically, these three persons and the different numbers (i.e., singular . . . versus plural . . . ) are considered as "the set of contextual anchors" of deixis. (Herman, 1994: 389) Mühlhäusler and Harré respond unequivocally to this type of argument, however, with the assertion that it is not the case at all that "all languages have pronominal categories involving at least three persons and two numbers" (Mühlhäusler and Harré, 1990: 64). Whereas Levinson's quarrel is with the adequacy of standard systems of classification of participant roles within address, Mühlhäusler and Harré's is with the Cartesianism that has come to underpin those systems. Thus, while Levinson's proliferation of roles is a valuable corrective to the narrow purview of the earlier accounts, it needs to be read through Mühlhäusler and Harré's propositions. Mühlhäusler and Harré observe that there can be no doubt of the existence of an "empirical" self, or that as embodied beings they themselves are things amongst things, their actions events amongst events. But don't I need to hypothesise another self, a transcendental being, unobservable, indeed, necessarily so, but a necessary condition for the possibility that my experience should exhibit the orderliness that it does? And is not that the being that I refers to? Everything I say about myself must be grammatically assigned to I, so must not the pronoun then function as some kind of referring device by which that inner core of being, my transcendental self is picked out? (Mühlhäusler and Harré, 1990: 20) Mühlhäusler and Harré's unequivocal answer to the final question is no. "It has long been realised that there is something very wrong with this line of thought" (Mühlhäusler and Harré, 1990: 17). Rather, what they show is that "the transcendental ego is a shadow cast on the world by grammar" (Mühlhäusler and Harré, 1990: 20). They make the implications of such a position to linguistic analysis quite explicit. "To use pronouns grammatically, correctly, one must deploy one's philosophical theories of what one is as well as one's knowledge of the social relations in which one stands to those with whom one converses" (Mühlhäusler and Harré, 1990: 16). Having observed that there can be no doubt that their own actions are events amongst events, there can likewise be no doubt that their participation in speech events entails the adoption of specifiable discourse roles structured by the grammar particular to the language and the social relations of the event, and of specifiable conceptions of the nature of their being (Being) within such roles. That is, to use pronouns, to say "I" and "you" and "they," is to mobilise particular ontological and epistemological conceptions of what it is to be a person. 11. The Specification of "You" and the Depth of Address The two paradigms of address discussed above, of the conduit and prototypical conversation, are the principal conceptions of address used in describing or accounting for the depth of the allocutionary function in "second-person" fiction. Both Fludernik and Kacandes, for instance, draw explicitly upon the conversational model. In turn, the addressee identity nexus and Hantzis's conception of "second-person" point of view proper as involving an oscillation of reference both owe much to the conduit model. 63 One might observe, too, though, that the two paradigms are not exclusive of one another. In so far as Kacandes must still identify who addresses whom, the conduit model in some measure subtends the conversational model. There is, however, a third approach. The depth of the effect of address of the allocutionary function can also be construed in the more formal terms of grammatical structure. Herman draws our attention to the fact that the "second-person" pronoun's deictic profile and its grammatical, morphosyntactic form can become disengaged from one another. By "morphosyntactic form," Herman means the aspect of language of which the pronoun "you" is a conventional constituent. Its proper place in the overall structure of the English language is within the grammar of address. Many instances of the "second person," he shows, exhibit an incongruity between grammatical form and deictic function in that they cease to retain the full shifting character of deixis expected of them. Such disengagement of form and function, in fact, lends authority to Kacandes's arguments. Dialogic reciprocity might be taken as a consequence of the full agreement in an utterance of deictic and morphosyntactic form, and non-reciprocity as a consequence of their disagreement. Indeed, Herman's continuum closely echoes Kacandes's model, his pole of incongruent form and function and its typical narratives corresponding with Kacandes's non-dialogic pole, his pole of congruent form and function corresponding with Kacandes's dialogic pole, and his doubly deictic case corresponding in type, if not in position on the respective continuums, to Kacandes's radical narrative apostrophe. 64 Herman's insightful move has been to recast this scheme in grammatical terms, moving away from the tendency to imagine - and circularly to analyse - its instances in terms of privatised selves acting out communicative acts. Herman's second move, following Hanks's lead, is to shift the underpinning grounds of deixis (for at least one mode of "second-person" textuality) away from conventional ego-centric conceptions of deixis towards one that is principally sociocentric.65 The case for reconceiving deixis as sociocentric rather than ego-centric might advantageously begin, as Peter Jones writes, with Voloshinov's and Bahktin's observations on the social nature of language. Jones writes: "Each and every utterance should be registered as a social event, as a unit of 'speech communication' whose essential quality is its 'addressivity,' that is, 'its quality of being directed toward someone' (Jones, 1995: 46). Volosinov properly asserts that the "organising centre of any utterance, of any experience, is not within but outside - in the social milieu surrounding the individual being" (cited in Jones, 1995: 46). Bakhtin puts it this way: from the very beginning, the utterance is constructed while taking into account possible responsive reactions, for whose sake, in essence, it is actually created. As we know, the role of the others for whom the utterance is constructed is extremely great [because] the role of these others, for whom my thought becomes actual though for the first time (and thus also for my own self as well) is not that of passive listeners, but of active participants in speech communication. From the very beginning, the speaker expects a response from them. . . . The entire utterance is constructed, as it were, in anticipation of encountering this response. (Bakhtin, 1986: 94) Jones concludes that to claim that "the speaker's utterance, including the use of deictic terms, is ego-centric or speaker centred is quite wrong; what the speaker says is 'address-centred,' the viewpoint and expected response of the addressee is already, as it were, built into the utterance itself" (Jones, 1995: 47). Reasserting the social, communicative nature of language, our starting point for the investigation of deixis, he writes, "would not be what is individual, but what is shared, i.e., social activity, practical cooperation in the world, mediated by language" (Jones, 1995: 47). For supporting evidence, he looks to Bruner's discussion in "The Organization of Action and the Nature of Adult-Infant Transaction" (1982) on the development of language. Here, writes Jones, language is shown to emerge between mother and child in common contexts of activity. These shared contexts are not perceptual phenomena but goal-directed activity structures in which words, including deictic words, emerge and come to mean what they do as "slices" or phrases of a particular action guided and controlled by shared intention and expectation. The deictic field would therefore be not so much a passive registration of a "canonical situation" or external environment but a synthetic "model" of the real domain being altered, or to be altered, by creative social action, a position which would allow for constant dialectical interaction between the activity domain and its model. (Jones, 1995: 47) Jones also proposes that we should consider deixis in the light of what Bakhtin says about the role in communicative language of "speech genres." Bakhtin describes these as "relatively stable and normative forms of the utterance." He writes that: Speech genres are much more changeable, flexible, and plastic than language forms are, but they have a normative significance for the speaking individuum, and they are not created by him but are given to him. Therefore, the single utterance, with all its individuality and creativity, can in no way be regarded as a completely free combination of forms of language, as is supposed, for example, by Saussure (and by many other linguists after him), who juxtapose the utterance (la parole), as a purely individual act, to the system of language as a phenomenon that is purely social and mandatory for the individuum. (Jones, 1995: 80-81). According to Jones, this approach to deixis would "attempt to discover the typical stylistic and generic resources available to users of a language . . . on the assumption that the speaker's very selection of a particular grammatical form is a stylistic act" (Jones, 1995: 47). Such an approach, Jones continues, would "encourage scepticism" towards the traditional view that there is a basic or general meaning for deictic expressions as Lyons posits in "Deixis and Subjectivity: Loquor, Ergo Sum?" (1982). And it would emphasise, instead, "the heterogeneity or meanings and uses of apparently one and the same expression" (Jones, 1995: 47). Jones proposes that, with these ideas in mind, "one might explore the hypothesis that deictic words are always sociocentric in orientation, because of the very nature of verbal communication" (Jones, 1995: 47). This approach would enable - and encourage - us to differentiate "speaker-" and "hearer-centred" and differently oriented deictic values "according to the nature of the activity, the nature of the participant roles, and convention" (Jones, 1995: 48), rather than to assume from the outset that deictic utterances must be ego-centric.
Fig. 17. Herman's continuum between grammatical form and deictic function. Herman's exploration of deictic values brings him to draw a continuum on which "the degree of concord between grammatical form and deictic functioning" is high at one end and low at the other (Herman, 1994: 393). He identifies five paradigmatic uses of the "you," two of which he places on each opposite pole, the fifth of which is placed somewhere between (see figure 17). 66 On the right-hand pole Herman places "occurrences of you in which the form of the expression exactly specifies its function," the pronoun's morphosyntactic form and its deictic function agreeing (Herman, 1994: 393). Agreement produces either fictionalised address, which concerns instances of direct dialogic discourse between characters, or apostrophic address (my mesofictional- or metafictional-"you"), which is address directed "toward an actual . . . addressee" (Herman, 1994: 394). On the left-hand pole are the terms that exhibit incongruity. In the generalised "you" case, the "you" pronoun takes on the now familiar role "of an impersonal or generalised or colloquial you," the pronoun virtually losing its deictic force (Herman, 1994: 392). In the "fictional reference" case, the "you" pronoun "entails the displaced deixis of an I —> you deictic transfer" (Herman, 1994: 392), in which the "you" points back at the utterer as a dissembling "I" within mere self-reference as opposed to self-address. That is, it retains a deictic element, but pointing to the addresser, holds the function grammatically proper to "first-person" deixis rather than "second-person." This would appear to be the case that Fludernik describes as "reflector-mode you," the "you" that functions as interior monologue in which "the narrative disappears entirely behind the thoughts of the protagonist 'you'" (Fludernik, 1994c: 451). Herman makes a point of distinguishing this case from self-addressed modes of "you," positing that "the phenomenon of self-address . . . constitutes a kind of hybridised category" that combines features of both case 2, fictional reference, and case 3, fictionalised address (Herman, 1994: n. 5, 405). Given that his principal example of case 2 is Butor's La Modification, it is to be inferred that the narrator must be more or less explicitly marked as a self-conscious "I," as in the texts discussed at the end of Chapter 6 (cf. Fludernik, 1994c: 451). But even in texts in which the "I —> you deictic transfer" is much less equivocal than that of La Modification, such as Kocan's The Treatment and The Cure (see pages 147-48 below), it is not necessarily the case, I suggest, that the address function is disengaged. Indeed, as is attested in Chapter 6, it is far from conclusive that the "you" of Herman's paradigm text, La Modification, itself relinquishes any of the "second-person" pronoun's effect of address. In contrast to Herman's proposition, one might consider the breadth of response Kacandes reports with respect to La Modification (see page 5 above). Such texts do disengage the deictic profile of you from its grammar, but each nonetheless carries to the reader the sense that an address is being made, an address that invokes the inference of particular anthropocentric, interpersonal relations, even where we struggle to identify the entities involved, and even where we struggle to identify the roles that might hold within the proliferation of cases revealed in Levinson's catalogue. Where deixis becomes incongruent with the grammatical profile of an address utterance, it can be shown that the allocutionary function is not necessarily undone as one might expect it to be. The "grammatical profile" of the "second-person" pronoun can, as Herman himself puts it, drastically undermine the pronoun's deictic functions (Herman, 1994: 390). This, I believe, is what Brian McHale is reacting to when he suggests that the reader is fully justified in endeavouring to identify the addresser and addressee and the nature of "the communicative circuit that joins them" in any "you" utterance (McHale, 1985: 95). If the deixis that is proper to the "second-person" pronoun can be disengaged, as it clearly can, then it must be concluded that the particular feature that enables the "second person" to convey "some sense of address, even in its most 'innocent,' impersonal instances" (McHale, 1985: 112), inheres in some way in the utterance's grammatical, morphosyntactic form, and operates, to invoke Bakhtin's term, much like a "speech genre." What I mean by this is no more than that we recognise or register an instance, a moment, of address in the utterance's formal structure, in its appearance. Even when the pronoun has adopted a figural or narrative modality and relinquished the conventional deictic function required for the full functionality of dialogic address, the insistence of the morphosyntactic form of address left as a trace in the "second-person" pronoun can carry the allocutionary function forward. On the other hand, the more forceful effect of address does, by and large, correlate with instances of greater congruence; and conversely, where the grammar and deixis of "you" in the generic cases display little agreement and there is little anthropocentric particularisation or specification, the allocutionary effect will be correspondingly shallower. 67
Fig. 18. Depth of Address model My suggestion, illustrated in figure 18, is that the force of address returns to narrative-"you" precisely for the "you"-protagonists' particularisation in narrative, and that its force will more or less correspond to the degree of particularisation of the protagonist. Hence, the effect of the allocutionary function in Maugham's "The Beast of Burden" (1925) is somewhat deeper than the effect of the aphoristic "you" for "one." Maugham's generic-"you" figure is subjected to a certain amount of particularisation in terms of circumstance, albeit that this particularisation is itself non-unique (see page 78 above). The depth of address model - set for useful comparison beside the generic-particular model - illustrates the way in which anthropocentric particularisation of the pronoun's referent puts Herman's model under stress. The anthropocentric particularisation of the pronoun's referent forces an effect of address back into the narrative discourse regardless of the pronoun's deictic incongruence. In skaz, caught between the generic and the mesofictional cases as it is, for instance, considerable agreement between form and function remains but the allocutionary force is now softened, as it were, by the generality of the address and generic identity of the addressee. The pronoun continues to designate an addressee in each utterance and to sustain a strong sense of dialogism. 68 But this reader/listener is no longer somebody in particular: he or she simply provides the opportunity for the tale to be told. Skaz-styled works usefully demonstrate another significant feature of the address function, however. Vernacular forms of discourse and address-related mechanisms (such as the speaker's immanence in vernacular constructions and the phatic function) enhance the allocutionary force of any utterance of address. This enhancement, indeed, is the reason skaz reaches so far toward a sense of dialogism in spite of the amorphous nature of its addressee: as inappropriate as it would be for the addressee to interrupt the speaker, the immediacy of the discourse seems to allow for such interruption. Indeed, in oral skaz, the addressee may very well speak, albeit that the sole appropriate form of such speech is encouragement and support for the story teller, utterances that articulate the listener's engagement with the story-telling itself. In reference to the vocative wing of the generic-particular model (which accounts for Herman's apostrophic case), it is proper and expected that the pronoun's referent be particularised. As Barbara Johnson puts it, this is part of the explicit, rhetorical undertaking of classical apostrophic address: "The fact that apostrophe allows one to animate the inanimate, the dead, or the absent [by addressing it directly] implies that whenever a being is apostrophised, it is thereby automatically animated, anthropomorphised, "person-ified" (Johnson, 1987: 191). In reference to the narrative wing, and disbarring the eventuality of some hybridised relationship such as explicit self-address, Herman's argument would conclude that, as particularisation increases, the sense of address that holds will remain shallow and easily passed over. Reading narrative-"you" texts, however, one tends to experience something quite different. In many instances, therefore, as particularisation increases, so, too, does the depth of the allocutionary function. That is, to some extent, in the same way that apostrophe will tend to anthropomorphise its object simply because it can, "second-person" utterances that refer to an anthropomorphic, particularised object will tend to be construed in some sense as articulating an address, just because they can. The reader registers the trace of address in the "second-person" pronoun's morphosyntactic form, and feels that he or she should be in the presence of address. 12. The Sociocentricity of Deixis I have not as yet spoken about Herman's fifth case, "double deictic you." Herman insists that this case can only emerge "against the background of the other [four] modalities" and their hybrids (Herman, 1994: 397). Its "perceptibility" within texts "derives from its (fluctuating) position relative to a ground" comprised of the other four functional types. He provides the following illustration from O'Brien's A Pagan Place (1984). You were saying goodbye to fields and trees, and even to headlands of fields where a plow never got and where not an ear of barley had chanced to grow. In all these corners there were bits of things, machinery, broken delft, cowhorns that had served as funnels, machine oil tins and the rags and remnants that the scarecrows wore. You felt a terrible burden as if something inanimate might speak or something motionless might get up and move. (Cited in Herman, 1994: 400) By the end of this passage, Herman asserts, the condition has been produced for a "doubly deictic projection of a fictional you onto the audience and vice versa" (Herman, 1994: 400). The passage cannot be completely reduced to the you encoding, either via reference to the fictional protagonist or via self-address, a participant who is located somewhere in the indexical field of the current discourse. Yet the you does not stand in for one, either, since the amount of circumstantial detail built into the description . . . cannot be reconciled with the notion of an impersonal or generalised you, nor for that matter with an apostrophic you in the strict sense. Again, the deictic scope of the you encompasses more than a particularised addressee or a specific discourse participant but stops short of including everyone whatsoever. The scope of you is modalised, as it were, such that it covers anyone who might conceivably be a participant in the discourse; you ranges over any context that might be activated and brought to bear on the discourse. (Herman, 1994: 400) The similarity of this description to Kacandes's notion of radical narrative apostrophe is clear. Herman's double deixis is named not for its hybridity, however, but for the reason that it "produces an interference pattern between two or more competing deictic fields, none of which can fully orientate the deictic transfers" (Herman, 1994: 398). Because of these competing fields, writes Herman, "the scope of the discourse context embedding the description is indeterminate, as is the domain of participants in principle specified or picked out by you," freeing the reader to implicate him or herself or any other available entity in the indeterminate deictic reference (Herman, 1994: 399). This fifth term, I suggest, is the most valuable element of Herman's formulation of the "second person's" address function. It is this term that shifts the model from the ego-centric, standard account of deixis to the more open, sociocentric deictic fields posited by Hanks. In "The Indexical Ground of Deictic Reference" (1992), Hanks proposes that: In theory at least, one could imagine any number of alternative indexical pivots, logocentric, person-centric, event-centric, and so forth. Given that acts of reference are interactively accomplished, a sociocentric approach is certain to be more productive than an egocentric one, even when the speaker is the primary ground of reference. (Hanks, 1992: 53) Although Hanks freely grants that deixis requires an origo for communicative sense to be made of the utterance, given that the notion of "pointing to" is rendered nonsensical without the specification of the instance of "pointing from," he proposes that one need not assume this origo to be the speaker, the "first person." Deixis is therefore, for Hanks, not to be characterised as being fundamentally "subjective" in the way that proponents of the standard account suppose it (Hanks, 1992: 52). Rather, he writes, it is better to think of deixis as "a framework for organising the actor's access to the context of speech at the moment of utterance" (Hanks, 1992: 61). To be sure, speaking subjects can and do act as the origo of deictic utterances, but it is neither necessary nor inevitable that they do so. One of Hanks's illustrations is of a fragment of conversation between two people working in a field, and would appear to involve the participant domain he refers to at one point as Other. The first person (designated A) says to the second (B) about a third person (C, who constitutes the participant domain of Other): "so then he says to me, 'Come here (to me)' he says, so I went to him" (Hanks, 1992: 55). In order to make the illustration more pointed in the context of a discussion of the "second-person" pronoun, one might rewrite Herman's illustration to make the implicit addressees more conspicuous. A says to B about C, "so then he says to me, you know, 'You come here to me,' he says, so I went to him." Hanks points to the segment marked as quoted speech, noting that, Although A utters the standard directive for summoning an addressee. . . , B understands immediately that A is not calling him to his side. He also understands that the referent of [here] does not refer to A's current locus, but to C's locus at the time of the original utterance. [ . . . ] (Hanks, 1992: 55-56) Likewise, while B would understand himself to be the referent of the first "you" uttered, he would also understand that the "you" within the segment of speech quoted by A does not refer to him, but to A himself. "It follows," writes Hanks, "that one and the same deictic is interpreted in one way in quoted discourse, but quite differently in direct [discourse]" (Hanks, 1992: 56). The type of shift in reference illustrated here, he continues, is well enough known, and "is what motivated Jakobson's original description of these forms as 'shifters'" (Hanks, 1992: 56). But Hanks goes on to insist that although the one deictic, "here" (and in my augmented version of his illustration, "you"), can be interpreted quite differently, this does not mean that it is necessary to construe "that the relational values reverse from use to use" as is the conception in standard accounts. Rather, writes Hanks, the shift can be accounted for by saying that quotation involves a transposition of the indexical ground of reference. The Relational features of the forms remain constant in both types of discourse, but the origo is projected in quoted speech from the actual utterance framework into a narrated one. (Hanks, 1992: 56) That is, the origo of the utterance can no longer be said to be exclusively held to the utterer, to be ego-centric, but might also be seen as involving an instance of Other-centred deixis. Moreover, as Hanks observes, speakers "routinely interact across boundaries of various kinds," speaking from room to room, across a work site, across open distances and so on, with deictic utterances reorienting the speakers "to a relatively great degree, creating a reciprocal or common focus of attention" between the participants where initially there was none (Hanks, 1992: 67). Thus, whereas "[i]n many deictic acts, an already constituted deictic framework is presupposed, in which interlocutors share certain relevant knowledge, immediate experience and engagement," in others, the origo is not already in place, so that the deictic acts must immediately create a framework, "an interactive relation that did not exist prior to the utterance" (Hanks, 1992: 67). This play between "presupposed" and "creative" aspects of deixis, Hanks argues, is ongoing, revealing that "the indexical origo is a dynamic ground, rather than a fixed object" (Hanks, 1992: 67). Herman's double deictic case of "you" is an attempt to conceive of the Protean-"you's" shiftiness precisely in terms of this dynamic ground. Kacandes's spectrum of reciprocity (a dialogic model) and Hantzis's "second-person" point of view proper (implicitly a conduit model) both continue to conceive of the problem through the ego-centric ground of the standard account of deixis - and indeed, both draw heavily on Benveniste's account of the shifter. In double deixis, writes Herman, "there is no longer any ultimate (nonimaginary) reference point anchoring localised . . . shifts in space and time." Neither a term of address nor not a term of address, doubly deictic you ranges over the middle portion of a continuum whose lower and upper limits, respectively, are marked by the two limit cases of virtualised and actualised you. Hovering between these two extremes, double deixis ontologically destabilises a modal system that can no longer be neatly divided into the virtual and the actual. (Herman, 1994: 398)69 Interpreting double deictic "you," Herman writes, "requires that we abandon what Hanks has characterised as the 'assumption of egocentricity' [in favour of a] 'sociocentricity of deictic reference'" (Herman, 1994: 400). Herman argues that, in reading O'Brien's narrative, we (the audience) are able to adopt the role of participant precisely because discourse in general encodes reference to a set of potential addressees even when pointing to an actual addressee. The doubly deictic you of second-person narratives suggests that there can be an addressee just because there could be other addressees: that is, what we deem to be actual speech situations are just part of a larger network of (more or less) virtualised speech situations toward which the current discourse is constantly tending and from which it never ceases to emerge. . . . [H]earing can no longer be neatly distinguished from overhearing. We are eavesdroppers on the discourse that addresses us and beckoned by discourse addressed to others. (O'Brien, 1984: 401) I suggest that there is a particular ambiguity in Herman's working through of doubly deictic "you" at this point that should not pass unaddressed. Double deixis seems to be characterised as a multiplication of "I"-"you" couples, therefore as nothing so much as a tabling of all who may speak from a deictic centre (i.e., as the origo) and all who may be pointed to from that centre. Herman is fully aware, however, of the much deeper implications of Hanks's propositions, and his final remarks make it clear that he intends these to be played out fully in the notion of double deixis: Hanks is not speaking merely of a loosening of deictic reference between available speech event participants. Herman proposes in the final section of "You and Double Deixis in A Pagan Place" (1994) that the ground of deixis - and of double deixisis not constituted by a range or set of more or less reciprocal relations of deictic self-centred-ness. He insists that the familiar use of speaker-centred deictic terms is not evidence for their primacy, nor for the primacy of egocentric orientation in general. And he makes clear, too, that linear models are not adequate to describe the address effects of double deictic "second-person" narrative. Rather, the allocutionary effect is better conceived of as specifiable within a field of intersecting constituents, or, more properly, amongst overlapping fields of deixis. As Herman elegantly and justifiably claims about his case of double deixis in his closing remarks, doubly deictic you reveals that context itself is never petrified and immobile, but dynamic and vegetable, living, growing with the multiplicity of linguistic forms it embeds and in which it is embedded. It resembles the lichen that you sometimes see growing on and through a wall, lichen that may have 'passed into the body of the stone and . . . become part of it. There was white and green and rust-coloured lichen and there were queer shapes, all watery at the edges, like the borders of countries on the school map.'" (Herman, 1994: 404)70 The deictic origo therefore must be understood to be dispersed and dispersing, rhyzomic, its indexical framework of reference always changing as "interactants move through space, shift topics, exchange information, coordinate their respective orientations, and establish common grounds as well as non-commonalities" (Hanks, 1992: 53). The origo, observes Hanks, is neither a psychological nor metaphysical given but is an entity "bound up crucially in the interaction between participants" (Hanks, 1992: 68). To summarise, then, I would identify three aspects of "second-person" textuality as contributing to the depth or strength of the effect of the allocutionary function. These three aspects, I propose, are most felicitously described in the context of a reference model of "second person" textuality. Such a model offers two advantages. First, the model permits the critic/reader to construe its categories of agency - the "you," the narrator, the reader and narratee, and the "no-body" of the figural, to name the categories at their most coarse 71as quintes |
