He Who Says “Ego”

The three misconceptions previously mentioned—the belief in an I that thinks, in thoughts prior to linguistic expression, and in language as a neutral medium of communication—all relate to a construal of the subject in substantial terms, a subject that may indeed come to itself self-reflexively in language but where the core of the self is posited as a prelinguistic datum. This position often relies on a form of consciousness that is directly present to itself, rather like Aristotle’s conception of deity (noesis noeseos). In opposition to this way of thinking I have sketched a view of self where language takes center stage, especially in the form of narration with its implied subject. As Calvin Schrag has expressed the matter, “The event of self-consciousness is inseparable from the history of saying ‘I’.”⁶ Such a history is primarily that of autobiographical acts, for a meaningful self-consciousness is, as previous chapters have attempted to show, synonymous with self-narration and therefore with self-interpretation. Benveniste sums up this linguistic position:

It is in and through language that man constitutes himself as a subject, because language alone establishes the concept of “ego” in reality, in its reality which is that of the being.

The “subjectivity” we are discussing here is the capacity of the speaker to posit himself as “subject.” It is defined not by the feeling which everyone experiences of being himself (this feeling, to the degree that it can be taken note of, is only a reflection) but as the psychic unity that transcends the totality of the actual experiences it assembles and that makes the permanence of the consciousness. Now we hold that “subjectivity,” whether it is placed in phenomenology or in psychology, as one may wish, is only the emergence in the being of a fundamental property of language. “Ego” is he who says “ego.” This is where we see the foundation of “subjectivity,” which is determined by the linguistic status of “person” (PGL, p. 224).⁷

Subjectivity is attained in discourse by assuming the role of “I” in that discourse. “I” designates this speaking subject at the instance of utterance, just as it designates other speakers in their turn. “‘I,’” says Benveniste, “signifies the person who is uttering the present instance of discourse containing ‘I’” (PGL, p. 218). But, as Benveniste also reveals, “I” always functions dialogically with an addressee, “you” (singular or plural), and it is here that language guarantees the possibility of sociality and intersubjectivity (PGL, p. 225). One cannot become “I” without an implicit reference to another person, an auditor or narratee—which may be the same subject qua listener. “I” functions in contrast to “you” in much the same way as “here” refers linguistically to “there” rather than to any fixed location. In Lacanian terms, “there” is the other of “here,” “you” is the other of “I,” and vice versa. Discourse always has its other, which is one way of restating the Saussurean claim that signifiers gain meaning in relation to other signifiers, to a chain of signifiers that is excluded by the utterance but presupposed by it.

“I” has fundamentally a locutionary reality, setting up what I shall call a subject of speech (le sujet de l’énonce), a form of subject that exists solely in the expression. In the ordinary train of discourse this linguistic subject tends automatically (habitually and implicitly) to be predicated by listeners and readers of its author, which I term the speaking subject, which may be the bodily site of the enunciation or the origin of inscription for written language. This form of predication can, I believe, be seen as fundamental for generating what are generally termed “persons” (embodied subjects). But we must not forget that the subject of speech has an important autonomy from its site of production—it may be reproduced in a text or on a tape recorder, and of course it may be purely fictitious.

The main point here is that the subject of speech does not bear a one-to-one relation to the speaking subject, as though the “truth” of the latter is necessarily mirrored in the former. Rather, we should see here a parallel to the dialectical mechanism of prenarrative-narrative that was discussed in the previous chapter. The meaning of the speaking subject (for itself and for others) is only given via its discourse, in which, if not identified with the thematic subject and content of the utterance (as in autobiography), it at least becomes the authorial subject of the utterance. In itself the speaking subject is simply a possible site of utterances, a semiotic body of potential gestures and articulations through which it will make itself known as a particular subject with particular concerns and perspectives on reality. The speaking subject, then, attains selfhood via its expressions—much as the prenarrative attains its expression and fulfilment in a narrative (such as autobiography or history).

Now the obvious rejoinder to this general linguistic position is, once again, that language is, to the contrary, simply more or less adequate to expressing what is prelinguistically already given to us, as individual subjects, in our experience. In other words, I already am prior to expression, and I employ language primarily to communicate with others. But let us look more closely at this prelinguistic realm. Though we undoubtedly have a bodily existence apart from language, I am claiming here that it is in and through language that the dimension of the subject, the self, is generated.

Consider this statement by Helen Keller, who began life as blind, deaf, and mute: “When I learned the meaning of ‘I’ and ‘me’ and found that I was something, I began to think. Then consciousness first existed for me.” Prior to this self-consciousness, she writes,

I did not know that I knew aught, or that I lived or acted or desired. I had neither will nor intellect. I was carried along to objects and acts by a certain blind natural impetus. I had a mind which caused me to feel anger, satisfaction, desire. These two facts led those about me to suppose that I willed and thought.

These statements bring out perfectly the effect of language for providing a position and identity for the human subject. Prior to the appropriation of “I,” she says, “my mind was in a state of anarchy in which meaningless sensations rioted, and if thought existed, it was so vague and inconsequent, it cannot be made a part of discourse.”⁸

If we can trust Keller’s description, there simply was no reflective consciousness, no point of view from which future acts are assessed and the past reflected on. Thus there simply was no “I” that she was for herself. The only “I,” or the only notion of personhood, that existed was the one predicated of her and filled out by others. Yet despite this situation it seems so natural for us native speakers to attribute selfhood, intention, and conscious deliberation to others that we do so to very young children and sometimes even to animals!

Another interesting point arises from Keller’s observations. What I have called the prenarrative level of experience seems not to have existed in any organized fashion; instead there is primarily a riot of sensations and impulses which, as Keller aptly put it, “cannot be made a part of discourse.” We find confirmation of this state of affairs in the work of Saussure:

Psychologically our thought—apart from its expression in words—is only a shapeless and indistinct mass. Philosophers and linguists have always agreed in recognizing that without the help of signs we would be unable to make a clear-cut, consistent distinction between two ideas. Without language, thought is a vague, uncharted nebula. There are no pre-existing ideas, and nothing is distinct before the appearance of language.⁹

We thus see that without language, without a modicum of self-narration occurring during the course of one’s life, even one’s unreflected or pre-conscious life loses structure, loses some of its implicit narrative.¹⁰ This leads us to the conclusion that if we have not been brought up with stories we cannot expect to find them in our lives, cannot expect to live them.¹¹ Indeed, the problem here is that the prenarrative and the narrative levels are always intertwined; their histories continually cross and intermix.

Let us be reminded of Benveniste’s central tenet, now that it might have a more concrete value:

it is literally true that the basis of subjectivity is in the exercise of language. . . . there is no other objective testimony to the identity of the subject except that which he himself thus gives about himself.” (PGL, p. 226)

We should especially note the implications of this statement. What, for example, are we to do with patients in a catatonic state? Are they still subjects, still persons? On Benveniste’s view the question remains open, but borders on the negative. They are generally granted subject status for us only because they once were speaking subjects—we tend to give them the benefit of the doubt. But if self-consciousness is correlative to language, then only a solipsistic interior dialogue can continue to sustain them as subjects, a fact that is most difficult to ascertain.

It is interesting—in films, for example—how even a computer can become a subject (and be treated as one at the discursive level) if it responds with a voice that uses the first-person singular rather than the drone of impersonal information we might otherwise expect. HAL in Stanley Kubrick’s 2001 is an obvious example of this subjectifying brought about in speech, and it often occurs without a visible and localizable physical body. Our positive responses to such, usually fictitious, cases can be revealing. They point, among other things, to a certain acceptable independence of the speaking subject from its concrete state of embodiment—something we encounter regularly with fictional characters in novels.

HAL was successful as a subject because “his” speech respected the dialogical nature of the personal pronouns; he could become “you,” the addressee. The “subjectivity” of HAL is different from the subjectifying of animals I mentioned earlier, for here it comes primarily from HAL “himself,” not from our own projecting. We are caught up in the dialogic situation his speech engenders; we ourselves are addressed as subjects.¹²

Jacques Lacan has said of communicative interaction that “if I call the person to whom I am speaking by whatever name I choose to give him, I intimate to him the subjective function that he will take on in order to reply to me, even if it is to repudiate this function.”¹³ It is surprisingly difficult to avoid this “intersubjective” dialogue altogether. Of course, much does depend on the status of the reply, for I may either recognize or refute others as the types of subject they set themselves up to be, or may deny their subjective status completely by simply not responding to the speech situation at all. This latter case, however, may still form two categories: (1) relegating the other to total object status (“only a machine”), which truly negates the other as subject; (2) alienation of the other, which still therefore presupposes subjectivity. This second category may, for example, take the form of a master-slave relationship where only a quasi-subjectivity is granted.¹⁴

Can HAL ever be a “real” person? Not by our standard societal definition, for this does seem to require, among other things, embodiment in what we take to be a human form. Persons tend to be commonly conceived as a soul or mind plus body; we encountered something similar with Locke (concept of “man”), and it goes back at least to Plato. Of course this state of affairs could change if our conception of embodiment changed.¹⁵

In a face-to-face dialogue it is the other’s “body” that speaks to me. The other’s “body” becomes both the site of narration and the site of ascription for subjectivity. By this I mean that the subject, the “I” of the other’s discourse, is attributed to a certain spatial location, the perceived origin of the voice. This physical body, the site of narration, thereby becomes endowed with the status of selfhood, becomes thereby a distinctly human body, a person rather than a mere impersonal mechanism or animal.¹⁶ Even when the speaker is absent some form of ascription to an embodied authorial origin generally occurs. This act appears to be fundamental to our society.

The act of ascription of selfhood is a prime factor in the generation of what we call “persons,” but it is also a factor in the formation of potentially troublesome concepts such as soul and mind. The soul (also the mind) has been traditionally viewed as distributed in the body in some way, though it is especially associated with the head and throat regions, both by the speaker and the addressee. William James points this out when he says that the spiritual self, or what is commonly taken for it, is really only those intimate and diffuse but continuous bodily motions or pressures occurring at the head and throat and does not necessarily indicate the presence of a soul substance. These motions are thus assumed to be the seat of the self, the feeling of subjectivity, or what James calls “the real nucleus of our personal identity.”¹⁷

The head is the most obvious site of ascription for a personal self that originates primarily in voice (mouth and throat), and whose world is eminently visual (eyes). We must be careful, however, not to reify these so-called “spiritual” entities—soul, mind, self—and also not to set up unnecessary dualisms of mind and body substances. As we hope to have shown, the self (mind could also be included) is not a substance or thing at all; to consider it so would be to commit Ryle’s category mistake. It is not therefore localizable in any but a derived sense. One becomes a subject for oneself, or one has a self, within a speech community where the “I” and “you” are played out. It is our inveterate habit of substantializing or reifying this “I,” of giving it a being other than as expressed in the social praxis of discourse and at the moment of utterance, that leads us to the belief in localization and also, because it cannot of course be discovered at that locale (consider Descartes’ glandular problem!), leads to an essential separability from the body. Ignoring the important functioning of language has created numerous such problems in the history of philosophy, problems that we are only now learning to look at from this new perspective—or simply to discard as totally ill-formed.

Signs of Derrida

Let us look more closely at the second misconception enumerated at the beginning of this chapter, particularly with a view to understanding further the nature of signification and meaning at the level of soliloquy—the “dialogue of the soul with itself.” Soliloquy is especially important to the question of self-identity conceived narratologically and also to the function of shifters (“I,” “you,” “here,” “there,” “now,” etc.). The second misconception can be formulated as follows: the belief in intentions or thoughts that exist prior to linguistic expression.

What is often supposed in ordinary discourse is that expression carries over in a communicative act what was initially given or known prelinguistically. This prelinguistic realm is considered to be a realm of thought or direct intuition that serves as a preconceptual origin and touchstone for the meaning and truth of our expressive utterances and statements. For perception this distinction is especially clear. I say “it is snowing outside,” and this statement is presumably verified in my own perceptual experience and can be similarly verified by the addressee if he or she looks out the window. This position is essentially the one proposed by Edmund Husserl, whose phenomenology is one of its most detailed expressions. If we pursue Husserl’s thought, primary in his earlier works, it will take us directly to a consideration of expression and soliloquy. We shall, in effect, be arguing against a Husserlian position.

For the early Husserl, the meaning (Bedeutung) of an expression is fulfilled in intuitive self-evidence or in the sense (Sinn) of extralinguistic experience, just as meaning may be taken as an expression of such a sense on the part of the speaker.¹⁸ A prime function of expression, which requires a signifying medium, is thus to communicate a pregiven sense or intended content, a sense that is presumed to be directly present to oneself and which one seeks to indicate to others. Language (and we will take exception to this view) is then conceived of, in its ideal state, as a transparent or self-effacing conveyance of meaning from one interiority to another, from one “soul” to another. The goal of this communication is intuition, conceived of as the presence of the state of affairs (the referent) in itself, though one may of course remain at the level of meaning and simply accept the other’s word. This latter situation is a very common practice; it indicates an important difference between meaning and truth. For Husserl, meaning is fulfilled in intuition, much as Kant’s “empty” concepts are, or in the presence of the object intended (an object that could also be fictional or intellectual). It is with fulfillment that the truth of an utterance is directly attained.

There are a few distinctions to be made here. Signs may be broadly divided into two primary camps, which I shall call, following Husserl, “indicative” (or “indexical”) and “expressive.” The former functions like my pointing finger, or like smoke that indicates fire; no intermediary level of meaning is required; one simply has to know the ostensive convention or make the relevant association: “Every sign is a sign for something, but not every sign has ‘meaning,’ a ‘sense’ that the sign ‘expresses.’”¹⁹ Expressions, on the other hand, involve meaning, and are not therefore in a one-to-one relation to things. Expression (Ausdruck), as in written and spoken language, operates through a degree of ideality; in fact, expressions refer in the first place to an ideality rather than to a reality. We can also say that expression expresses the meaning of things or states of affairs; it raises the sense of things to the level of communication.²⁰ Meaning thus requires the existence of a signifying medium (or legible structure) that allows for recognition and repetition of signification and meaning within a community. Without this iterability there could be no meaning, no communication from person A to person B.

It is essentially this iterability that allows meaning, and hence expression, to break with the empirical. “It is snowing,” is meaningful whether it is snowing or not; it is meaningful now as later.²¹ The ability of writing to communicate over centuries is a clear indication of this break, a break even from the moment and context of inscription and its psychological associations and implications.²² Thus Derrida: “A written sign . . . is not exhausted in the present of its inscription.”²³ It is of the essence of writing to be able, potentially, to transcend its particular context or site of production, to communicate in the absence of author, author’s intention, and implied referential situation.

What is particularly problematic here is the presumption by expression theories of meaning to conceive of expression as a duplication or reproduction of a prior stratum—“to repeat or duplicate a sense content which does not wait for speech in order to be what it is,” as Derrida says.²⁴ Expression is thus restricted to the model of communication, a communication of what is already prefigured in the interiority of consciousness and which is simply mirrored forth in the linguistic utterance. Truth then becomes a matter, on one side, of the adequacy of the expression to the intended sense and on the other of its adequacy to the object referred to. The dissociation that may occur with writing already points to a serious undermining of both these aspects of truth, and also of the conception of language as communication, that is, the passage of meaning from one soul to another.²⁵

Already in Husserl’s work, however, the notion of a neutral mirroring is problematic, for the expressed gives something of a new form to the preexpressed:

A peculiar intentional instrument lies before us which essentially possesses the outstanding characteristic of reflecting back as from a mirror every other intentionality according to its form and content, of copying it whilst colouring it in its own way, and thereby of working into it its own form of “conceptuality.” . . . Expression is not something like a coat of varnish or like a piece of clothing covering it over; it is a mental formation exercising new intentive functions on the intentive substratum. . . .²⁶

More and more the initial instrumentality of language gives way to a creative function that generates rather than mirrors a pregiven meaning. But expression not only generates meaning, perhaps more importantly it also generates the subject and object (qua intended) presupposed by it. Was there in fact a subject preceding expression? This is a position that previous chapters have continually argued against. “There is,” states Derrida, “no constituting subjectivity. The very concept of constitution itself must be deconstructed.”²⁷ Let us now follow this constituting subject, necessarily presupposed by Husserl, back into soliloquy.

Expressions may rely on intuitive fulfilment for their truth value, but this is not what makes them meaningful. Meaning, as Saussure has pointed out, is a matter of the juxtaposition of traditional signs in a more or less grammatical chain (syntagm). Hence, even in interior “monologue” what one expresses may be quite meaningful. But the question to be faced concerns the communicative value of the expressions. Consider Husserl:

One of course speaks, in a certain sense, even in soliloquy, and it is certainly possible to think of oneself as speaking, and even as speaking to oneself, as, e.g., when someone says to himself: ‘You have gone wrong, you can’t go on like that.’ But in the genuine sense of communication, there is no speech in such cases, nor does one tell oneself anything. . . . In a monologue words can perform no function of indicating the existence of mental acts, since such indication would be quite purposeless. For the acts in question are themselves experienced by us at that moment.²⁸

If, as Husserl claims, such communication is considered to be the expression of an already known intention, then communication to oneself is a gratuitous act. What follows from the act, what it indicates, is already present to oneself.²⁹ In order to follow this argument further let us turn briefly to the indexical function of expressions.

Expressions can, indeed must, be able to function also as indices. They may, for example, indicate attitudes, moods, or states of mind; bodily and facial “expressions” also do this.³⁰ Expression may also indicate the sheer otherness or the congeniality of the other. They may indicate the “intention” to describe a certain object, or a certain memory, feeling, or perception. They may further indicate a meaning that the speaker seems not aware of, and so on. These indicative functions are very often associations that experience teaches us, and they may also be subsidiary to the express meaning of the utterance.

The question to be faced is whether, qua listener, the speaker stands in a privileged relation to his own “thoughts,” his own intentions, as Husserl would have us believe. Or is the speaker in the same position any other listener to such an utterance might be in? Does speech in general, like writing, also have a deferred or nonpresent origin? Is the subject produced in and through the signs, and not vice versa? It is significant that language is, to use Husserl’s expression, interwoven (verflochten) with many other act strata.³¹

When people seek to express themselves do they really check, by something like a backward glance, with the preexpressed sense they seek to communicate? Does what is said never surprise the speaker? Is it not the case that the meaning, even the feel or tone, of the expression itself guides one’s next utterance, by the way the utterance relates to the speech context for example? It appears to us that what one “intends” is not at all clearly given unless one has, or has had, the expression for it. It is usually only after the demonstrated inadequacy of an initial expression that one says “what I meant to say is . . . ,” thus feeding in the intention after the fact.

Seeking an expression is very often a matter of attaining a degree of univocity, such that the expression cannot be taken wrongly by a listener. This seeking functions by exclusion and by knowing what one has said rather than what one supposedly “intended” to say. The adequacy of one’s expression is a question of how one understands that expression oneself. If it is ambiguous, one rephrases it, and so on. It is interesting that when someone says “I have it on the tip of my tongue,” they do not yet know what it is, do not yet have it, and they may only come to know what they “mean” in the moment of expression. It is as though the coming of language sets the sense free, brings it to the light of consciousness. Yet another example would be: “I know what I want to say, just give me time.” How often it is that “what I wanted to say” gets worked out in the actual expressing.

These phenomena of language do not, I maintain, occur only in public speech but also in soliloquy. There is, to disagree with Husserl, no essential difference between the two. It is interesting to note, however, that in public one is often much more careful in one’s formulations for they are questioned and corrected not only by oneself but, more importantly perhaps, by other people; that one in fact “does not know” becomes apparent more readily than in soliloquy. It can take discipline even to think in complete sentences to oneself.³²

Rather than posit something like a private language or a privileged and mysterious self-knowledge, it is more economical (in Occam’s sense) to suggest that in solitary monologue one’s expressions first render the meanings of one’s experiences or states present to oneself. This situation is not significantly different from expressing oneself to others. One essentially becomes an interlocutor to oneself.

Much of what I am saying here can be seen to derive from what was said in the preceding chapter, for it is especially clear in Taylor’s description of the relation between emotion and expression. For Taylor, as we saw, man is defined as an interpreting animal, a language animal. It is this thorough embeddedness in language that is often overlooked by Husserl. We could also say it is because language and signification have already occurred that one has a “sense” of knowing beforehand. This phenomenon relates back to what I have called the prenarrative (quasi-narrative) level of our experience.

We can now take a further look at the language of the soliloquizing subject. Meaning is not a free-floating X but is indissolubly linked to a basis in materiality—to both the iterable mark on the paper and to the phonetic material of voice (e.g., one identifies the same word in different modulations of voice). It is this materiality of the sign that is iterable, that has a certain self-identity in its occurrences despite variations in voice and accent.³³ Ideality, to follow Derrida’s lead, is a matter of the seemingly infinite repeatability of signs, their freedom from any particular utterance and any particular speaker. Ideality, therefore, does not properly apply to meaning, for meaning is thoroughly contextual and syntagmatically dependent.³⁴ In other words, meaning (where it is not what we commonly call “self-evident”) is a matter of interpretation, where interpretation that seeks understanding is essentially an act of translating the given expression into what one sees to be an equivalent expression; it is putting the expression into one’s own words, one’s own language.³⁵

Here we find another major indexical relationship: that existing between the materiality of the sign and the phenomenon of expressed meaning. One must see the vocal gestures or the written script as potentially meaningful, as indicating an “intention” toward meaning. It is especially the self-effacing nature of the phonetic medium that leads to our belief in something like an unmediated presence of meaning, to the belief in extra-linguistic sense (Sinn). It is as though the materiality of words immediately passes away once spoken, leaving behind the pure stratum of meaning, of which it was simply the carrier. But this is far from true. As Merleau-Ponty discovered,

The wonderful thing about language is that it promotes its own oblivion: my eyes follow the lines on the paper, and from the moment I am caught up in their meaning, I lose sight of them. . . . Expression fades out before what is expressed, and this is why its mediating role may pass unnoticed. . .³⁶

“My words,” writes Alfonso Lingis, “are ‘living,’ animate with my own life; they do not quit me, do not exteriorize themselves from my own breath.”³⁷ Those iterable words that are always at my disposal are my lifeblood, for it is here that a certain self-consciousness arises and is constantly renewed in the form of hearing oneself speak. It is perhaps this relationship, which Derrida classes under acts of “auto-affection,” that best founds our sense of subjectivity or self-consciousness. “This auto-affection,” writes Derrida, “is no doubt the possibility for what is called subjectivity or the for-itself. .. without it, no world as such would appear.”³⁸ Thus, if there is a presence of the subject to itself, it is the presence of the voice; it is here that I find myself expressed, where I hear myself expressed. The “I” appears in this auto-affective relation.³⁹

The Alter Ego

If auto-affection is the possibility of subjectivity, this subjectivity finds its release, its expression of itself, in acts of signification. The feeling of subjectivity that we have more or less continually, I contend, is quite simply the possibility of signification, of expression, what might be called vouloir dire or a wanting and being able, in most cases, to say or express. But this subjectivity does not know itself outside the fulfillment of its desire to express.⁴⁰ It is at this level of desire that the use of the word intention becomes serviceable, but not as denoting some form of private language or transcendent subjectivity.

It is in the actual expression that I take my place as a subject among subjects; a place that is prepared by language itself. This preparation we have already seen in the function of personal pronouns, in the dialogical unity of “I” and “you.” Just as the “I” gives voice to the silence of subjectivity, so “here” and “now” give voice and definition to my spatiotemporal being.⁴¹ But what, we might ask, is the difference between the child who simply mimics the word I and the adult who expresses himself thereby? Before we can consider this question, we must distinguish between the casual user of the word and the one who, like Descartes, is philosophically fascinated by the “I,” by the cogito.

The casual user is designated by the “I” but is not held by it; he or she is, rather, caught in the drift of the conversation and the topic at hand. The casual user is not usually concerned with what “I” actually means or what it indicates. As was said before, when one is asked what the “I” stands for, the common answer is some fairly vacuous variant on “me!” It usually takes either a philosophical mind or a significant event of some kind to prompt a further questioning of the “I.” There are two types of questioning that can occur here. One is the explicitly philosophical kind that asks after “I-ness” in general; the other answers “who am I?” in terms of an autobiographical or narrative account.⁴² In earlier chapters we were focusing primarily on the second narrational type of questioning; here we are pursuing the philosophical question ourselves.

When the child mimics the saying of “I” he may be on the way to authentic expression (wherein the speaking subject identifies with the subject of speech), but insofar as his vocalization only mimics the phonetic material this gesture can be considered no differently than other sounds he makes; it is neither an expression nor normal speech. The first vocal gestures of a young child are on a par with the spontaneous gestures of his limbs. Such gestures are indices and not true expressions (as these were defined in relation to Husserl), and are to be interpreted in a scheme of associations. It is only from the second year and later that the child is capable of clearly distinguishing his own person and his own perspective from that of others, while allowing others their own perspective. As Merleau-Ponty has written,

The pronoun I has its full meaning only when the child uses it not as an individual sign to designate his own person—a sign that would be assigned once and for all to himself and to nobody else—but when he understands that each person he sees can in turn say I that each person is an I for himself and a you for others.⁴³

It is perhaps true, however, that if the child had no prelinguistic sense of self (no matter how vague), he could not develop into that which language offers him. It will therefore be of use here to outline what Jacques Lacan calls the “mirror stage,” for it presents us with a prototypical situation of l-identification at an age preceding language acquisition.⁴⁴

Lacan distinguishes between what he calls the symbolic, the imaginary, and the real in human experience. The first two realms are defined primarily in opposition to the real or that which is always outside representation and signification. The real is analogous to the drives and desires that, in Freudian theory, are “tamed” by civilization; it also approximates a mute nature in the nature-culture split.⁴⁵ Both the imaginary (images, perceptions) and the symbolic may serve in the formation of a “subject,” but the symbolic is the primary order.

Language is, for Lacan, the privileged symbolic medium, but symbolization extends to rituals, ceremonies, conventions, and such like. By “symbol” Lacan means the realm of signification generative of meaning through a system of differential relations (much as Saussure defined language). Entry into the symbolic begins with the acquisition of language, and from that point on, says Lacan, the real is gradually left behind. Reality is, as it were, redefined and alienated in the new social and cultural order of the symbolic:

Symbols . . . envelop the life of man in a network so total that they join together, before he comes into the world, those who are going to engender him “by flesh and blood”; so total that they bring to his birth . . . the shape of his destiny; so total that they give the words that will make him faithful or renegade, the law of the acts that will follow him right to the very place where he is not yet and even beyond his death. . . .⁴⁶

The symbolic is thus where the child, as a member of a family unit of such and such a type, is constituted even before it is born. Becoming the subject of this symbolic prefiguration (i.e., becoming what it signifies) is one of the child’s passions as well as one of its torments, and is well documented, for example, in many novels that deal with the problem of finding one’s own identity against the impositions of others.

The symbolic does not directly represent or correspond to the real, for it generates a level of signification, and therefore meaning, in a more or less closed network of mutual relations that both refigure and transcend the level of the real. This mutual relation of signifiers is such that the signified is always another element in the signifying network, that is, another signifier or group of signifiers. (It is to this symbolic realm that Benveniste’s account of the subject truly belongs.) Lacan accordingly sees in the expressions of subjectivity a split that is symptomatic of this real-symbolic division, a split between the embodied speaking subject (real) and the subject as signified in, say, language (symbolic). This latter division is essentially what we have already argued for in terms of the division between the speaking subject and the subject of speech. The embodied subject, in effect, is externalized in language (and in other signifying systems) and identifies with the externalization, the projection.⁴⁷ This stage of identification is productive of what we will call the spoken subject, the final stage of the linguistic auto-affective relation.

This alienation of the symbolic from the real is best exemplified in the earlier “mirror stage” of imaginary representation. The young child is thought to have no conception of itself (in the linguistic sense of concept), and its bodily image of itself (if it can be said to have one) is at first highly fragmented into the various auto-affective relations pertaining to its own bodily functions, particularly the touching-touched relation. Unlike its perception of others, who may have a certain visual totality, the child is presumed to have no vantage point from which to view itself in a like manner. Some form of visual reflection corrects this deficit.

Lacan contends that the mirror stage occurs first at about the age of six months.⁴⁸ It involves the child’s identifying with its specular body image:

This jubilant assumption of his specular image by the child at the infans stage, still sunk in his motor incapacity and nursling dependence, would seem to exhibit in an exemplary situation the symbolic matrix in which the I is precipitated in primordial form, before it is objectified in the dialectic of identification with the other, and before language restores to it, in the universal, its function as subject.

What this experience, which is essentially that of objectification, yields is a visual or imaginary (in Lacan’s sense; not to be equated with fanciful) alter ego set over against oneself, one’s bodily feeling being, that one then identifies with. Correlative to this development in the imaginary there is also a significant reorganization of the child’s spatial field in relation to this newfound “sense” of self. Says Lacan, “the mirror-image would seem to be the threshold of the visible world.” One important effect of this imago is a break in the syncretic identification with others around him, a separation of his life and actions from theirs.

Important here is that the mirror image is not experienced as separate from the child’s identity, and yet it is a displacement from the immediacy of the tactile body (the real). Lacan locates this meconnaissance at the origin of self-consciousness, and its general structure of displacement (the “fragmented body”) is carried through into the symbolic stage. From the narcissism of the mirror stage, for example, the infant gradually transfers its ego ideal onto other people, especially the mother. The symbolic does not, however, totally replace the imaginary, for the identification, say, with iconic role models and ideals continues throughout life.⁴⁹ Consider in this regard the seductive effects of advertising and films, and also the quasipresence that one experiences in front of photographs and the like. In many societies the imago is potentially dangerous (as is one’s name) if it gets into the wrong hands. The mirror stage is highly indicative of the role that representation of oneself (doubling-mimesis) will play throughout one’s cultural life, and of which narrative recounting is a primary form.

We can see in Lacan’s account of the mirror stage a few traits that we have noted before. The I or ego is not the product of a gradually evolving, self-generating consciousness, as though self-consciousness simply enlarges itself from itself in the course of the child’s experiencing. There is in the mirror stage a fundamental dialectic with an other that precludes such an autonomy of consciousness. As Lacan points out, we should not “regard the ego as centered on the perception-consciousness system, or as organized by the ‘reality principle’—a principle that is the expression of a scientific prejudice most hostile to the dialectic of knowledge.” Rather should we start “from the function of meconnaissance that characterizes the ego in all its structures. . . .”⁵⁰

Merleau-Ponty proposed a similar view when he considered the child’s perception of others: “the perception of others cannot be accounted for if one begins by supposing an ego and another that are absolutely conscious of themselves, each of which lays claim, as a result, to an absolute originality in relation to the other that confronts it.”⁵¹ Merleau-Ponty’s role of the body image and the image speculaire in the initial relation of a child to the other, and to itself, perfectly parallels the function of Lacan’s mirror stage.

The specular I and the general structure of consciousness that it maps out represent a loss of presence, especially if we equate presence with a unity of being that overlaps with itself, that is transparent to itself and knows itself as origin. If Lacan’s category of the real stands for something like an originary self-enclosed being, this is not a form of being that has self-presence. Only by the detour of the other is self-presence attained. Earlier I noted that Ricoeur, in his account of hermeneutics, also insists on some such detour if self-knowledge is to be had. As we have seen, this self-presence is precisely a presence grounded in an identity given through difference. This conclusion relates back especially to the illusions of intentions and of the supposed thinker behind the thinking.

Let us now draw some of our conclusions together. Our investigations up to this point reveal a subject that can be conceived, and can “conceive” of itself, in a number of ways. Each of these ways is related to certain concrete experiences; they are not intellectual abstractions. There is, first of all, the “I am I” experience of syncretic unity (in the child this is primarily characterized by “introspective” feelings and impulses). Second, there is the auto-affective stage prior to the mirror stage, where no stable or holistic self-identification occurs. Third, there is the mirror stage, which sets up, on the plane of the imaginary, a representational self-image that the child identifies with. Fourth, beyond the mirror stage comes the realm of role models that the image identification is transferred to; one gains selfhood in and through other persons. Finally, through the acquisition of language there is a gradual and complex identification with voice, thought, and abstraction on the plane of the symbolic. This identification is based especially on the speaking-hearing dyad, which, in expressions of meaning, promotes the belief in an interiority of consciousness or, in a word, mind. The assumption of a substantial self behind expression occurs partly by overlooking the essential materiality of the signifier.

The symbolic level also opens positions prefigured for the subject in the form of personal pronouns, and these have been shown (in the first person) to be dialectical. In interior monologue this dialectic still occurs—in the form of talking to oneself. Here the alienation, or duplicity, of the mirror stage continues. One may still ask: “Is the I that speaks the same as the I spoken about?” With this entry into language proper (the symbolic) all other modes of I-identification tend to fall prey to the linguistic and to the type of understanding that the linguistic affords. Only perhaps in dreams and “mental” disorders do we find what seems to be a predominance of the more archaic modes.

It is often considered important, in matters pertaining to self-consciousness and self-understanding, that the relation of the preexpressed and the expressed is such that the latter should mirror the former. But we have seen that this model disregards the essential relation of ourselves to language. The disclosive power of language is formative of the subject,⁵² of a speaking subject that defines itself in its own expressions and identifies with the subject there portrayed. Earlier chapters have sought to delineate the preexpressed or prenarrative realm in terms of both the prior functioning of language and the quasi-narrative structure characteristic of experience itself. We have seen that the prenarrative nature of experience serves as a basis for interpretive narrational activity (as for recollection), but such narration cannot be said to aim simply at mirroring the prenarrative level. We shall now consider, taking psychoanalysis as an example, what is meant by saying that narrative can be the truth of the prenarrative.

Narrative and Truth

An important question that arises from the contemporary emphasis on the self as a narrative construct concerns the adequacy or truthfulness of the narrative accounts we give of ourselves. What, for example, stops our self-narrations and self-characterizations from becoming, in many cases, mere inauthentic flights of fancy or sheer fictions? If, on a fairly radical view, the self is taken to be a product of narrative emplotment, then clearly this question cannot be answered by a simple appeal to what the narrative is supposed to copy or represent—as though the self were first given to us outside or prior to the narrative. It is not immediately clear, however, just what a narrative account of the human subject could be adequate to; and if we were to grant some prenarrative status to the human subject, it is also not clear whether the imposition of narrative structures falsifies the “truth” of this latter subject. What we shall examine here, then, is precisely this important question of adequacy and various problems that surround it.

Guiding our present investigation is this question: to what degree can the truthfulness of a self-narration be considered more a matter of pragmatic and creative adequacy than of a correspondence to the way things actually were or are? We have already seen, in my treatment of memory, that although the past is a constant horizon and support for the present, it is not thereby given with fullness of meaning to reflection and recollection. Recollection, I maintained, is both selective and interpretive. We do indeed remember that certain events have occurred, but understanding their import implies the further task of discerning a chain of events or a story to which they belong.

Self-understanding also requires that we see some form of causality (and rationality) operating in our lives. For human actions this causality takes the form of motivations or purposes. In The Phenomenology of the Social World, Alfred Schutz makes a useful division of human motivation into two aspects: the “because-of” and the “in-order-to” motives. The former motive is oriented to the past, while the latter is futural. Understanding human action necessarily involves the explication of these two aspects of motivation. Without the meaning conferred by such an explication we would have, at best, only a chronicle of events.⁵³

A central problem with motivation is its not being fully conscious to the actor. Motivational contexts usually extend beyond anything we explicitly formulate, and part of the reason for this is simple forgetfulness. How often have we read in a novel something like this: “Although he didn’t know it yet, it was his growing love for her that drove him to such extremes of behavior.” Other problems with describing motives concern the possibility for fabrication and duplicity. For example, one’s explicit reasons for acting can disguise a deeper, less conscious motive. This disguised motive may also be the product of repression, say, rather than just a simple oversight.

Perhaps the very suggestion of fully accounting for motives is in the end doomed to failure or is at least highly problematic, for such a task would seem to require a foundational subject that has the sources of its own acts potentially within scrutiny. The meaning of our acts, however, as this is worked out in terms of because-of and in-order-to motives, is a product of retrospective and prospective emplotments that draw upon the prenarrative past, refiguring it in light of the present demand for sense and coherence. Here again we find the dialectic of the prenarrative and narrative, a dialectic that is, to borrow a useful phrase from Merleau-Ponty, one of creative adequation.

This dialectical situation places us in the proverbial chicken-or-egg dilemma—the hermeneutical resolution of which is to say that we cannot have one without the other. We undoubtedly act based on our prenarrative context, but the question of motivations immediately involves us, as self-conscious human subjects, in our awareness and expression of such motivations. As human subjects we not only act but do so within a more or less detailed plan or emplotment of the action. The question of “truth” thus involves us in the question of the adequacy with which our explicit narrations map onto or otherwise follow from prenarrative experience.

As stated, however, this approach involves us in the problematic epistemological stance of a correspondence theory of truth. The correspondence theory is, however, only tenable if our prenarrative experience has meaning for us outside our interpretations or emplotment of it, but, as I have argued in this and earlier chapters, such is not the case. The truth of our narratives does not reside in their correspondence to the prior meaning of prenarrative experience; rather, the narrative is the meaning of prenarrative experience. The adequacy of the narrative cannot, therefore, be measured against the meaning of prenarrative experience but, properly speaking, only against alternate interpretations of that experience.⁵⁴

Our task here, then, is to pursue what is meant by a creative adequation between prenarrative and narrative experience, with the emphasis on the word creative. To illuminate this relationship I shall begin by taking a further look at psychoanalysis and conclude with an examination of historical writing.

If the telos of the symbolic, as we saw earlier from Lacan, is to generate a subject whose domain is in the order of signification, this transformation must occur at the expense of what he calls the real: drives, instincts, primary desires (keyed especially to the sites of bodily processes, but also to external objects of intimacy and gratification), and the body’s rhythms generally. This symbolic subject is always on the way to becoming the Cartesian subject or Husserlian transcendental ego that, from the secure platform of the cogito, is assured of its own unity, homogeneity, and epistemic centrality. What holds the symbolic subject back, what checks its flight, are the transgressions wrought by the more “primitive” level. As Julia Kristeva says, “anguish, frustration, identification or projection all break down the unity of the transcendental ego and its system of homogeneous sense and give free rein to what is heterogeneous in sense, that is, to the drive [Trieb].” In such transgressions, especially if they are extreme, “the speaking subject undergoes a transition to a void, to zero: loss of identity, afflux of drive and a return of symbolic capacities, but this time in order to take control of drive itself.”⁵⁵

This shift, Kristeva contends, is what inaugurates new signifying processes, particularly of a creative or poetic kind. Poetic language with its reliance on metaphor and metonymy is, in its revitalizing of our symbolic capacities, essentially a revolutionary practice, overturning the categories with which we commonly describe ourselves.⁵⁶ What poetic discourse (especially lyric) establishes, to borrow from Kristeva, is a subject-in-process, a subject still finding or refiguring itself. This unsettled subject manifests itself as a speaking subject that diverges from normal referential and communicative discourse. In this respect, poetic discourse often operates, much like Freud’s discourse of the unconscious, through displacements and condensations that may defy both semantic and grammatical categorial interpretations.⁵⁷ What “speaks” in such instances is a state of being anterior to the Cartesian subject. What is said may be epistemically unprecedented.⁵⁸

Whereas poetic discourse gives willing voice to otherwise unformed desires and emotions, that which is repressed seeks a voice for what the conscious subject has, for one reason or another, avoided or put aside. In operation these two processes can be very similar.⁵⁹ Lacan states the psychoanalytic model as follows:

Undoubtedly, something that is not expressed does not exist. But the repressed is always there—it insists, and it demands to come into being. The fundamental relation of man with this symbolic order is precisely the same one which founds this symbolic order itself—the relation of being to non-being.

That which insists on being satisfied can only be satisfied through recognition. The end of the symbolic process is that non-being comes to be, that he is because he has spoken.⁶⁰

For Lacan the unconscious develops from the split that the symbolic introduces into our being. The unconscious thus evolves dialectically with the expressed; it is the other side, as it were, of the expressed. In this way, as Lacan says, “the unconscious is structured like a language”—a phrase that might usefully be said of prenarrative experience generally.⁶¹

The origin of this view is found in Freud’s notion of unconscious “dream-thoughts” that undergo transformation and censorship in the dream-work and which interpretation seeks to uncover.⁶² Psychoanalysis is thus a process of disclosing the “discourse” of the unconscious that motivates and subverts our explicit discourse, particularly our self-narrations, and this means (at least for Lacan) being sensitive to the metaphoric and metonymic transformations that occur as this other discourse enters conscious expression. In psychoanalytic practice the analysand should come to recognize and appropriate this other discourse; this may be seen as the central moment of “curing” the analysand. As Lacan succinctly puts it, “Analysis can have for its goal only the advent of a true speech and the realization by the subject of his history in his relation to a future.”⁶³

It should be clear how this model of the subject fits in with our previous discussions of the prenarrative level.⁶⁴ The psychoanalytic prenarrative is a part of one’s own history, one’s experience, that is refused anything but an oblique entry into one’s ongoing and conscious life story. Though it has its roots in a perhaps instinctual bodily basis, this prenarrative nevertheless has conscious recognition as its goal and is already a structuring force in one’s self-conception. What psychoanalysis is premised upon and constantly stresses is the resistance of the subject to its own truth: “One is never happy making way for a new truth,” says Lacan, “for it always means making our way into it; the truth is always disturbing.”⁶⁵

Earlier we talked of this truth of the subject in terms of an act of creative adequation, and this tends to involve, at least in the psychoanalytic case, overcoming prior and perhaps well-established interpretations of ourselves. This is also a reason why literature, at its best, is both disturbing and liberating. We shall now pursue this notion of truth a little further, keeping psychoanalysis as our guide.

Psychoanalysis is known as the “talking cure” precisely because its analyses are carried out primarily in the realm of discourse and dialogue—the discourse of the analysand with himself and with the analyst. It has become increasingly evident that a primary aim of the analyst is the unfolding of a life history, a history that does justice both to the past and to the present. Analysis can take diverse routes, but the end result is a narrative account of the analysand’s life wherein the analysand finds him or herself adequately reflected and can accept this representation as biographical and, perhaps more fruitfully, as a basis for future action. As Roy Schafer writes, “It has been becoming increasingly clear in recent years that clinical psychoanalysis is an interpretative discipline whose concern it is to construct life histories of human beings.”⁶⁶ Why this should be so is that the human subject, as we have stressed, exists and knows itself as the implied subject of its own discourse and narratives (though often these are told by other people). “We are,” says Schafer, “forever telling stories about ourselves. In saying that we also tell them to ourselves, however, we are enclosing one story within another. This is the story that there is a self to tell something to, a someone else serving as audience who is oneself or one’s self.”⁶⁷

We have already discussed this self in terms of the implied subject (especially signified by the personal pronouns) and the speaking-listening dyad. What Schafer is pointing to is that the reality of the subject for itself is primarily linguistic, derived from its self-narrations. Schafer rightly applies this thesis also to other persons: “The other person, like the self, is not something one has or encounters as such but is an existence one tells.”⁶⁸ Even where this story of other persons is not explicitly told there is, nevertheless, the implicit assumption that it could be told.

Let us consider an example from neuropsychology. In The Man Who Mistook His Wife for a Hat, Oliver Sacks discusses various cases of memory disorders caused by Korsakoff‘s syndrome. Of one patient with a memory span of only a few seconds, Sacks notes:

Unable to maintain a genuine narrative or continuity, unable to maintain a genuine inner world, he is driven to the proliferation of pseudo-narratives, in a pseudo-continuity, pseudo-worlds peopled by pseudo-people, phantoms.

Such a patient, says Sacks, “must literally make himself (and his world) up every moment.” The problem here is that memory loss makes it impossible to link different narrative instances around a common theme or development, and yet some sort of narrative is necessary for a sense of identity and purpose. The identity of the patient, for himself, can only be maintained through continuous narrative activity, through staging dramas in each succeeding moment: “The world keeps disappearing, losing meaning, vanishing—and he must seek meaning, make meaning, in a desperate way, continually inventing, throwing bridges of meaning over abysses of meaninglessness. . . .” For the patient these bridges are not mere inventions; they are the world as he knows it, as he interprets it at that moment, and the same will apply to his sense of selfhood.⁶⁹

What this example illustrates is both the ongoing need for narrating experience in order to exist as a meaningful human subject and the function of narrative in generating a continuity of identity, of self. Indeed, a good case can be made for viewing narrative understanding as the most adequate approach to the human domain. To understand ourselves we must grasp our own implicit history, for to be human is not simply to have a history (in a certain sense animals have this), but to be cognizant of this history. This is why the “linguistic turn” of hermeneutics, unlike much of Anglo-American language philosophy, goes hand in hand with narration and therefore with historical analysis and description.

The continuity afforded by narrative, however, can be a feigned one—especially from the perspective of other people. Sacks’s patient had no knowledge of his dispersed self other than through the reports given by other people. The narratives he invented were simply fragments or residues of his past life and past occupations (his fragmented habitus), which were then arbitrarily imposed on the present. These stories (and dramas) were created (and acted out) by the patient not only to give himself a role to play but also to integrate his surroundings into something familiar, something that makes sense. Socially, however, the identity he created was a failure, for there was an imposition of roles on other people that were simply not appropriate and an adoption of roles by the patient that, for an observer, were absurd or misplaced.

Because of the physical damage to his brain, Sacks’s patient could not be cured of his delusory fabrications; anything told to him was forgotten after a brief interval and the fabrications began anew. Such person lives at the level of shifting surface phenomena. He lacks the depth, the richness, and the constraints that the past usually imposes on one. Psychoanalysis, on the other hand, is aimed at people who have the possibility of augmenting their life stories in light of present practices and expectations as well as their past. But there are numerous interpretive barriers to be encountered—and not only, for example, in the recounting of dreams, for even in the narrating of what took place yesterday there are problems of interpretive distortion to be overcome.

As we saw earlier, even memory retrieval involves a degree of emplotment if what is remembered is to be interpreted and understood in the context of an ongoing life.⁷⁰ The psychoanalytic analysand has the additional problem of the involuntary masking of certain events and interpretations. Certain avenues of interpretation of the past may be closed or inhibited because of traumatic experience in the past, for example.⁷¹ The analyst is especially seeking to facilitate the movement of such repressed contents into explicit speech and recognition. But here again there is an interpretive problem. Analysts must interpret what is told to them out of their own interpretive schemes, with consideration of a direction the analysis is to take and on the basis of their own life experiences. The problems multiply when we consider the stories involved in the psychoanalytic dialogue.

There is, first of all, the presumed story “waiting to be told” which contains the key to anomalies or inconsistencies in one’s present behavior; this is the repressed story, and it belongs at the prenarrative level. Second, there is the story told by the analysand. This story may be a pure fabrication, but even if it is a report of what the analysand distinctly remembers there will still be, in the telling, impositions of both style and context onto what is told, as well as selectivity. Also, in Habermasian terms, one will likely understand out of an “interest.” Thus the so-called facts are told in a certain manner, style, or genre, and for certain effects, to illuminate certain points, and also out of certain interests that inform the dialogical situation. And no doubt there could be other factors involved that we have not mentioned, such as one’s emotional state during the telling. Third, there may be the further story which the analysand constructs on the basis of his or her initial disclosure. This latter story would be the story of what the story told “actually means” to the analysand upon reflection. Fourth, there is the story as heard by the analyst, the meaning of which is inevitably different from the analysand’s, for it is heard against a different background, out of a different context and interests. Finally, there is the underlying story that the analyst seeks behind what is told—the subtext. In classic psychoanalytical theory, which Freud saw as an archaeology, this latter story should coincide with the first story, the repressed story which is the “truth” of the analysand’s past.⁷² What, then, are we to make of the above proliferation of narratives?

Truth, as the term is commonly used, relates to the adequacy of statements for conveying the way things actually were or are. The archeological model of psychoanalysis would similarly commit one to what might be called “historical truth”: the correspondence of descriptions to the analysand’s past. What has become evident to numerous psychoanalysts, however, is that this historical truth is little more than “narrative truth.” As Donald Spence remarks, it “is more appropriate to think of construction than reconstruction; to give up the archaeological model; to think of an interpretation as a pragmatic statement that has no necessary referent in the past; and to replace historical truth by narrative truth.”⁷³ In other words, the archaeological-historical method commits one to a merely factual view about a person’s past, thereby reducing the contextual significance of the present and future for the very meaning of the past. The archaeological model seeks story number one (above) as though it were fully formed and had only to be brought to consciousness and recognized for what it is. Such a view, as we saw in the chapter on time and memory, disregards the futural nature of human existence and the hermeneutic dimension of interpretation. The past has meaning only in light of what precedes it and what follows it (Schutz’s because-of and in-order-to motives). This requirement is what makes all recollection that seeks understanding a narrative endeavor, a matter of emplotment. Story number one is only a quasi-narrative, a story still to be told, and told from a certain perspective.

Much of the past, including motivations and possible complexes, is only on the way to language, and the carry-over is often a rather difficult, creative, and prolonged task (if psychoanalysis is anything to go by!). As Spence maintains, “The construction not only shapes the past—it becomes the past in many cases because many critical early experiences are preverbal and, therefore, have no proper designation until we put them into words.”⁷⁴

The analysand’s explicit associations and recollections are like so much disorganized material that needs to be understood in light of a narrative that holds it together in a development that yields familiarity, meaning, and, accordingly, understanding. This is a process that in our ordinary lives we commonly achieve satisfactorily. The methodological result to be gleaned from this is that psychoanalysts should not orient themselves toward revealing a final story that supposedly duplicates a repressed past but should enable the analysand to overcome problems in the present by allowing the formation of a therapeutic narrative that nevertheless gives meaning and direction to the analysand’s life.

Narrative truth is thus more a matter of facilitating understanding and integration than of generating strict historical verisimilitude (supposing this were even possible at our level of investigation). To quote Spence again:

Narrative truth can be defined as the criterion we use to decide when a certain experience has been captured to our satisfaction; it depends on continuity and closure and to the extent to which the fit of the pieces takes on an aesthetic finality. . . . Once a given construction has acquired narrative truth, it becomes just as real as any other kind of truth.⁷⁵

Spence’s last remark may appear somewhat extreme, but one must remember the context within which it is written. When dealing with the meaning of a person’s life (or of the past) we can only have interpretations that, given what we otherwise know of that life, afford us a satisfactory comprehension of it (especially of motives). As I argued earlier, interpretations vie for credibility not simply in their accounting for the known details of a life but, more importantly, in relation to other and perhaps more viable interpretations. Spence’s use of the term real may still be puzzling, but what is implied is that a narrative account that fulfills the conditions he mentions (continuity, closure, etc.) is the optimum in our understanding of a person’s life. That is, the historical account must become narrational if a comprehensive understanding of individual lives is the goal.

While one may agree with Spence’s general position, it should be noted that certain questions are nevertheless raised by his account. First, there is the problem of the relativism of narrative interpretations. No one account can be regarded as the final truth. This is, however, a limited relativism, for there will exist extranarrative elements (e.g., knowing that, in relation to dates, and places) for which one seeks a narrative emplotment. Presumably the narrative must respect both the temporal sequence of these elements and their content, and it must also link them in a way that is intellectually and emotionally acceptable. The narrative should, in other words, provide a feasible context for the exhibiting of the elements in their causal and temporal connections and should reveal how the past is operating on the present and on the expected future (though perhaps with apprehension). As Spence indicates, the final judgement of a narrative is its acceptance by the one whose experience it recounts and whose reflected life it becomes, even though this acceptance may not be easily won. It should also be noted that one’s own acceptance of a narrative may be significantly affected by whether or not other people accept one’s account.

Another possible problem with narrative relates to its aesthetic and rhetorical appeal. Once certain experiences are brought into some form of narrative sequence and closure there may be a tendency to accept uncritically the finished product. This situation is due primarily to the persuasive rhetorical character of narratives. A related problem is that any series of events may be amenable to numerous and different tellings, to different ways of filling in the gaps left by recollection. This question of open-ended interpretation is, however, common not only to self-understanding but to all fields of the humanities and cannot be avoided.

The kind of truth proposed here is, as Spence mentions, a pragmatic one. Psychologically, narrative is aimed not at achieving a mirror image of one’s history but at generating a plausible account of the details of that history and allowing one to have an understanding of oneself that facilitates the overcoming of psychic blockages and allows one to function satisfactorily in the present. In fact, a fictional narrative could serve the same end if it addressed the right questions, situations, and conflicts. As C. G. Jung stated at the beginning of his autobiography, “Whether or not the stories are ‘true’ is not the problem. The only question is whether what I tell is my fable, my truth.”⁷⁶ We are after all story-telling animals, speaking as much in allusions, symbols, and metaphors as in the logic of “objective facts.” Narrative truth is thus a matter of adequacy and fit to what is otherwise given, and this is so not only for psychoanalysis but also for historiography and any form of recollection that seeks to unfold a past history that aims at more than a mere chronology.

We turn now to a broader discussion of historical narrative in general, especially with a view to addressing the ideological distortions that always threaten self-narration.

Like self-narration, the writing of history is a way of consolidating a past, a tradition, and therefore an identity. Like self-narration, history is also concerned with a form of archaeology that operates on a prenarrative subtext, though in this case the prenarrative material consists primarily of material artifacts in addition to memorial traces. However, both self-narration and history writing have a common grounding in the sense of our lives as being temporally circumscribed.

In much the same way that contemporary philosophy has brought into question the status of an underlying self or soul substance and has, as I have done here, sought to approach an understanding of the human subject via the paradigm of language, so many historiographers and philosophers of history have problematized historical research by investigating the nature and presuppositions of the historian’s use of both language and narrative. The dream of a history which, in Ranke’s words, reports the past wie es eigentlich gewesen becomes questioned.⁷⁷ History, considered hermeneutically, is an interpretive discipline that should not hope to coincide with past events but must make do, as the French historiographer Paul Veyne contends, with narrative reconstructions in language: “Knowledge of the past is not an immediate datum, for history is an area in which there can be no intuition but only reconstruction.”⁷⁸

Given that history writing is a narrative endeavor,⁷⁹ akin to the novel, we are led to ask after the status of its “referent.” Hayden White states the problem in this way:

the problem of narrativity turns on the issue of whether historical events can be truthfully represented as manifesting the structures and processes of those met with more commonly in certain kinds of “imaginative” discourses, that is, such as fictions. . . .⁸⁰

If we consider this question from the point of view of a form-content distinction, it is an easy matter to say that formally historical discourse borrows from literature, but that nevertheless its content is drawn from “reality” rather than from the imagination. That is, the content of history is found rather than invented; its final referent is not simply a product immanent to one’s narrative, but is external to the recounted story.

The answer I have already formulated to this question of reference is, on the contrary, that narrativity is a principle of intelligibility and not simply a vehicle for a pregiven and evident sense. As I have argued, the given (human actions, transactions, and so on) has a quasi-narrative status that has yet to be brought to explicit narrative understanding. This latter process, however, will not tolerate a description in terms of a simplistic dichotomy of form and content. Narrative expression is not mere communication of information but is a constitutive and synthetic activity. Historical narration takes its lead from artifacts but must aim beyond them to a synthesis that yields a satisfactory coherence, directionality, and intelligibility; otherwise history would be a mere cataloguing or dating and could not hope to rise beyond the chronicle stage. Peter Gay says as much when he writes: “Historical narration without analysis is trivial, historical analysis without narration is incomplete.”⁸¹

What historical narrative generates is not a neutral mirror of the past but a seeing of the past as something: as a gradual emancipation from certain class structures, as an unfolding tragedy, as dominated by certain religious beliefs, etc. The artifacts may be seen to justify such interpretations or stories, but they can usually be seen to justify numerous other stories as well. As with recollection, present interest and the conceptual tools of the present set parameters to what material will be deemed relevant, what story will be told and the style (or genre) of that story. Hayden White makes a similar point in his collection of essays Tropics of Discourse:

Histories are not only about events but also about the possible sets of relationships that those events can be demonstrated to figure. These sets of relationships are not, however, immanent in the events themselves; they exist only in the mind of the historian reflecting on them. Here they are present as the modes of relationships conceptualized in the myth, fable, and folklore, scientific knowledge, religion, and literary art, of the historian’s own culture. But more importantly, they are . . . immanent in the very language which the historian must use to describe events prior to a scientific analysis of them or a fictional emplotment of them.” ⁸²

The unavoidable and hermeneutical “seeing as” of historical narrative is what gives to the fragments of the past a significance beyond their mere occurrence.

But of course the historical past generally does not consist of mere disparate events and archival material. The past is, for most societies at least, already historicized, already told and continually updated. As with our self-narrations, one’s present account is largely a reworking of stories already related, already participated in—only the amnesiac begins, as it were, ab nuovo. As we noted earlier, experience is irredeemable qua experience and, in addition, it is the significance or meaning of the experience and the world that one seeks to present and understand via narration.⁸³

This intertextual and literary nature of historical narration has not been overlooked in contemporary scholarship. Roland Barthes, for example, goes so far as to bring historical narrative and fictional narrative into the same camp, with a view to contesting the former’s claim to scientificity and objectivity:

Does the narration of past events, which, in our culture from the time of the Greeks onwards, has generally been subject to the sanction of historical “science,” bound to the underlying standard of the “real,” and justified by the principle of “rational” exposition—does this form of narration really differ, in some indubitably distinctive feature, from imaginary narration, as we find it in the epic, the novel, and the drama?⁸⁴

The object of Barthes’s criticism is twofold: a rejection of extralinguistic referentiality and a critique of the ideological use of historical discourse. We can go along with both criticisms, but only to a certain degree.

Barthes sums up his rejection of reference in this way:

Claims concerning the “realism” of narrative are therefore to be discounted. . . . The function of narrative is not to “represent,” it is to constitute a spectacle. . . . Narrative does not show, does not imitate. . . . “What takes place” in a narrative is from the referential (reality) point of view literally nothing; “what happens” is language alone, the adventure of language. . ⁸⁵

This claim is aimed at narrative in general; although it may appear to apply primarily to literary fictions, it is nevertheless, for Barthes, also applicable to historical discourse.

Barthes’s principal point here is that the meaning of a narrative is a product of its language and thus cannot be said to mirror the nature of “real,” extralinguistic, past events. In historical discourse we construct a spectacle that via the authority of its author (its academic situation, and so on) is deemed “historical” and is thereby granted a referentiality to the “real” world—much as the implied subject of a personal narrative is reified into an existent soul substance, a thing. Like Mink (“Stories are not lived but told”), Barthes views narrative as an integrative way in which the past is given a meaning it otherwise lacks, a meaning that is actually a play of language that is far from ideologically innocent in its “recounting”: “As we can see, simply from looking at its structure . . . historical discourse is in its essence a form of ideological elaboration.”⁸⁶

Such historical narratives have a performative dimension that often serves surreptitiously to fashion and promote a certain image of man. One has only to consider mythological world views to appreciate how the individual’s image of himself and his social relations are delimited and perhaps constrained by the parameters placed on his “reality” by the society’s “histories.” In our own time, histories differ depending on whether they are told by the East or the West, the rich or the poor, them or us.

I have already said a good deal about the illocutionary force of narratives in chapter 2. However, what is overlooked in Barthes’s strongly semiological-structural account is that narrative expression is constantly interwoven with what he calls “reality.” There is still in Barthes a dichotomy of the narrated (language) and the real (nonlinguistic reality). If this “real” is consistently examined, however, we find that it cannot be exempted from at least a quasi narrativity (this was my argument against Mink). Historical narratives, like their personal counterparts, need not be free floating but, as I have said, can draw on the narrative structure of human time itself, on the story that is already evidenced in the purposive structure of human events.

Historical discourse, like self-narration, falls into that intermediary realm between fact and fiction; this is what leads Paul Veyne to say that “history is a true novel.”⁸⁷ What distinguishes history from fiction is that the events related in the former are presumed to have actually taken place and were, at the time, witnessed or undergone by certain real persons. An additional factor is that the historian is often concerned to give precise reasons and documentary evidence for why a particular description or explanation of events is offered over others. What unites history with fiction is its dependence on narrative discourse and creative synthesis in order that events have meaning and purpose. As with traditional fiction, history seeks both closure and completeness but can attain them only through selection and by applying the formal beginning-middle-end structure of narrative—which then implies “discovering” such teleologies in the events of the past. As we learned from Maclntyre, it is often the story within which events are framed that first gives them their importance.⁸⁸

This interweaving of history and fiction is taken up in an interesting way by Ricoeur in the third volume of his Time and Narrative, and it will be pertinent to look at some of his conclusions here. Ricoeur’s hermeneutical project has always been concerned with the pervasiveness of the “seeing as” structure in human experience and, concomitant with this, the necessary mediatory role of imagination as it generates adequate representative figures for the past. It is Ricoeur’s thesis that “seeing as” and productive imagination are common to both history and fiction such that history is seen to be “quasi-fictive” and fiction “quasi-historical” (3:190f). These two genres of narrative are shown to be mutually dependent and together serve the important role, at the reception stage, of refiguring our experience in its temporal dimensions.⁸⁹

History relies on imagination for its attempt at a reenactment (Colling-wood’s term) of the past, a reenactment that aims at making the otherness of the past less foreign: “It is always through some transfer from Same to Other, in empathy and imagination, that the Other that is foreign to me is brought closer” (3:184). The other must be brought into some significant relation to my present, must be reconstructed (configured) within a life context (social, cultural, etc.) that can be understood by extension from my own). In the last resort, the historian presents the past such that one gains some sense of actually being there, albeit in the mode of the “as if” (3:185). To aid in obtaining this type of fulfillment, historical imagination also borrows from cultural traditions various genres and rhetorical devices of literary emplotment: tragedy, comedy, romance, irony, analogy, and the like. The employment of such devices, sometimes unconsciously, often follows from the motivation and interests that serve to isolate the historian’s field of investigation; it is, for example, understandable that an account of the Holocaust will take on the rhetoric of tragedy.

The above factors explain why many great works of historical writing could be read as though they were novels. On the other hand, however, it is easy to see why a novel such as Tolstoy’s War and Peace could be read as history. In fact, the traditional fictive novel, maintains Ricoeur, usually depicts events and states of affairs in a way that makes them seem plausible or possible given our prior knowledge of the world. This is, of course, a point stressed long ago by Aristotle in his Poetics. Ricoeur, however, enriches this claim by adding that what is encountered in the fictional work can be read as a “quasi-past,” in this case the past of the narrative voice: “Fictional narrative is quasi-historical to the extent that the unreal events that it relates are past facts for the narrative voice that addresses itself to the reader” (3:190). Thus, through a certain willing suspension of disbelief, we read the fictional work “as if” it were a recounting of an actual past. This, says Ricoeur, is part of the pact we enter into with the author when reading such works. A major value of the fictional, to return to Aristotle, is to disclose possibilities in the real past which, to follow Ricoeur, aid in the imaginative refiguring of the reader’s own world. In this regard, one has only to think of the way stories and myths enter into the sense a community has of its own historical identity.

Ricoeur’s work clearly offers a firm basis for further considerations of the interpenetration of history and fiction and of their mutual effect in forming our own sense of the temporality and historicality of experience. This interweaving has, especially in the latter half of the twentieth century, been put through interesting and provocative variations in the works of various postmodern or metafictional novelists, often with a more theoretical aim of problematizing the entire history-fiction dichotomy. Salman Rushdie’s Midnight’s Children, for example, explicitly weaves together historical fact and fiction into a narrative that appears self-consciously aware of its fictive transgressions while being sure of its possible effect in refiguring our sense of the past.

One might well claim, in light of our above considerations of history and narrative, that the problem with conceiving historical discourse (and our own biographical self-narrations) along narrative lines is that (1) this does not satisfy our more positivist longing for empirical exactitude and disinterestedness, and (2) it opens the gates to potential abuse: to constructing a past that suits, say, one’s ideological purposes. This latter point indicates a potential disfiguring and not merely refiguring effect of both historical and fictive narratives.

In response to the first of these objections, I shall only say here that empirical exactitude makes for boring history, history without life’s drama.⁹⁰ History without interpretative narrative emplotment, which gives meaning to the events related, would in fact be an impoverished account of human experience and social transactions, especially if we are correct in granting such transactions a quasi-narrative status in the first instance. Similarly, empirical exactitude applies only to dates, places, and documented evidence. The more global meaning of these events and reports is developed in the story, which is told after the fact and which cannot be said to simply correspond to any of “the facts.”

Thus, one does not report an already constituted meaning of the past; rather, one seeks a narrative that synthesizes the various threads of the past into a coherent, meaningful, and plausible account. Such plausibility may depend on factors (e.g., future events) not at all present to the actors of the events being considered. Furthermore, disinterestedness may be an ideal here, but it is hardly a practical one from a hermeneutic point of view.⁹¹ The positivist’s claim applies primarily to the necessary empirical analyses that precede the actual writing of history.

The second objection, ideological abuse, is one that continually threatens any discourse concerning a cultural study of man. As human beings we are quite simply prey to fallibility, to self-deception, and to self-edifying discourse; we have no neutral vantage point from which to make final judgments. Hayden White adds a new and valuable slant to this ideological component of history when he says:

it may be observed that if historians were to recognize the fictive element in their narratives, this would not mean the degradation of historiography to the status of ideology or propaganda. In fact, this recognition would serve as a potential antidote to the tendency of historians to become captive of ideological preconceptions which they do not recognize as such but honour as the “correct” perception of “the way things really are.” ⁹²

The legitimization of one narrative over another is often due not to its correspondence to “the way things really are” but to its pragmatic and comprehensive nature. Is it edifying, without being narcissistic or egotistic? Does it make sense of what we otherwise know? Is it useful in furthering other and interesting interpretations? It would seem that a primary way to overcome, at least in part, the ideological use of language is to open ourselves to alternative viewpoints and world views, and to alternate interpretations. Anthropological studies and the reading of literature, for example, especially serve in this regard to broaden our knowledge of human society and sharpen our critical capacities with respect to restrictive and dogmatic narratives. A great deal could be said concerning this problem of legitimation, for it has far-reaching social and political ramifications; here I can only point the reader to the various works of Jurgen Habermas as an introduction and working out of some of these important questions.

We have seen that the adequacy of our self-narrations is not a matter of carrying over into language what we already know of ourselves, but is instead a matter of a creative adequation that first generates an explicit sense from our otherwise mute (though significant) prenarrative experience. Imaginative elaboration and potential distortion cannot be exorcised from this latter process. Authenticity, after all, is not the mere recounting of one’s past but, as Heidegger has said, also the projection of one’s possibilities. And besides, the meaning or story of the past, provided we are not hard-core objectivists, continues developing until that life of which it is a constitutive part itself draws to a conclusion.

The Cogito

Before bringing together the results of our investigation of the self, it is appropriate to return at this point to a brief examination of the Cartesian cogito, if only because Descartes inaugurated much of what I am arguing against. In earlier chapters certain problems surrounding the Cartesian cogito were noted, and we are now in a position to pursue these issues more thoroughly. My primary concern will be to bring out, in accordance with our recent investigations, the consequences of the spoken quality of the cogito. In Descartes’s texts it is this spoken quality that is overlooked.

What in particular the Cartesian philosophy instigated was a shift toward an indubitably given subject that could itself become the ground for a more scientific, primarily epistemological philosophy. The Cartesian moment of self-certainty is the guarantor of truth and the overcoming of both relativism and scepticism. This turn toward the subject as what is immediately accessible had the correlative function of relegating objects to the status of mediated, secondary phenomena:

although the things which I sense and which I imagine are perhaps nothing at all apart from me, I am nevertheless sure that those modes of thought which I call sensations and imaginations, only just as far as they are modes of thought, reside and are found with certainty in myself. (Med. III p. 91)

The orientation that Descartes gave philosophy provided the impetus for the unfolding of the Kantian system and continued its effect at least up to Hegel.

There are two problems (though in essence they are one) that arise from the Cartesian position; both are metaphysical. In dispossessing objectivity of a reality other than that of a cogitatum, philosophy became burdened with the problem of accessing the “things in themselves.” A decisive split results between spirit and nature (res cogitans and res extensa).⁹⁴ The second problem concerns the other pole of the cogito—not the status of what thought intends but the I or ego that intends it, that is cognizant of it. If that which is intended has determinations but is ontologically dubitable, the I (ego) has no determinations but is indubitable.

We might wish to see thoughts (cogitata) as determinations of this ego, but they are really only occasions for its apperception. The ego is a res, a something that thinks, a subjectum whose thought is immediately given to itself. The cogitata may be predicated of the ego (as its acts) but they are still nevertheless known by it as objects. That Descartes isolates both I and mind from its acts is clear from the following statement: “Nor can the faculties of willing, perceiving, understanding, and so forth be called parts of the mind, for it is the same mind which wills, perceives, and understands” (Med. VI, p. 139). The mind thus has ontological priority over its own acts.

The ego is, for Descartes, not itself an object but a transcendence. As Kant was later to say, the “I think” must be capable of accompanying all of my representations, but it is not itself a representation.⁹⁵ The essence of the Cartesian subject lies in this moment of self-consciousness, this presence to its representations. However, the I or ego itself appears to be contentless and fundamentally ahistorical, for content and change will properly pertain only to the temporal and spatial representations that the ego has. The ego is thus posited as the ontological ground for the very possibility of both history and representation. Descartes’s I in fact remains anonymous unless we turn back to the cogitata themselves, and also to their historical context and their production.⁹⁶

For all its absoluteness the Cartesian cogito is fraught with frailties. Consider dreams. Can I always actualize the “I think” during dreams? Am “I,” in fact, the dreamer? As is known, Descartes does not clearly address this problem. In saying that even if what I take to be my waking life is nothing but a dream, I nevertheless cannot doubt that I am, Descartes is referring to the “I” that exists even within deception, but not to what we properly call dreams. Similarly, there are also psychological cases such as split personality, where we find two or more possible and even mutually exclusive cogitos!

What needs to be brought out in the Cartesian philosophy is the status of the cogito as language event. As Descartes writes, “I am, I exist, is necessarily true every time that I pronounce it or conceive it in my mind” (Med. II, p. 82). This pronouncing seems to us all-important, for it is difficult to see how conception can be achieved outside this pronunciation (be it aloud or to oneself). Descartes does not come back to pronunciation other than to say:

I am, I exist—that is certain; but for how long do I exist? For as long as I think; for it might perhaps happen, if I totally ceased thinking, that I would at the same time completely cease to be. (Med. II, p. 84)

This statement should remind us of Helen Keller’s observations; it points to the fact that we can take Descartes’s statement more literally than he perhaps intended. For Descartes, however, res cogitans still harbors traits from scholastic ontology, especially unity and indestructibility.

But who or what is the “I”? Is it solely an impersonal transcendental ego, or is it the person we call Descartes? In the text we clearly have a sense of both meanings, though they are not explicitly separated. One of the “egos” clearly has a history, and Descartes in fact offers an autobiographical account of himself in the opening sections of the Discourse on Method. Here the I is not the bloodless and anonymous epistemically foundational subject; and is this I not found throughout the Meditations as the subject who actually inquires after itself, the narrator? It is this latter I—let us call it Descartes—whose history leads itself to pose the question of certainty in a fight against scepticism. It is Descartes who has perceived, doubted, and dreamed, and it is he therefore who embodies the context within which radical doubt may arise. The I of the cogito is located ontologically outside the narrative of Descartes’s systematic meditations and yet is temporally a product of it, parasitic on it. As Merleau-Ponty remarks, “The question is how subjectivity can be both dependent yet irremovable.”⁹⁷

As in the case of Kant, the Cartesian “I think” is a manifestation of the spontaneity of the transcendental ego, the sheer mineness (and hence unity) of experience. But this I is as empty as the bare form of temporality. It is only, and here we refer back to Husserl, in the “unity of a history” that individual persons are manifest, and it is here that the Discourse on Method in fact begins; cogitos come later. An important question to ask at this point concerns the explanatory value of the I disclosed in the cogito. Can the unity or synthesizing powers it exemplifies be demonstrated? Is it in fact anything more than a mode of address? In sum, what does the “I” refer to, what does it designate?

The cogito can be said to be to the extent that I pronounce it, but it is more than this, for I may be just blindly repeating Descartes’s phrase, much as a child repeats its name without knowing the significance accruing to it. The I must become my I; it must be indubitable in my own intuition, my own experience. The cogito must be the expression of my own being, a reenactment of the ontological moment of self-certainty. As Merleau-Ponty has written, “The cogito at which we arrive by reading Descartes is . . . a spoken cogito, put into words and understood in words. . . .”⁹⁸ Thus the “I” is not before the words, just as I never discover myself at the origin of the words I say. The cogito must be spoken, for it is only in the spontaneous upsurge of language and expression that I find myself.⁹⁹ I then acquiesce to the logic of the expression, think myself into being as it were. I become the “I” spoken of. But this “I” is nothing more than an index of the person who speaks it, qua speaker: the speaking being.

The self-referentiality of the “I,” as has already been suggested, lies initially not in a prelinguistic or transcendental subject but in the auto-affection of speaking-hearing. This phenomenon is certainly a relation of oneself to oneself, but, unlike the relation touching-touched, language also designates the subject of this relation by the personal pronoun. One does not simply hear; one also names oneself with the universal subject “I.” Again, the reification of this universal subject seems to me an adequate explanation of the problematic transcendental ego and, as was previously stressed, this arises from a forgetfulness of the constitutive power of speaking; one believes that saying “I” relates back to something other than a speaking subject at the moment of speaking. (Many of the problems surrounding this speaking subject have already been investigated in earlier sections, and will not be repeated here.) The iterability of the cogito, the ability we have continually to repeat or reactivate it, reinforces our belief in the transtemporal unity of the ego and creates the illusion of having a stable identity (a self) throughout the flux of empirical differences.

Historically, the idealist metaphysics of the subject began to be undermined shortly after Hegel, with the rise of dialectical materialism coming out of Feuerbach and receiving its definitive form in Marx. A parallel rejection of the Cartesian standpoint occurs in nineteenth-century positivism. But it was not until the twentieth century that language was clearly perceived as playing a central role in this deconstruction. Heidegger’s replacement of the subject-object dichotomy (along with all its metaphysical baggage) with the notion of Dasein as being-in-the-world, for example, inaugurated a new era in European philosophy by resituating the subject inextricably within both a world and a language.

The Semiotic Subject

It is time to draw together the various threads of the previous sections into a model of the human subject that respects its situatedness in language and signification. I will call this subject the semiotic subject. The label semiotic is preferable to any derivation from the word language since, while remaining within the realm of signification, its extension is broader. Though language is perhaps the most important signifying system, it is clear, to take one example, that art in its various nonlinguistic forms can also express the subject.¹⁰⁰

A person is a being of semiosis, a living body of gestures and articulations that exists in extensive interaction with other acting bodies and the products of semiosis—speech, texts, art works, and meaningful action generally. The development of the person will depend on a reflective grasp of, and habitual participation in, this network of social communication and praxis. The human subject must thus be situated within the structures that sustain it rather than posited as transcendent to them; it must be implicated in the production of such structures but need not be taken as foundational.

I have tried to show how the subject, in losing its autonomy, is both decentered and split, coming to itself across the divide demanded by expression, not in the immediacy of self-transparent intuition. While I have disparaged authorial intention, I have nevertheless sought to retain the notion of intention, of vouloir dire, because it establishes the speaking subject’s essential “wanting to be”—regardless of the degree to which this is determined beforehand by one’s habitus. Though structuralism, for example, has gone a long way toward eradicating the causal efficacy of the subject by reducing it to a puppet of structural systems, it is clear that such an impersonal approach is of limited applicability.

There is, as we saw earlier, a spontaneity of the subject that has to do with the way it gradually appropriates and refigures its social world. This is not to say, however, that the subject can be extracted from this network of processes and significations and still retain an identity as subject, for this spontaneity remains one with the malleability of the social structures themselves. Social structures undoubtedly determine individual possibilities (as is the case for language and history, to which we are subjected); they do so, however, only because, on the other side of the coin, the individual has those possibilities open to it. Such possibilities are what ground the notion of intention. In fact, a more favorable interpretation of structuralism would be to see it not as committing one to the rejection of the subject (and subjectivity) but simply as refusing to separate the subject from the social order, thereby rejecting the notion of a free or autonomous foundational subject. In talking only of structures, structuralism will therefore imply subjects to which these pertain, for the two cannot, in the end, be separated.¹⁰¹ Whether we turn to anthropology or linguistics it is obvious that subjects are always implied; the problem to be dealt with is the function and place of these subjects.¹⁰²

This question of the efficacy and value of the subject ties directly into the question of authorship (and ultimately into authority). It is fairly common, especially in literary circles, to hear of the “death of the author,” the notion that one should examine texts on their own merit rather than viewing them as mirrors of a constituting consciousness. The same argument would hold for speech. There is much that we can agree with in this stance, but it will be useful to consider exactly where this position leaves us with regard to the speaking subject. We shall refer to both Roland Barthes and Michel Foucault on this issue.

Writing, says Barthes, “can no longer designate an operation of recording, notation, representation, ‘depiction’ (as the Classics would say)”, for the “scriptor” does not exist outside the scene of writing.¹⁰³ The author must “die” for the text to begin its own life, a life without final closure, without a final signified content. The text’s future lies in the hands of its readers. But while the psychophysical author (the person) is left behind, what we have called a subject of speech remains:

Linguistically, the author is never more than the instance writing, just as I is nothing more than the instance saying I: language knows a “subject,” not a “person,” and this subject, empty outside the enunciation which defines it, suffices to make the language “hold together,” suffices, that is to say, to exhaust it. (IMT, p. 145)

Parallel to the way Barthes moves from author to scriptor, we must note here a similar move from the speaking subject to the linguistic subject. Just as it may seem banal or even tautological to talk only of a scriptor in the context of writing, so it may also appear obvious that in a purely linguistic context we can only talk of subjects of enunciations and not flesh and blood persons; in a text there are only signified subjects, subjects of speech. This situation need not, however, deny the reality of authors or persons insofar as it restricts its claim to the linguistic.

Barthes’s linguistic “I” is nothing other than the subject of speech we isolated in Benveniste, an “I” that is usually taken to denote a speaking subject. Barthes’s point is that the “I” of a literary text, be it a character or narrator, cannot be innocently ascribed to a “real author” (a phrase we will use for the flesh and blood author). The real author is left behind once the text goes public. Nevertheless, the causal efficacy of this author is attested to in Barthes’s remark: “His only power is to mix writings, to counter the ones with the others. . .” (IMT, p. 146). The author is thus a confluence of intertextuality. He or she cannot escape linguisticality and intertextuality even on the expressive plane:

Did he wish to express himself, he ought at least to know that the inner “thing” he thinks to “translate” is itself only a ready-formed dictionary, its words only explainable through other words, and so on indefinitely. . . . (IMT, p. 146)

Thus, beneath the written (and spoken) words there are only more words, words that are the life of the author, words that even replace passions and feelings. Barthes writes:

Succeeding the Author, the scriptor no longer bears within him the passions, humours, feelings, impressions, but rather this immense dictionary from which he draws a writing that can know no halt: life never does more than imitate the book. . . .” (IMT, p. 147)

This rejection of authorial intention as being anything other than a potential expression is something we should now be familiar with. The real author has no access to thoughts other than those he himself expresses, in language or otherwise.

We have already seen that even passions and feelings are largely inseparable from interpretation and that the so-called neutral impressions of perception are themselves framed by a certain world view which cuts out and defines elements in the perceptual field. Emotions and passions are best seen as a motivation or impetus for expression and action within the symbolic and cultural spheres.¹⁰⁴ The development of our expressive capacity goes hand in hand with the broadening of affective experience. It is in this respect that life can be viewed as imitating literature (or some other signifying medium such as song or film), for literature provides us with a rich vocabulary for articulating, and thus interpreting, experience in ways previously unsuspected.

But whereas language may only “know” a subject, the reader to whom the text is destined also knows persons—the embodied speaking subject. The body is, as we saw earlier, both the site of narration and the site of ascription or predication for utterances. (A parallel can be made for action generally, in terms, say, of a “site of production” and a “site of responsibility”; this would yield an agent.) This ascription is clear for face to face dialogue, but it also pertains to literature. One generally assumes fictional utterances to be the product of a (fictional) speaking spatiotemporally located human body (exempting some science fiction). To put this another way, we know that texts imply authors—authors who are usually named at the beginning of the text. It is to this named or supposed author that the text is attributed, and with this attribution goes a degree of responsibility for that text. This is not to deny, however, that the author is in turn defined in and through the text. We would also not want to deny that a psychological reading can yield insights into the character or personality of the real author, though the proof of such claims would presumably have to be corroborated by something other than the text itself if that text is not to fool us into misguided assumptions.

Barthes explicitly separates, somewhat boldly, the authorial function of texts from that of the persons who write them. Even his autobiography, Roland Barthes, is prefaced by the statement: “It must be considered as if spoken by a character in a novel.”¹⁰⁵ While accepting the general framework of this separation of author-writer, we nevertheless wish to propose a closer link between the two categories, for the human subject is a semiotic subject that must become its own author in order to define itself. As Foucault has said in his influential essay “What Is an Author,”

the subject should not be entirely abandoned. It should be reconsidered, not to restore the theme of an originating subject, but to seize its functions, its intervention in discourse, and its systems of dependencies.¹⁰⁶

The path we have been pursuing throughout this work is one that describes just these dependencies of the subject on language and discourse. We can only definitively separate the subject from discourse at the expense of that subject, for the subject is an implicate of its discourse. Foucault’s remark points to a needed rehabilitation or reinscription of this subject into the realm of signification (literary or otherwise). The “author function,” as Foucault says, is one of the forms in which the subject is manifest, though not necessarily in a one-to-one relation.¹⁰⁷ Texts, like speeches and even casual discussions, have social conditions and expectations attached to them. Each mode of signification may have its particular restraints and freedoms. Each may set up a different subject as its implied origin. Consider, for example, the difference between a poem, a scientific manual, and a charter of rights. Even from the point of view of the speaker, there are different personae, different roles to be assumed, in these different discourses.

Is there a central organizing “I” behind these various roles? Only if we should choose or be led to believe so: if we believe that the “true me,” for example, appears under such and such conditions. Later, of course, we may come to see the folly of this belief, or revise it in light of certain experiences or theories. Either way, however, it is not that we behave contradictorily in our lives, for this is still to postulate a subject at the center of things that is enduring and is able to contradict itself. It is rather the case that social reality demands, during our life (even during a single day), that we be different, diverse, that we assume various guises or roles as our own. As Barthes remarks,

This is why, when we speak today of a divided subject . . . it is a diffraction which is intended, a dispersion of energy in which there remains neither a central core nor a structure of meaning: I am not contradictory, I am dispersed.¹⁰⁸

Even the body can be seen as plural, dispersed among a repertoire of roles such as the sensual, digestive, athletic, sick, emotive, mimetic, gestural, and so on.¹⁰⁹

Given this diversity, the possibility of unity for the subject can only arise through two primary channels: routine activity at the level of praxis and acts of self-narration. In the first case, identity is a matter of repetition, of having a schedule that one repeats daily, weekly, etc. Because this type of identity is largely unconscious, it will take something like sickness to bring our implicit dependence on it to the foreground. My earlier discussion of habitus has much to do with this form of identity. The second option, self-narration, is the properly conscious form of human identity. Only here is the implicit order (or disorder) and structure of our lives taken up into conscious understanding. It is also in narration that we seek to tie together the more disparate strands of our lives, of our history.

I would like now to present a model of the human subject, the semiotic subject, that respects Barthes’s observations while fulfilling Foucault’s demand for a resituated subject. This model, which I employed in earlier sections of this book, is based on a tripartite division of the subject: the speaking subject or material agent of discourse, the subject of speech or purely linguistic subject of the discourse (designated by personal pronouns and other deictical indications), and the spoken subject or subject produced through or by the discourse as a result of its effect on a reader-listener.¹¹⁰

Thus, for example, in the case of self-narration, of the past, the speaking subject is myself qua language user and “repository of images” (and hence conditioned and restricted by that language, by tradition, and by past experience). This narrative then sets up a subject of speech, the character signified by the pronoun I and involved in a certain narrated life situation. What then makes this narrative personally historical or autobiographical is that I correlatively become the spoken subject of the narrative—just as a spectator might identify with some character in a play or film. Of course this third stage might be thwarted; the identification might not occur or it may occur when the subject of speech is a fabrication or lie with no direct relation to one’s actual memories or prior narratives.

A diagram may help us see the relations between the three subjects more clearly.

THE SEMIOTIC SUBJECT

The diagram shows two primary realms: signification and body. Through expression the embodied speaking subject enters the realm of signification. The utterance, if it is autobiographical, relates back to the embodied subject in two ways: by implication and by participation. The first generally occurs on the part of an addressee or receiver. If I write a letter to a friend, for example, I become for the reader the implied author (2) of the correspondence. If the letter were anonymous the exact site of ascription for the implied author could be lacking; this would lead to implied author (1). A more common academic use of implied author (1) is the literary one, where a text is said to set up an implied author that cannot be naively identified with the real author. We tend to attribute, in one way or another, almost all texts to an implied author.¹¹¹ One may even adopt the receiver standpoint to one’s own utterances.

By participation is meant the various forms of identification that the embodied subject has in relation to an utterance and its subject of speech (a relation that may also be seen as one of mimesis, a term that is central is Ricoeur’s more recent work¹¹²). Very often the immediacy of listening to speaking carries one in an unbroken and unreflective reciprocity forward into further articulations. Here there can be immediate identification with the subject of speech. But such immediacy can break down.

When narrating one’s past it is often the case that first attempts are unsatisfactory: the recollections are seen as too sketchy or perhaps as fabrications; there is a rejection of the implied subject (of speech) as being or properly representing oneself. This rejection of one’s own narrative is, as we have seen, central to the theory grounding psychoanalysis. Another revealing example of participation occurs in acts of rage (or drunkenness) where one’s emotions cause one to say things that at the time one positively identifies with, and which later seem exorbitant and excessive. The act of participation can also take us out of ourselves. In reading a novel a certain relinquishment of self¹¹³ occurs that is necessary for us to participate in the values, characters, and action of the drama. The sympathy we feel for a certain character, say, is a form of this participation in narratives that are not of our own making.¹¹⁴

The spoken subject that participation results in is particularly evident in cinema. Consider how we often, on the visual level, identify with the camera’s perspective, especially if the shot is taken from approximately head height. Again, consider how we can walk out of an adventure film feeling somewhat inspired by the hero. In literary criticism this manipulation of the viewing-reading subject is recognized as an essential function of texts. Texts set up not only an implied author but also an implied reader. In a text, the presenting of material in a certain way may not only place restraints on the reader but also set up certain expectations and biases in the reader.¹¹⁵

The embodied subject can be characterized in a general way, following Heidegger, as being in a certain state of mind with its moods and degrees and modes of understanding.¹¹⁶ These general characteristics are what provide the impetus and background for expression, much as unexpressed emotions serve such an impetus in Taylor’s analysis, and they are what develop and change in acts of participation.¹¹⁷ I have labeled the relation between the spoken subject and the speaking subject as desire, for this seems to capture the impetus that changing states have for the subject. Earlier I had occasion to define subjectivity as being the possibility of expression. Here we see, in more detail, the dynamics of this more or less continuous state of the subject.

For structuralism (and Barthes in his earlier structural phase), semiotics, and narrative theory (as applied to literature), it is the subject of speech that has been especially emphasized. Indeed, it is here that the stories we tell of ourselves appear in the public arena, and hence where the linguistic subject is constituted. But rather than leave this subject of speech floating in linguistic space, which is the sense one gets from Barthes, the above model attempts to integrate the body back into the equation, not of course as the positivist material body of science but as the speaking-feeling embodied subject (the person). Earlier I talked of the body as both the site of narration (the speaking subject) and site of ascription (implied author (2)) for the subject; it is here that our commonsense notions of ourselves as embodied subjects are satisfied.

What the above model of the semiotic subject seeks to emphasize is the division of the subject into different moments with no central and organizing core. Subjectivity is itself blind without mediation through the realm of signification, but signification is not a neutral mirroring process. Subjectivity, as a form of vouloir dire (a wanting to say, to be, to do), is manifest as the speaking embodied subject that seeks to carry over into expression the implicit truth of itself (its implicit history or story).¹¹⁸ This expression is a creative adequation to what is only schematically given in a quasi-narrative form. The “I” then exists in its communicable form as the subject of speech. Already there is a split here between the speaking subject (what Roman Jakobson called the “subject of the enunciation”) and the subject of speech (the “subject of the enounced”).¹¹⁹

In self-narration the final stage of the semiotic subject is identification with and appropriation of this linguistic subject through the reading-listening process. A split or noncoincidence in the subject is also apparent here due to the interpretive nature of this participation. One may not, for example, accept the expression as an adequate representative of oneself, which may cause the cycle to continue again. This cycle of ever new signification and appropriation is, of course, none other than the dynamic framework within which personal development takes place.¹²⁰ To return to Charles Taylor: man “cannot be understood simply as an object among objects, for his life incorporates an interpretation, an expression of what cannot exist unexpressed, because the self that is to be interpreted is essentially that of a being who self-interprets.”¹²¹