Paper to be published in Theory & Psychology, 2002

Running Head: From Thought to Voice

From "the thought is the thinker" to "the voice is the speaker"

William James and the Dialogical Self

John Barresi

Department of Psychology

Dalhousie University

Acknowledgement: I wish to thank Ray Martin, Dan McAdams and Hubert Hermans for their helpful comments on ms. versions of the present article. Please send reprint requests and other correspondence to John Barresi, Department of Psychology, Dalhousie University, Halifax, NS B3H 4J1, or

KEYWORDS: William James, Bakhtin, Hermans, Dialogical Self, Consciousness, Self


In 1890, William James provided a rich account of self in his Principles of Psychology. A hundred years later, Hermans and his colleagues have provided an equally rich account of self, most particularly in Hermans and Kempen's The Dialogical Self (1993). One of the key distinctions in the former work is that between the "I" and "Me," a distinction that Hermans builds on in his concept of the dialogical self. Nevertheless, there are differences between the two theories of self that are developed out of this distinction. In the present article I look closely at some of the differences between the two theories, and also look closely at Bakhtin's contribution to the development of the dialogical self. Through a careful comparison of the works of James, Bakhtin, and Hermans, I try to indicate how our understanding of the self has changed over the past hundred years. At the same time, the comparison of some of the central ideas of these three authors provides a basis for a critical evaluation of certain aspects of Hermans' theory of the dialogical self.

From "the thought is the thinker" to "the voice is the speaker"

William James and the Dialogical Self

Over a hundred years ago, in his Principles of Psychology (1890), William James put forward a fascinating account of the self. In that theory, he makes a distinction between two aspects of self, the self as subject, or the "I," and the self as object, or the "Me." James goes on to investigate the nature of these two aspects of self. He concludes that the me comes in three basic types: the "material me", the "social me", and the "spiritual me." As for the I, James concludes that, at least for the purposes of psychology, there is no need to postulate a subject of experiences, a metaphysical I that goes beyond the physical being who does the thinking. Rather, he concludes that 'the passing thought ... is itself the thinker' (1890, p. 401). A hundred years later, Hermans and his colleagues (Hermans & Hermans-Jansen, in press; Hermans, Kempen, & van Loon, 1992; Hermans & Kempen, 1993; Hermans, Rijks, & Kempen,1993) have put forward another fascinating account of the self, which they call the "dialogical self." They claim that this notion of self has its roots in James' distinction between the I and Me, but claim that it also depends on Bakhtin's (1973) theory of the polyphonic novel. In summarizing how James' I and Me relate to their use of Bakhtin's notion of the multiplicity of voices in a polyphonic novel, Hermans claims that "the narrative 'translation' of these terms [i.e., I & Me] . . . lead to the notion of the polyphonic novel" (Hermans, et. al.,1993, p. 210). According to this translation, each me in James becomes a character in the polyphonic novel of self, and each of these me's has, not a thinking I but, a speaking voice to represent its point-of-view vis-a-vis other characters and their voices in the polyphonic or dialogical self.

But, can the transition from James to the dialogical self be that simple? Can a hundred years of thought about the self be reduced to a transformation from "the thought is the thinker" to "the voice is the speaker?" And does this result in a conception of a self as a "polyphonic novel" in Bakhtin's sense of this term? In their initial empirical study of the dialogical self, Hermans, Rijks, & Kempen (1993) state that they are "approaching the self as a polyphonic novel; that is, a novel where different voices, often of a markedly different character and representing a multiplicity of relatively independent worlds, interact to create a self-narrative" (p. 208). This idea, which has its basis in Bakhtin's (1973) theory of Dostoevsky's poetics, presupposes that the self is constituted in some fashion out of a multitude of voices, each with its own quasi-independent perspective, and that these voices are in a dialogical relationship with each other. In elaborating what they mean by a "voice" and its implications for the dialogical self the authors state: "In order to become dialogical, personal meanings (e.g., an idea, a thought about something, a judgment) must be embodied. Once embodied, there is a 'voice' which creates utterances that can be meaningfully related to the utterances of another voice. It is only when an idea or thought is endowed with a voice and expressed as emanating from a personal position in relation to others that dialogical relations emerge" (p. 212-3).

The notion that an "idea or thought" becomes a "voice" according to this theory, confirms Hermans' suggestion that there is a simple and direct connection between the theory of the dialogical self and William James' theory of the self. Indeed, he provides further confirmation when he suggests not only that the polyphonic novel is a "narrative translation" of James' concepts of I and me, but also that the "I is the author" and the "me is the actor." If, as James suggests, each thought is the thinker, and each thought as I thinks for its own particular me, then by converting this relation into one of "author" and "actor," where the author and actor are one, we have not multiple thoughts as thinkers, but multiple authors writing or voicing the point of view of their actor me's. If, in addition, we assume that the variety of thoughts or ideas come into dialogue with each other, then the result would be a polyphony of voices, representing each of the thought-ideas, or I-me's.

However, this "translation" of James into Bakhtin and the dialogical self may be too quick. What we need to do is to take a closer look at both James and Bakhtin as well as at how Hermans actually uses the concept of dialogical self, once he puts his sources aside. In what follows, I first look at James' theory of the I-me distinction, and, in particular, at his concept of "the thought is the thinker." I follow this by looking at Bakhtin's work, where I will focus on his early epistemological investigations, as well as on how his theory of the polyphonic novel relates to these early ideas. I then return to a consideration of Hermans' theory of the dialogical self and evaluate it in light of what we learn about James' and Bakhtin's theories.

James' theory of the self

The main distinction that James draws at the beginning of his chapter on Self in Principles of Psychology (1890) is between the self as known (or me) and the self as knower (or I). He begins his discussion of the self as known with the claim that: "In its widest sense a man's Self is the sum total of all that he CAN call his, not only his body and his psychic powers, but his clothes and his wife and children, his ancestors and friends, his reputation and works, his lands and horses, and yacht and bank-account" (1690, I, p. 291). What makes "all these things" part of a man's Self is that they "give him the same emotions. If they wax and prosper, he feels triumphant; if they dwindle and die away, he feels cast down, - not necessarily in the same degree for each thing, but in much the same way for all" (pp. 291-2). What determines the boundary between self and not-self is one's emotional attitude about an object or thought. The things, people, or thoughts, with which one identifies are quite literally part of Me, so long as what happens to them is experienced as something happening to me. They are the self, though some of these objects may be more centrally the self than others. Indeed, later on, after he distinguished among the material, social and spiritual selves, he focuses on how we decide wherein the central part of the self resides, and, again, it is the extent to which we identify with (or find "warmth and intimacy" in) particular aspects of self that ultimately decides the issue.

This issue of what is most central to the self plays itself out in a number of ways in James' discussion of the me. One way it does so is very closely related to later developments in the narrative approaches to self, particularly as found in McAdams' narrative theory of personal identity (e.g., McAdams, 1990). James provides a wonderful account of how we choose among rival or conflicting selves to create an identity for our selves:

I am often confronted by the necessity of standing by one of my empirical selves and relinquishing the rest. Not that I would not, if I could, be both handsome and fat and well dressed, and a great athlete, and make a million a year, be a wit, a bon-vivant, and a lady-killer, as well as a philosopher; a philanthropist, statesman, warrior, and African explorer, as well as a 'tone-poet' and saint. But the thing is simply impossible. The millionaire's work would run counter to the saint's; the bon-vivant and the philanthropist would trip each other up; the philosopher and the lady-killer could not well keep house in the same tenement of clay. Such different characters may conceivably at the outset of life be alike possible to a man. But to make any one of them actual, the rest must more or less be suppressed. So the seeker of his truest, strongest, deepest self must review the list carefully, and pick out the one on which to stake his salvation. All other selves thereupon become unreal, but the fortunes of this self are real. Its failures are real failures, its triumphs real triumphs, carrying shame and gladness with them. (James, 1890, Vol I, p. 309-10)

In this account of rivalry among empirical selves (or selves as known) and its resolution, James makes clear that the alternative selves are "characters" satisfying literary ideal forms, and that a person is forced to choose among these characters in cases where they come in conflict with each other for domination of the individual's behavior. It is also clear from his account that ultimately some character or characters are chosen and subsequently reaffirmed as the identity upon which the individual must "stake his salvation." McAdams recognizes that James' account here comes close to his own notion that the resolution of the Eriksonian identity crisis involves the selection or creation of a narrative personal identity that becomes reaffirmed or revised throughout adulthood. However, for McAdams this identity typically involves more than one character or 'imago' that can complement each other but also sometimes oppose each other in the individual's constantly updated narrative of identity (McAdams, 1990, p. 173f).

Hermans (Hermans & Hermans-Jansen, in press) also discusses this description of conflicting selves in James. However, what is important from the point of view of the theory of the dialogical self, is that there are multiple selves here, each of which could be given a voice and allowed to engage in a dialogue. Whereas, for McAdams it appears that there is a kind of magisterial (or Bakhtinian monological) I who generates a narrative self, and constantly updates it, for Hermans there are merely different me's depicted in each of these characters ( but cf. McAdams, 1995; 1997), who must be given the opportunity to speak for themselves, by being provided an Ior I-position to speak for them.

In James' own account of conflict and resolution among possible identities, he makes a general point that connects this discussion to the issue of I-Me relations. He closes the very passage just quoted with the following observation:

This is as strong an example as there is of that selective industry of the mind on which I insisted some pages back (p. 284 ff.). Our thought, incessantly deciding, among many things of a kind, which ones for it shall be realities, here chooses one of many possible selves or characters, and forthwith reckons it no shame to fail in any of those not adopted expressly as its own. (James, 1890, Vol. I, p. 310)

James is here referring to his discussion in the previous chapter on the Stream of Thought. In that chapter he outlines five properties of the stream of thought, the fifth of which is that "it is always interested more in one part of its objects than in another, and welcomes or rejects, or chooses, all the while it thinks." (I, 284).In that discussion, as in the present case, the focus is on selection among alternative possibilities. James does not emphasize what will become clearer in his discussion of the I, that each thought is, in a sense, an independent being, whose preferences and choices may be uniquely its own. While each thought connects to past thoughts each thought is metaphysically distinct. There is no absolute necessity that its preferences will be those of previous thoughts, though it has a strong bias in that direction.

The issue of what is most central to the self appears once again in James discussion of the spiritual self, which he defines as "either the entire stream of our personal consciousness, or the present 'segment' or 'section' of that stream, according as we take a broader or a narrower view " (1890, Vol. I, p. 296):

When we think of ourselves as thinkers, all the other ingredients of Me seem relatively external possessions. Even within the spiritual Me some ingredients seem more external than others. . . . The more active-feeling states of consciousness are . . . the more central portions of the spiritual Me. (James, 1892, p. 181)

These active-feeling states are "the very core and nucleus of our self, as we know it."

And these states are "often held to be a direct revelation of the living substance of our Soul." But whether this is so or not is "an ulterior question," a question James attempts to answer when he turns to what constitutes the self as knower, or I.

James sums up his view of the self as knower or I and its relationship to the me:

The consciousness of Self involves a stream of thought, each part of which as 'I' can remember those which went before, know the things they knew, and care paramountly for certain ones among them as 'Me,' and appropriate to these the rest. (1892, p. 215)

The I which knows these past thoughts and appropriates them, "for psychological purposes" is neither a "Soul" nor "transcendental Ego" outside of time. "It is a thought at each moment different from that of the last moment, but appropriative ofthe latter, together with all that the latter called its own" (1892, p. 215).

James' logic here is that, since the stream of thought is constantly changing, there is no reason to suppose some "fixed" entity beyond the stream itself. Rather, there are "pulses of consciousness" or thoughts, which are unified in themselves, involving, among other things, the immediate awareness of the body, and that these thoughts, as I's can remember and appropriate prior thoughts in the stream:

The nucleus of the 'me' is always the bodily existence felt to be present at the time. Whatever remembered-past-feelings resemble this present feeling are deemed to belong to the same me with it. (1890, Vol. I, p. 400)

Despite this statement, James does not consistently assert that the me associated with a particular I must be based on present bodily feelings. The exceptions that he makes include cases where present feelings go against one's general feelings, so that one's current feelings are a surprise to one's self. In such examples, one might say that a continuation of a prior thought is reflecting on a new thought which is itself based on current feelings. So the perspective here is not that of a "current" I and associated Me, but an I-me combination which has its base in a remembered continuity with the past.

Let me summarize what I take to be important about James' view, with respect to its anticipation of later developments. First, for James the me or self as object is very complex, and brings together a variety of conceptions of self - e.g., material, social, & spiritual selves. And for James these multiple sources of self can come into conflict with each other (e.g., his discussion of rivalry among selves), with some having greater priority than others (in particular the spiritual over the other types of selves). With respect to the I his position suggests even more diversity in perspective, and potential conflict among selves, because, for James, each thought or pulse of consciousness in the stream of thought has an independent existence as a representative of a particular point of view in space and time, even though it generally appropriates to itself other past thoughts (or I-me combinations), depending on their "warmth and intimacy," when compared to the point of view of the present thought (I-Me). However, what is also obvious, but somewhat less developed in James' general account, is that even during a brief period of time, which may be viewed as being dominated by a single thought, or I-Me, that there can be opposing thoughts. This becomes apparent in his examples of mental illness and forms of dissociation in his chapters on self, as well as in other works, in particular his Varieties of Religious Experience (James, 1902), where he provides extensive discussions of internal conflict in his chapters on the "sick soul" and "divided self." Such discussions suggest that James and Hermans' dialogical self may not be all that far apart (see, Barresi & Juckes, 1997).

Bakhtin's theory of the self

Hermans built his account of the dialogical self by developing ideas found in Bakhtin's major work on Dostoevsky's poetics (Bakhtin, 1973), which was initial published in 1929. Considered by many to be his most important work, it is here where Bakhtin develops his mature conception of the polyphonic novel, and of Dostoevsky's role in developing that genre. However, in recent years, there has been a growing interest among Bakhtin scholars in his earlier, more epistemologically oriented works, which were not published during his lifetime. In particular, in his essay, "Author and Hero in Aesthetic Activity," Bakhtin (1990) presents a sophisticated view of self and other understanding, that is now thought to provide the philosophical core for his later intellectual work. In the present discussion, I will draw on this early essay and show how his later work was built on these early insights.

The key epistemological theme that Bakhtin develops in "Author and Hero in Aesthetic Activity" is the radical difference that exists between our understanding of self and other. Although his terminology differs from that which I have used in recent articles (e.g., Barresi & Moore, 1996; Barresi, 2000), he pursues a theme similar to one developed in these articles. The main related distinction that I make is that between "first person" and "third person" information about self and other. Adapting this distinction to Bakhtin's essay, by first person information I mean the experience and viewpoint that an individual has about his or her own activity as he or she is engaged in it. It is the concrete epistemological standpoint that the person has when engaged in current activity. By third person information I mean the experience that another person has of an actor when observing the actor engaged in an activity. Now, it seems both to Bakhtin and to myself that, in order to fully appreciate a person's activity, whether that person be self or other, what is required is some sort of integration of first and third person information. The actor cannot fully appreciate self, without, in some way, integrating third person information about self with first person information about self. But, according to Bakhtin, the actor can never fully succeed in integrating such information. Nor can the observer fully integrate these two types of information. The epistemological positions of the actor and the observer can never be merged, yet there is a constant need to do so. I see this as Bakhtin's enormous insight. It is one that soon leads him to discover the importance of the dialogical relation between self and other in the quest for further knowledge of self and other.

In my own previous work (see Barresi & Moore, 1996), I have suggested that an understanding of the activity of self or other always involves some sort of integration of first and third person information. In particular, in the first instance, what we do is combine concrete first person information about self, with concrete third person information about another individual engaged in the same kind of activity as we are, in order to understand what it means for self and/or other to engage in that activity. The infant first comes to understand its own or another's activity through a process or imitation or spontaneous mimicry of the activity of another person. However, when it comes to understanding self in circumstances where we are the only individual engaged in the activity, we must combine directly perceived or current first person information about self with imagined third person information about self in order to understand our intentional activity. And we must integrate imagined first person information about another person, with directly perceived third person information of the other person's actions in order to understand the other's intentional activity. But in these circumstances where self and other are engaged in different activities an epistemological problem arises that I did not previously consider, but which seems to be at the center of Bakhtin's view.

While Bakhtin agrees that understanding self and other involves the attempt to integrate first and third person sources of information, he is much more skeptical than I have been in my previous writings about the possibility of full integration of first and third person information, whether it be about self or other. According to Bakhtin, we can never understand the other's activity in the same way that we understand our own activity, because of the distinct epistemological positions that we have with respect to self and other. With respect to self, we are always situated in our current position in space and time and bound to our current future-oriented perspective on action. The world about us is viewed as a medium for action, rather than an environment that determines action. The future appears open ended with respect to one's self, but not in our perception of the other, whose behavior appears to be determined by the environment. Nevertheless, we try to perceive ourselves from a third person perspective, and we try to understand the other from a first person perspective. In engaging in each of these activities we "transgress" the boundary between self and other. We require the other's valuations and expressions of behavior toward us, to place ourselves more fully in the world with the other, and we use our capacity to empathize with the other in order to more fully appreciate the activity of the other, including activity directed toward us. But, according to Bakhtin, neither of these attempts to cross over from self to other, or to complete an understanding of self or other is fully successful. With respect to the other there is, in a sense, a sort of objective completion, because we can fill-in an internal life to the objective other, and can thus conceive of the other as a complete being in the world. But, with respect to self, however hard we try to get outside of ourselves, partly through adopting the other's viewpoint of self, we can never fully enclose ourselves in the world. However hard we try to observe our own activity from the other's point of view, we always bring that viewpoint within the horizon of our own current activity, which is the ultimate epistemological position for all activity, including our attempts at understanding of self and of other.

In his essay on "Author and Hero," Bakhtin argues that successful aesthetic activity requires that the author always keep some distance from his hero, so that the hero's inner perspective of his activities can be 'completed' and placed in the wider environment of the overall narrative - whether this be in an autobiography, novel or other narrative. He complains here that Dostoevsky fails at doing this because, as author, he empathizes too much with his heroes and thus adopts their own point of view, not providing the necessary completion, from an external authorial viewpoint. But by the time that Bakhtin wrote his book on Dostoevsky's poetics, he has reversed his opinion. In that book, he praises Dostoevsky for overcoming traditional forms where the author has unified 'monological' authority over his heroes. Instead, Dostoevsky has created a polyphonic novel, where the heroes are their own authors and on an equal footing with each other and with Dostoevsky. Each hero depicts the world from his own point of view, and various heroes are put face-to-face in dialogue with each other, but with none of the heroes, nor Dostoevsky, himself, allowed to have the final word. Each viewpoint of reality is allowed to stand on its own, contested by alternative viewpoints, but not fully contained within any particular hero's (or the author's) view:

Here [in Dostoevsky's novels], we don't have a great number of destinies and lives developing within a single objective world, enlightened by the consciousness of the author alone; rather we have a plurality of consciousnesses, with equal rights, each with its own world, combining in the unity of the event but nonetheless without fusing. (Todorov, 1988, p. 104)

We can see here, how Bakhtin's epistemological position appears in the polyphonic novel. The point is that no first person perspective gets completed by being enclosed within the perspective or consciousness of another person, or, indeed, in the consciousness of the same person at a later time. Rather, these perspectives or consciousnesses meet each other in space and time, in narrative (and deeds) often directed at each other, but never displacing or overcoming the other. Bakhtin's last fragment, written shortly before he died, indicates that this theme is one that he held throughout his life, and one that constantly refreshed his thought:

There is no first or last discourse, and dialogical context knows no limits (it disappears into an unlimited past and in our unlimited future). Even past meanings, that is in those that have arisen in the dialogues of past centuries, can never be stable (complete once and for all, finished), they will always change (renewing themselves) in the course of the dialogue's subsequent development, and yet to come . . . . The problem of the great temporality. (Todorov, 1988, p. 110)

The overall implication of Bakhtin's epistemological viewpoint and its later development into a dialogical principle is that the individual acting at a particular point in space and time is creative with language and in deed, and that the self here is a unique particular perspective on reality, an "author" but not a "hero" and as such cannot be an object to him or herself - or a hero in the story of his or her life. In fact, the action is done on a much more narrow plane, pursuing immediate goals. Only later will these acts be interpreted in narrative, but no narrative will fully capture them - either in autobiography or biography. Any act will always be open to revised interpretation. So the person is always dealing with an open future and the past cannot "determine" it - as there is no fixed interpretation of the past, since the past is constantly creatively reinterpreted by the actor as author if by no-one else.

Any two people must be "open ended" concrete particulars. We try to "enclose" each other in our conscious narratives, while not being able to 'enclose' ourselves. Yet - and this is the other side of the coin - we can't really know ourselves without having the other person's input. When we look in the mirror we can only see a "personification" or performance of our self - we can never really get hold of ourselves, because we can't see ourselves "naturally" at all. It is only in actual interaction or dialogue with others that we can come to know the third person point of view of who we are. And without it our vision of our self is incomplete. But, even with it, it is incomplete - because of the openness of the future, and the openness of interpretation of the past.

The key problem that I see now with my own previous accounts of the understanding of self and other (Barresi and Moore, 1996; Barresi, 2000), and I think this applies to Hermans' conception of the dialogical self as well, is that I put too much stress on how imagination can fill-in missing information. Actually, imagination can never fully achieve that job. As a result, no two speakers or actors can ever completely agree in their interpretations of each other's activities. So self and alter are always essentially in dialogue. And this applies not only to two distinct individuals, but also to the same individual across time, that is between past, present and future selves. Empathy can only go so far. At some point it breaks down, otherwise self and other would fuse and not have distinct viewpoints. Self and other, or present self versus past or future self, always have distinct and irreducible consciousnesses. And each has its own inner story or narrative to tell that cannot be captured within the consciousness or global narrative of another. So the story is never told that fully integrates first and third person perspectives. But despite this limitation of imagination, there is a sense in which we know more than we can say. The characters that we create or the models we create of others do seem to have a life of their own. Writers often take this view, as do people with imaginary friends. The very success of Dostoevsky at producing a polyphonic novel suggests that creativity can somehow move beyond the boundary of self to construct an quasi-independent other. And one can suppose that our models of real people are somehow richer in structure than mere imaginative construction. How this is possible is unclear, but it may be similar to how we predict where a ball will land that is thrown in the air. Obviously, it isn't imagination or logic that gives us the answer. We have a model that can simulate, or at least anticipate, physical reality, richer than either imagination or logic, that we can rely on. Even so, such models can never fully simulate or anticipate reality, at least insofar as they originate within our selves and our limited epistemological perspective rather than depending on direct experience of reality outside of our selves.

Hermans, James, Bakhtin and the dialogical self

Returning now to Hermans' theory of the dialogical self: The weakness that I see in Hermans' current formulation of this theoretical framework is the assumption that he makes that an individual can adopt a narrative or authorial stance, somehow above the characters that make up the polyphonic and dialogical self, and can freely move the narrative "I-position" from one character to another to give each their own voice. According to this conception, an individual at a particular time can speak in his or her own dominant voice or character, as well as in the voice of various alter-ego's with almost equal facility. Although Hermans & Kempen (1993) indicate that, because of dominance relations among different characters or selves, some of their voices are undeveloped relative to others, they also suggest that, when we adopt a narrative and reflective stance above selves, we become a 'Self' that is relatively free to move the I-position among characters or selves, even to weaker ones, so that each can express its own view in the first person, and can comment on and evaluate various points-of-view of other selves. Yet, in apparent contradiction to this general principle, they concede that even this "Self" can sometimes be dominated by subselves, so that it might be better to conceive the Self as "in the middle of a highly dynamic field of criss-cross dialogical relationships among possible positions, subjected to influences from all sides" ( p. 98).

This final dynamic description appears more consistent with the position held here. If my account of James' notion of the thought as thinker and of Bakhtin's epistemology of self and other is correct, then it is not just some of the time that we are dominated by some idea or self. Rather, each and every time that an individual engages in reflective valuations from the points of view of self and others, the person engaged in this enterprise has his or her own current consciousness, and its associated sense of self, as the source and ultimate arbiter of the alternative characters that it creates as well as the voices that it gives to these characters. There is no way to move outside of the situated consciousness of the present speaker/knower. To suggest that this can happen is to assume that the present consciousness can enclose itself within a model it creates of another consciousness. Though Bakhtin thought that Dostoevsky may have come close to providing an existence proof of this possibility, ultimately, the assumption of one consciousness fully enclosing another, is incoherent. Rather, what Dostoevsky accomplishes is to create the aesthetic illusion of a real dialogue among independent selves. And perhaps that is as close to the ideal as we can ever come, to escaping the limits of the immediate self: that is to create an illusion of dialogue among independent selves by our attempts to give voice to different characters or selves within us that we create, or to models of the characters of others who we know. If that is so then Hermans' methodology, while not fully achieving the illusion it attempts to create, of a moving I-position among characters, comes as close as we can to obtaining true dialogue among selves.

I believe that Hermans' methodology, though in fact limited epistemologically, is productive in expanding the horizon of reflective self-consciousness by its very attempt to construct a polyphonic novel of self. Though not fully achieving independent voices without 'ventriloquism' that involves the author's own current voice behind all created characters, it does open up possibilities for relatively independent self-valuations, to provide alternative 'third-person' perspectives on self. And it seems to me that Hermans' methodology does achieve more than might be expected if only one real voice were involved in the dialogical relations that are generated. For instance, in the case of "Mary" and the "witch" described in Hermans & Hermans-Jansen (1995), as a therapeutic technique, Mary is asked to give an independent voice to a side of her self that is labeled the witch. Even though it is Mary, herself, from her own current spatial-temporal standpoint, who must shift I-positions between her own view and that of the witch, it is, nevertheless, the case that each position at the start of the inquiry appears to have a relatively independent set of valuations and affective judgements of each other's valuations. Perhaps more important than whether these valuations and judgements are really independent of each other is the effectiveness of this methodology in getting Mary to reflect deeply on two aspects of her personality, and, ultimately, to obtain a more unified and adaptive integration of these two aspects of self. This is particularly seen in the valuations and affective judgements generated a year later in therapy, when the two selves are so integrated that Mary is unsuccessful in her attempt to give the witch and herself independent voices. As Mary and the witch have become one, the ability of Mary to create the illusion of another self fails. Perhaps, unlike Dostoevsky, whose personality may have maintained and perhaps even further developed its polyphony of selves in his later life, Mary as an author of herself has become more integrated, so her polyphonic self has become more a unified monological self.

Indeed, for many of us, it is a goal we try to achieve in life - to unify the multiple thoughts or voices that we have within us into a purposive self whose narrative is that of a unified self. And it is only when the internal dialogue becomes monologue that we experience integrated selfhood. But such integration never lasts. Things change, new possibilities arise, and limitations in our monological view of self discovered. So the dialogue continues between the present self and past and future selves as well as with others. Bakhtin is right, there can never be closure on the self or full identification with the other. The present self is always open to the future, and the past can always change its meaning. And there are always others - real others - with whom to engage in dialogue and mutual interaction. Even if the self could enclose its own past and future into the present, it could never enclose the consciousness or activity of the other. The other is always an unfinished unknown. And just as our engagement with the other is undetermined, so to is our relationship with our self in that engagement. And so the dialogue continues. . . and so also the thinking thought. . . .


Bakhtin, M. (1973) Problems of Dostoevsky's Poetics, (R.W. Rotsel, trans). USA: Ardis.

Bakhtin, M. (1990) Art and Answerability: Early Philosophical Essays by M.M. Bakhtin, (edited by Michael Holquist and Vadim Liapunov). Austin: University Of Texas Press.

Barresi, J. (2000) Intentional relations and divergent perspectives in social understanding. In S. Gallagher and S. Watson (Eds.), Ipseity and Alterity, Special Issue of Arob@se: Journal des lettres et sciences humaines, 4, 74-99.

Barresi, J. & Moore, C. (1996). Intentional relations and social understanding. Behavioral and Brain Sciences, 19, 107-122.

Barresi, J. & Juckes T. (1997) Personology and the narrative interpretation of lives. Journal of Personality, 65, 693-719.

Hermans, H. & Hermans-Jansen, E. (1995) Self-Narratives: the Construction of Meaning in Psychotherapy. New York: The Guilford Press. (pp.187-195).

Hermans, H. & Hermans-Jansen, E. (in press) Dialogical processes and the development of self. In: J. Valsiner & K. Connolly (Eds.), Handbook of Developmental Psychology. London: Sage.

Hermans, H. & Kempen, H. (1993) The Dialogical Self: Meaning as Movement. New York: Academic Press, Inc.

Hermans, H., Kempen, H. & van Loon, R. (1992) The dialogical self: Beyond individualism and rationalism, American Psychologist, 47, 23-33.

Hermans, H., Rijks, T, & Kempen, H. (1993) Imaginal dialogues in the self: Theory and method. Journal of Personality, 61, 207-235.

James, W. (1900) Principles of Psychology, Vol. I. New York: Henry Holt and Company.

James, W. (1892) Psychology. New York: Henry Holt and Company.

James, W. (1902) Varieties of Religious Experience. Cambridge, MA: Harvard University Press.

McAdams, D. (1990) Unity and purpose in human lives: The emergence of identity as a life story. In: A. Rabin, R. Zucker, R. Emmons & S. Frank (Eds.), Studying Persons and Lives. New York: Springer Publishing Company, pp. 148-200.

McAdams, D. (1995) What do we know when we know a person? Journal of Personality, 63, 365-395.

McAdams, D. (1997) The case for unity in the (post) modern self: A modest proposal. In: R. Ashmore & L. Jussim (Eds.), Self and Identity: Fundamental Issues. New York: Oxford University Press (pp. 46-78).

Todorov, T. (1988) Bakhtin: The Dialogical Principle (W. Godzich, trans.). Minneapolis: University of Minnesota Press.