Hermans”s Dialogical Self: Meaning as Movement
Narrative Subject: Between Continuity and Transformation
Similarly to McAdams, Hermans addresses primarily two aspects of change within his model of dialogical self: a developmental aspect that he traces from birth to maturation in adult years, and a functional aspect, when the dialogical self is already in place but can modify itself depending on the context and events in a person”s life. In contrast to McAdams, however, Hermans”s model is not organised according to the progression through stages, and strives to accommodate not only movements towards coherence and continuity but also movements towards decentring and discontinuity. However, integration remains an ideal course of development in Hermans”s model as well.
Hermans and Hermans-Konopka address the developmental aspects of the dialogical self in their book Dialogical Self Theory: Positioning and Counter-Positioning in a Globalizing Society. They show that the emergence and development of a dialogical self is predicated not only on the acquisition of language, but also on a number of intersubjective interactional processes that start with the birth of an infant. In tracing these early predecessors of the later dialogical self, Hermans highlights the embodied nature of the dialogical self, the aspect that marks another significant difference from McAdams”s model.
Hermans and Hermans-Konopka locate the earliest social precursors of the dialogical self in the neonatal imitation behaviour that involves a variety of facial, hand and finger movements and vocalizations that replicate and tune in to the behaviour of an adult caregiver. However, they highlight that young infants are also capable of provocation, actively seeking and eliciting reactions from others. As such, the succession of imitation and provocation is a precursor of turn-taking behaviour and exchange that characterize later dialogical processes. Hermans and Hermans-Konopka see a clear sign of growing intersubjectivity in an infant in the socially elicited smiling in face-to-face interaction with social partners that emerges between the first and second months. As they note, “Smiling emerges as a sign of mutuality and signifies an affective turning to the other” (Hermans and Hermans-Konopka 2010: 202). Another crucial step in the development of a dialogical self takes place around 9 months when the child starts to perceive others as intentional. Not only does the child begin to construe others as intentional but also becomes able to perceive objects in the environment from the perspective of the other. This step is facilitated by the growing ability of the child to direct attention to the objects pointed out by a caregiver. In this process, gazing becomes a form of perspective taking. A particularly significant step in the development of the dialogical self takes place when the joint attention between caretaker and child becomes directed to the child itself. Consequently, a child becomes capable of self-reflective attention and forms an indirect perception of herself, that is, the perception through the perspective of a significant other. As Hermans and Hermans-Konopka conclude, “Along these lines the phenomenon of joint attention paves the way for inclusion of the other-in-the-self as a constitutive part of the self”s extension to the world” (Hermans and Hermans-Konopka 2010: 207).
Hermans traces further forms of non-verbal dialogue through giving and taking between mother and child in the first year of life. Hermans argues that in the processes of giving and taking of objects, the interaction between a significant other and a child has a clear dialogical structure, although the child has not yet mastered language. While the infant does not understand the words uttered by a parent, he or she reacts to the intonations and gestures of the caregiver. Even if this interaction is not yet sign-mediated, it represents a co-regulated activity, and as such works as a bodily foundation for dialogical activities. Hermans further links the ability for reversal of action that emerges at this stage with the Vygotskian account of interiorization and specifically its implications for self-regulated behaviour. According to Vygotsky, later in life the child has to take the position of the other towards herself in order to achieve control of her own behaviour.
During childhood years the extension of the self and emergence of new positions within it are further stimulated by role-play. Role-play generates multiple opportunities for exploring new positions and their inclusion in the self. With the development of linguistic capacities, the dialogical self receives a further boost. Caretakers routinely enter into conversations with young children that provide initial training in verbal dialogue. Such conversations are often characterized by a high number of questions that an adult asks a child, but they also provide a model for a child to learn how to ask questions of a conversational partner. As Hermans and Hermans-Konopka argue, it is an important stage in the emergence of a dialogical self: “there is somebody who is competent as a conversational partner and via whom the self is driven into dialogicality” (Hermans and Hermans-Konopka 2010: 211). Moreover, through these initial dialogical interactions the child becomes “increasingly aware of the other”s attention and intention as belonging to the other” (Hermans and Hermans-Konopka 2010: 211). Drawing on Vygotsky”s ideas about interiorization, Hermans and Hermans-Konopka suggest attention and intention function as central factors in organizing the dialogue first with a conversational partner, and later between the child and herself. Hermans concludes that, “The interiorization of the dialogue with the other, into the self, functions as a developmental basis for the dialogical relationship with the other-in-the-self” (Hermans and Hermans-Konopka 2010: 211).
The linguistic competencies of a child, role-play, family dynamics and socialization through kindergarten and school lead to the formation of various positions within the emergent dialogical self. But in order to account for and understand the organization of dialogical self from a developmental perspective Hermans and Hermans-Konopka postulate a specific type of position, promoter positions, that focus “particularly on the temporal aspects of the self and organize the self over longer time-frames” (Hermans and Hermans-Konopka 2010: 228). Drawing on the work of Valsiner, an important contributor to the development of the dialogical self theory, Hermans and Hermans-Konopka delineate the following characteristics of promoter positions. Promoter positions are open towards the future and they have a generative potential. They are able to produce new qualitatively different positions within the self. They facilitate the integration of new and already existing positions in the self, in such a way that a more adaptive self emerges. They occupy a central place in the position repertoire and have the potential to reorganize the self towards the higher level of development. They ensure the continuity of the self, while at the same time they are able to accommodate the discontinuity.
Promoter positions play a key role in the multi-level model for the development of the self that Hermans and Hermans-Konopka introduce to account for the temporal aspect of dialogical self functioning. Graphically, the model represents a cylinder that consists of a number of circles positioned on top of each other and representing different developmental levels of the self (Hermans and Hermans-Konopka 2010: 237). Hermans and Hermans-Konopka postulate three types of movements within the model: progressive, regressive and balanced. Progressive movements of the self entail moving up to a higher level of development, while regressive movements imply moving down to a lower level of development. The self can also be involved in balanced movements, when it moves within one and the same level. The development of self as a whole is not determined by age and does not entail movement through specific stages, but is prompted by important life events and the way the individual responds to these events. During the life course an individual may experience many upward or downward movements. The presence and availability of promoter positions are key in facilitating progressive movements. They can mitigate the impact of disorganizing events on the self, by making it possible to negotiate challenges and instigate the innovative dialogue with the potential new positions. Consequently, the movement towards a higher level of integration of self is stimulated. However, if promoter positions are not available, there is a risk that as a result of a disorganizing event the self will move down to a lower level of integration.
Furthermore, Hermans and Hermans-Konopka postulate centring and decentring movements on each level of self-organization. Centring movements go in the direction of the centre of the self, towards order and perfect integration. Decentring movements go away from the centre and undermine an existing organization and integration of the self. Each of these movements has positive and negative functions. Decentring movements engage with the increasing diversification of positions, and can potentially facilitate innovation of the self. However, they can lead to disorganization, chaos and fragmentation if they become overly dominant. Centring movements restore the organization of the self when an existing order is disrupted. However, if they become overly dominant, the self is at risk of becoming rigid, fossilized and inflexible. Hermans and Hermans-Konopka highlight that while centring movements are emphasized in the modern model of the self, and decentring movements are characteristic of the postmodern model, his dialogical self model is constructed in such a way that it can incorporate both types of movements and, therefore, negotiate between modern and postmodern perspectives.
Overall, Hermans”s model of the dialogical self appears to be considerably more open and dynamic than McAdams”s notion of identity as a life story. However, closer analysis of Hermans”s engagement with temporality, change and transformation reveals some limitations. In outlining these limitations it is instructive to examine foundational assumptions about the dynamic aspect of the dialogical self presented by Hermans and Kempen in The Dialogical Self: Meaning as Movement. Inspired by Bakhtin”s dialogism, the dialogical self model postulates that the I can move, in an imagined landscape, from the one to another position, from which different or even contrasting views of the world are possible. Moreover, like the voices in Dostoevsky”s novels, from different spatial positions, multiple Is may enter into dialogical relationship, agreeing or disagreeing with each other, which may lead to the emergence of meanings that are not given at only one of the available positions.
In adopting Bakhtin”s theorizing, Hermans and Kempen follow in particular Michael Holquist”s (1990) architectonic conceptualization of Bakhtin”s idea of polyphony, which accords great significance to spatial rather than temporal organization. Addressing the spatialization of temporal positions, Hermans and Kempen draw further analogies from the modernist novels of Gustave Flaubert, James Joyce, Virginia Woolf and Gertrude Stein. These literary texts allow one to move beyond the understanding of narrative as intrinsically temporal, organized along the stages of beginning, development and ending. Instead, if time and space are accorded equal importance, narrative can be construed as open-ended. Such a view is essentially in accord with Bakhtin”s thesis of “unfinalizability”, which Hermans and Kempen adopt as a guiding principle in their model:
Narrative, conceived as a multiplicity of I-positions, means, in fact, that each I, as an author, has its own story to tell. This implies that there is not a single and final ending. Rather, a complex narrative with ongoing dialogical relationships between several positions assumes an open process that resists not only a final unification, but also a final completion. (Hermans and Kempen 1993: 60)
As a general outline of a theoretical framework, this premise appears to be promising, particularly with regard to addressing the possibility of change and reinstating an open—rather than a closed—temporal dimension. However, some of this conceptual richness gets lost in the development of a working, operationalized model.
The first theoretical move that Hermans and Kempen make to circumscribe the Bakhtinian notion of dialogue is through aligning dialogue tightly and inseparably with the dimension of power, or dominance. Drawing on the perspective of child psychology, and more remotely on Mead”s social constructivism, Hermans and Kempen outline the following developmental progression: when children have reached the stage at which they are able to talk and think about themselves as I and Me, they have also reached a period in which society begins much more forcefully than before to influence and organize their world. Children start to be involved in dialogues, or rather, begin to be placed into various positions such as child, pupil or friend by various members of the community and family, parents, teachers and peers. In these positions children are addressed, not in a neutral way, but in ways that indicate approval or disapproval. In being able to address a child in a variety of positions, the community begins to impact on the organization of the self, making some of the I-positions more dominant than others. Hermans consequently argues that “dominance, as an intrinsic feature of dialogue, not only organizes but restricts the multiplicity of possible positions in the process of socialisation” (Hermans and Kempen 1993: 73). From this point, Hermans focuses on the operation of power in dialogue, delineating symmetrical relationships of turn-taking and asymmetrical relationships of dominance. As a result of such narrowing and qualifying of the notion of dialogue and tailoring it to the needs of social psychology, in its final elaboration the dialogical self is defined by Hermans and Kempen as follows:
In the concept of “dialogical self” the notions of intersubjective exchange and dominance as the main features of dialogical relationships are applied to the self, considered as a multiplicity of I-positions. This means that a more or less intensive exchange between positions is supposed, with explicit attention to the relative dominance between positions. (Hermans and Kempen 1993: 78)
It can be seen that at this point Hermans”s definition of dialogue departs significantly from Bakhtin”s position. Dialogue, just like story, can be operationalized in various ways: just as story can be reduced to a set of consecutive functions, dialogue can be reduced to turn-taking. But as Bakhtin reminds us:
The internal dialogism […] cannot fundamentally be dramatized or dramatically resolved (brought to an authentic end); it cannot ultimately be fitted into the frame of any manifest dialogue, into the frame of a mere conversation between persons; it is not ultimately divisible into verbal exchanges possessing precisely marked boundaries. (Bakhtin 1981: 326)
By contrast, in Hermans”s definition, dialogue is aligned with a stimulus—response, behaviouristic model, even machine-assisted interactions or computer interfaces. The complex and multi-layered original notion of dialogue, which, as Bakhtin insists, does not presuppose any resolution—even a dialectical one—is dissected and re-packed into a neat algorithm of an act-response sequence. The overriding Bakhtinian concern with freedom is substituted with a concern with dominance.
These limitations restrict both the notion of dialogue and the notion of self. Having delineated the structure of dialogue, Hermans sets out to “find in the realm of the self a basic structure that is similar to the basic structure that can be found for the notion of dialogue” (Hermans and Kempen 1993: 146). Drawing on previous research, Hermans and Kempen arrive at a description of self in terms of two basic motivational characteristics: one that strives for autonomy, and another that aims at interconnectedness. This allows Hermans and Kempen to postulate that, “The description of the self in terms of two basic motivational characteristics is highly similar to the two defining characteristics of the concept of dialogue that we have extensively discussed: dominance and intersubjective exchange” (Hermans and Kempen 1993: 147). Reciprocally, Hermans and Kempen assert the following isomorphism between structures of self and dialogue: “Self and dialogue have on a basic level two characteristics in common: (1) the separateness and autonomy of the self correspond with dominance in turn-taking behaviour; and (2) the openness and participation in the self correspond with the intersubjective exchange in dialogue” (Hermans and Kempen 1993: 147). Finally, Hermans and Kempen outline a research methodology to study the dialogical self, which can be formalized and operationalized in a structural fashion:
In sum, given the myriad of possible stories, which differ among people and also within the multivoiced mind of an individual person (each position has an own story to tell), it makes sense to distinguish two levels of functioning in the dialogical self: a manifest or surface level, and a latent or deeper level. The phenomenological variety of narratives, both between people and within the same individual, are on the manifest level. On a latent level of functioning, however, a limited number of basic forces or motives is supposed, influencing the content and organization of the stories on the manifest level. Note that the distinction between manifest and latent level applies not only to individual stories but also to collective stories (myths, fables, fairy tales, famous films, rumours in town, stories about economic recession, etc.). (Hermans and Kempen 1993: 148)
Similarly to McAdams, Hermans resorts here to a structuralist-style two-level narratological model grounded in the distinction between surface and depth. This has implications for the temporal dimension of Hermans”s dialogical self as well: as has been previously discussed, the assumption of a “basic” paradigmatic underlying structure that underpins various surface manifestations implies achronic temporality, temporality that is in principle closed. Starting with a valorization of Bakhtin”s concepts of unfinalizability, open-endedness and indeterminacy, Hermans finally arrives at a theoretical model that can hardly be reconciled with these categories.
To some extent this can be explained by the fact that Hermans draws on a particular aspect of Bakhtin”s work, which reflects Dostoevsky”s idiosyncratic manner of writing. As Morson (1994) observes, Bakhtin concentrated on a “highly intensified present”, allowing historicity and temporality to recede into the background. Hermans might be replicating the same tendency.
This tendency becomes particularly obvious when one looks at the large body of empirical research that was inspired by Hermans”s writing. A substantial proportion of these studies concentrate on the phenomenon of “I positioning”, exploring which positions the moving I can occupy in a dialogical self. Positioning was introduced as “a dynamic alternative to the more static concept of role” (Hermans et al. 1993: 217). Following Hermans”s emphasis on dominance and exchange, particular attention was paid to dominant or submissive, dependent or independent, comforting or threatening positions and shifts between them. These studies adopted Hermans”s methodology of the personal positioning repertoire, in which the participant nominates from an extensive list of dialogical positions (e.g. idealist, fearful, creative, vulnerable) those positions that are recognized as significant and which can then be rated and explored both quantitatively and qualitatively. This line of research is characterized largely by “the combination of an empirical conceptualization of positioning in terms of constructivist, binary scales, with the idea that positioning takes place in a space defined by a moral order” (Raggat 2006: 19). This research programme, therefore, focuses on the structural, atemporal organization of self. Within this context the issue of change and development is reformulated as the issue of a shift from one position to another. The development of personality is represented by a series of snapshots that remain disconnected from each other.
This goes against Bakhtin”s insistence on constantly developing, unresolved, open-ended character of human life—what Bakhtin claims Dostoevsky”s characters, as models of a truly dialogical and multivoiced approach to consciousness, embody: “inner unfinalizability, their capacity to outgrow, as it were, from within, and to render untrue any externalising and finalising definition of them” (Bakhtin 1973: 59). As Bakhtin further explains:
A man never coincides with himself. One cannot apply to him the formula of identity A=A. In Dostoevsky”s artistic thinking, the genuine life of the personality takes place at the point of non-coincidence between a man and himself, at his point of departure beyond the limits of all that he is as a material being, a being that can be spied on, defined, predicted apart from its own will, “at second hand”. (Bakhtin 1973: 59)
The issue of unfinalizability is particularly important if we want to draw some implications from the take on narrative in literary criticism to its appropriation in psychology. Since in the literary realm the model of completed and finished human life is a possibility in some genres, in psychology, which deals with a living, struggling human, such a model is problematic.
As with McAdams”s appropriation of the concept of story, Hermans”s attempt to transpose the idea of dialogue into the psychological domain demonstrates that such operations are accompanied by some loss of conceptual richness and explanatory potential.