Decentering Therapy: Textual Analysis of a Narrative Therapy Session†
We thank Dr. Patricia Bell-Scott and the anonymous reviewers for their insightful comments in the preparation and revision of this article.
Abstract
Postmodern models of therapy stress the participation of the clinician in a nonhierarchical, non-objectifying role, and highlight the therapist's embeddedness in the same processes of social construction as are the individual and the family. While much theory has been published in recent years, the actual conduct of a therapy session derived from the premises of postmodernism remains unclear. We investigated how a postmodern therapist manages talk in an actual session. We used textual analysis to examine a couples therapy session conducted by a prominent narrative therapist. Analysis of the talk led to descriptions of the couple's and therapist's agenda, and their interaction. The therapist's agenda is described in terms of “decentering” both the local unfolding narrative and its embeddedness in larger cultural stories. Five conversational practices: matching/self-disclosure, reciprocal editing, turn management to de-objectify, expansion questions, and reversals are examined. These practices inform the deployment of a decentering agenda in this specific text.
The “postmodern” movement in family therapy emphasizes an anti-objectivist epistemology, highlighting the fluidity and embeddedness of social interaction in the construction of meanings we assign to phenomena (Paré, 1995). Meanings are then viewed as restraining and organizing behavior. From this perspective, the role of the therapist shifts from a systems analyst who diagnoses a family dysfunction and then intervenes to correct it, to a participant-observer of family interaction, whose goals are to “perturb” meaning systems (Varela, 1989) and to expand families' “reality” (Anderson & Goolishian, 1988). Numerous dialogues in MFT journals compare the merits of instrumental, hierarchical and/or strategic roles to the “non-interventive” and participatory stance of postmodernism (see for example Atkinson, 1993; Duncan, 1992; Goldner, 1993; Goolishian & Anderson, 1992; Held, 1992; Nichols, 1993; Simon, 1993). Leading postmodern family therapists characterize their practice in novel ways. For example, Anderson and Goolishian (1988) described the role of the postmodern therapist as a “master conversationalist.” Hoffman (in Simon, 1992) contended that her role is like “a friendly editor.”White and Epston (1990) see their function as “liberating subjugated knowledges and life stories.”
While theoretical guidelines for therapy (Anderson & Goolishian, 1988; White & Epston, 1990), typologies of questions (White, 1993), and case examples (Friedman, 1993; Penn & Sheinberg, 1991) are available, little or no process research has examined the management of talk by a “postmodern” therapist. In addition, postmodern ideas offer many therapists a clinical position that integrates and accomplishes social critique (Hare-Mustin, 1994; Kogan, 1996; Weingarten, 1995; White, 1993). How egalitarian roles, non-objectifying interventions, and/or social critiques are accomplished in postmodern therapy remains unclear. What occurs in interaction with a “friendly editor,”“master conversationalist,” or a “liberator of subjugated stories?” How do these agendas and theoretical premises manifest in therapy talk, and what is unique about the production of discourse in postmodern therapy?
To explore these questions, we examined a couples therapy session by a prominent narrative therapist as a text. That is, we focused on how language and discourse functioned to create possibilities for meaning and interaction. This may be contrasted to examining a session for the motives, cognitions, or dynamics that language is representing. Drawing upon interpretive methods from conversation, discourse, and narrative analysis, we hoped to provide a useful account of how talk among the couple and the therapist created a context in accord with some of the premises of postmodernism. This article will provide a review of postmodern premises that inform both these recent developments in therapy theory as well as some of our analytic tools and questions. We then offer a detailed account of our methods in constructing our description and interpretations of the text. This description articulates a “structure” to the interactions in the session based on the notion of a textual agenda, a consistent pattern of effects a particular speaker has on the discourse. The therapist's agenda we describe as “decentering” in its effects on the unfolding context created with the couple. We then provide further details of the therapeutic agenda by discussing specific rhetorical practices by which a decentering agenda was accomplished.
POSTMODERN PREMISES
Originally a movement in the humanities, postmodernism incorporates the critiques of “modernity” from such diverse thinkers as Jacques Derrida, Michel Foucault, and Jean Francois Lyotard (Sarup, 1993). In common, they share an anti-objectivist, anti-foundationalist stance with regard to what is knowable in the world. Questioning the “grand narrative” that, by employing reason and scientific method, we may come to know or even more closely approximate an objective reality (Lyotard, 1979), postmodernism proposes that we live in a world of multiple realities (Hoffman, 1990). Accordingly, “reality,” as such, is context-dependent, socially constructed, and mediated or constituted primarily through language (Anderson & Goolishian, 1988; Flax, 1990; Gergen, 1985; Sparkes, 1992).
The emphasis on language as the medium of social construction has led to the use of discourse-based metaphors for understanding the social world (Bové, 1990; Lax, 1992; Sarup, 1993). Discourse refers to any related system of thoughts or ideas as manifested in language, be it written or oral, and the associated social practices that accompany that system of meanings. Providing a semiotic medium for reality construction to occur, various discourses compete, reinforce, overlap, include, and exclude others. From the play and interplay of discourse emerge social institutions and individual identity (Foucault, 1980). Discourse is not a narrow function of the social structure, but is created, maintained, remanufactured, and reshaped by social interaction at all levels of context.
The emphasis on construction and context leads to a radical departure from Enlightenment conceptions of the individual subject. Descartes proposed an abstract, disembodied self (“I think therefore I am”) that transcends the body. This self is also typically viewed as core, essential, and stable from context to context. Postmodern thinkers posit a “decentered” or a dialogical self (Bakhtin, 1986; Day & Tappan, 1996; Gergen, 1991). Rather than a unity that is influenced by various contexts, selves emerge from discourse. The self actively incorporates, reproduces, and evolves the contexts in which it participates (Sarup, 1993). Individual experience cannot be separated from the contexts and conversations that give experience meaning and direction (Gergen, 1985). Self is not a social product but a social accomplishment, consisting of culturally contingent micropractices for its production (Garfinkel, 1967; Gubrium & Holstein, 1995). Personal identity, be it gender, racial, sexual, based on age, or any other factor, is an activity, rather than a thing (Gergen, 1985).
Postmodern therapists deploy these ideas through the metaphors of story, narrative, conversation, myth, and text. Unfortunately, at times, clinicians use these metaphors in a form that reifies them as modernist conceptions or handy techniques rather than as an “epistemological stance” (Griffith & Griffith, 1992). Individual identity is not solely constructed by stories a client tells herself about herself. Her sense of being an individual with ideas, meanings, goals, and reactions is an effect of social interaction that emerges from and reproduces discourse. This does not imply isolation or dismiss the actual context of human behaviors around a person. A common criticism of postmodernism charges that this view means anything goes and all meanings are relative (Held, 1995). While meaning is relative in the largest sense, the person experiencing an identity does so in a social context of other meanings and the behaviors configured by them. Since discourse produces systems of behaviors and conditions and delimits what behavior may occur, they have quite concrete effects (see Foucault, 1979; for therapy examples, see Towns, 1993, and Hare-Mustin, 1994). For example, discourses regarding women as property or as objects of men can be implicated in violence toward women in this culture. A woman cannot simply be free of violence by telling herself a different story. She is compelled by the prevailing power relations in her context and must deal with the concrete effects of this discourse in her treatment by others. The same discourses are reproduced by these interactions and produce an individual's sense of who and what she is. The extent to which a person may resist participation in a particular discourse is negotiable to some degree, and can provide space for alternative actions and an alternative sense of identity (White, 1993; White & Epston, 1990).
Given that discourse produces and is produced by systems of power relations with “real” effects in individual's felt experience of their lives, we are engaged in ongoing social processes that objectify our experience in a form that we may act upon it (Gubrium & Holstein, 1993). We define and deploy “issues” at play in our interactions, and behave accordingly. We construct normality and deviance, usually unconscious of the construction process, leading to a sense of the “naturalness” of privileged discourses (Foucault, 1980). Individuals and institutions constituted through the interplay of knowledges and associated practices recreate and reinforce a sense of what is right, normal, or natural. A postmodern thinker may “problematize” or deconstruct these systems of meanings and behaviors. “Situating” phenomena in historical, macrosocial, and/or discursive context unmasks the “naturalness” effect of beliefs that condition behavior (Shumway, 1989). The epistemological conditions that enable specific knowledges to manifest becomes explicit.
Deconstruction or decentering is closely related to the postmodern skepticism of monovocal and universal truth claims. Postmodern critiques reject the idea that a singular meaning or essence can be derived from any phenomenon (Sarup, 1993). “Facts” never speak for themselves. The truth status or value of something is produced in relation to both local and cultural levels of social construction. Some knowledges (and knowledge-producing systems) are central and given more legitimacy. Harper (1994) noted that a personal sense of “decenteredness” has been common for marginalized social groups. Restrained from participating in the “grand narratives” of history, constructing alternate visions of the world, and questioning the naturalness of meaning constructs emerging from the cultural center is not new. The dominant European culture has commonly “decentered” the experience of African Americans. The notion of “centered” may also be associated with an individual participating in normative discourses rather than marginal ones. At one level centeredness is a myth, at another level this metaphor can draw out the implications of participation in particular narratives, and how individuals conceive of their position vis-à-vis what is “natural” or “objective.”
WHEN TEXT IS THE “FIELD”
A textual or discursive analysis differs from the typical qualitative studies becoming more prominent in MFT literature (for example, Metcalf & Thomas, 1994; Sells, Smith, Coe, et al., 1994; Smith, Sells, & Clevenger, 1994). Many of these studies are ethnographic — using interviews and field notes to gather data information on a local culture. The goal is to understand the participants (in MFT studies, the clients or the supervisees). The researcher analyzes recordings, “field” notes, and interview transcriptions for various themes and patterns of content that represent the experience of participants. The transcription is used as a record of what happened, and the language of the text is important to the extent that it corresponds to a phenomenological reality of the participants. In contrast, the present analysis treats the transcriptions of the session and the videotape as a particular example of discourse. We avoid making claims for the text as a description of the participants' internal states or personal motives. We are less interested in what this session “really” means, or what people “really” experienced, than in how meaning “got done” by the participants, especially the therapist. Our focus is on language and communicative action itself. The discourse is the field, rather than being an attempt to represent it.
We derived guidance for our inquiries in large part from conversation analysis (CA) as well as narrative- and discourse-based schools of inquiry. CA, a system for analyzing transcripted portions of spoken discourse (Sacks, Schegloff, & Jefferson, 1974) was an outgrowth of ethnomethodology (Garfinkel, 1967). In ethnomethodology, social actors are regarded as producing, manufacturing, and performing a social reality. Rather than looking at interaction as reflecting underlying principles or as organized by systemic contingencies, ethnomethodology and CA focus on the local management and production of discourse. Context is not a bucket, but a performance: we accomplish who we are and what we are on an ongoing basis through interpretive practices (Garfinkel, 1967; Heritage, 1984). The practices members use to construct a situation are not chaotic or without structure. However, this structure is organized and constituted within face-to-face interaction. Members' actions are reflexive— referring back to the immediate and proximate context, which is being sustained through the members' action (Maynard & Clayman, 1991). The job of the ethnomethodologist is to discern this order and to describe the activities that create and sustain it.
In conducting CA, an investigator examines how talk is used by participants to produce meaning. Early forms focused on everyday conversation, mapping out pauses, turn exchanges, and other paralinguistic features of talk (Heritage, 1984). Transcriptions attempted to code for the actual speech of the participants, with all the pauses, overlap, “umms,” breaths, and so forth included in the transcript. All features of talk are considered relevant and contribute to the production of meaning. The sequencing of talk from turn to turn is a key feature of the creation of a meaningful reality. Each utterance contextualizes and is contextualized by the previous ones. Thus, all communication is both “context-shaped” and “context-renewing” (Heritage, 1984, p. 242).
A traditional CA involves extraordinarily detailed transcriptions, focusing on a microlevel view of interaction that may neglect the functions of longer sequences of talk exchanges. In addition, this method has been criticized for being too reflexive, in that it can conceal the embedded nature of talk in interaction, and the contested and political nature of these interactions (Bologh, 1992; Pollner, 1991; Rogers, 1992). Heritage (1984) noted that, in ethnomethodology, actors appear to “suspend differences of perspective and interest,” reacting in a cooperative and harmonious manner to sustain a common intersubjective world. “The chessboard of meaning is never willfully overturned by the person in the losing position” (p. 73). Conflict, oppression, and differential social power arrangements are lost in the microscopic analysis where, from turn to turn, each actor is responding and participating in the unfolding of meaning. The embeddedness of local construction processes in larger sociocultural restraints is neglected. The approach becomes “a view from nowhere,” as if interaction has sprung into being de novo (Haraway, 1991, Ch. 9).
Like others (Potter & Wetherell, 1987), we have borrowed elements of CA that seemed most useful. Our transcripts are not as detailed as a traditional CA, though we do try to preserve much of the surface “disorganization” of the talk. As therapists, we also embraced the focus on the sequential, turn-to-turn production of meaning. To address the issue of a “view from nowhere,” we used notions from discourse and narrative-based inquiry (Ewick & Silbey, 1995; Potter & Wetherell, 1987; Riessman, 1993; Smith, 1990), which include elements of the postmodern theorizing. Dorothy Smith (1990) elaborated a sociological approach that analyzed social reality as “textually” produced. Smith rejected the overly reflexive elements of ethnomethodology, wishing to theorize more clearly how subjectivity is both locally produced and embedded in a social structure. She described how our everyday practice of referencing things as “here” or “there” or “now” is a socially structured activity. The spatial or temporal referencing of a situation describes a shifting center or “null point” to our shared subjective space. This has important ramifications for experience and how the possibilities for conceiving of action or meaning can be influenced by the social structure while still being locally constructed.
In Smith's (1990) textual analyses, the narrating voice of a story becomes a “null point” of reference from which procedures of accounting occur. Absorbed in a story and often unaware of the narrator's positioning moves to direct the reader, a person just follows the story. A textual analysis describes and interprets these moves. Another influence on our analysis is Michel Foucault, who is widely referenced in narrative and discourse analytic approaches. For Foucault (1980), the “center,” be it the narrating vantage point from which a story is being told or the normative process of a culture, represents singularity and hegemony. The center of a text is the story being told, the margins are the story left untold. The most common example perhaps is the telling of history from the vantage point of the group in power. The “center” of a culture is an abstraction. It indicates a directionality enforced by social process toward the “appropriate,” the “good,” and the “natural” state of affairs (Foucault, 1979). A postmodern analysis of a text may attempt to reveal these procedures: Whose story is being told? Are other perspectives being elided? How are characters and perspectives presented within the text, and how might that presentation lead to various constructions of meaning and interaction? Like Smith, Foucault (1980) noted that meaning, while produced locally, is embedded in “strategic envelopes” of possibility such that not every meaning has an equal chance of being produced. How a “character” can be positioned in a story or how a position will be interpreted and reacted to is not solely reflexive but, rather, embedded in a larger social order.
METHODOLOGY
The selection of theoretical lenses and material for analysis reflected several issues. One author (J.G.) had a background in CA, and had focused efforts on explicating “the seen but unnoticed practices” by which prominent therapists achieved effects (Gale & Newfield, 1992). This was coupled with the other author's (S.K.) interest in critical and postmodern theory, especially ways to relate the immediate interaction in the therapy room with larger political issues (Kogan, 1996). The choice of a prominent narrative therapist offered an ideal opportunity since this model treats interaction as both constructed and as political. To explore the performance of a postmodern therapist, a videotaped couples' therapy session was procured, conducted by Michael White (who is referred to as the therapist, or “TH”). White's work is clearly articulated theoretically and clinically as drawing from postmodern and social constructionist influences of, for example, Jerome Bruner, Michel Foucault, Clifford Geertz, among others (White & Epston, 1990; White, 1993). This particular session occurred at a national family counseling conference where White and two other “distinguished” therapists interviewed the same couple over a period of 3 days in front of a large audience. The couple, in therapy for the previous 2 years, were flown in to participate in the conference, and their present therapist moderated the event. This session was the first of the series, and commenced after a brief lecture about narrative therapy. Videotapes of the session were released for commercial sale.
We chose a textual versus an ethnographic approach for three reasons. First, our main theoretical interests involve describing communicative practices. Had the participants been available to interview, we would tend to see that data as “talk about the talk.” It would represent another opportunity to examine reality-constituting procedures, only in another context — with the researchers rather than in the original setting. The interviewees would be producing accounts about the therapy, which would be linked to the interview setting. Second, as a commercially available educational video, the participants would not be available to anyone else studying the session to learn about therapy. Nevertheless, a great deal of education and information dissemination can occur without one having access and opportunity to interview everyone. Textual analysis offers a means to examine critically any strand of discourse we encounter. And third, we wish to de-emphasize the specific personalities of the participants, especially that of the therapist. Rather than see our analysis as the discernment of expert techniques, we wish to present a way of thinking about therapy and language that encourages readers to do their own critical thinking about how to practice therapy.
Initially, the videotape was viewed in its entirety several times by the authors. Subsequently, three 12-minute segments of videotape were transcribed according to CA convention, from the opening, middle, and ending sections. Early transcriptions mapped the conversation at a “mild” depth, recording pauses over one second in length, marked changes in inflection, and nonverbal behavior around turn-taking issues (see Table). We examined and revised the transcribed text and videotape over the course of 3 months. We analyzed how the talk accomplished specific meanings among the participants via accounting procedures, turn exchanges, location and types of interpretive resources, and other paralinguistic features. This involved asking: “How does each conversational move function in the text? How does it affect the next utterance or recontextualize a prior utterance?” We closely studied exchanges that involved accounts of the problem, interruptions, overlaps, and apparent changes in style. We examined how these procedures related to specific narratives, using the notions of the “null point” and the center: “Who gets to be the narrator of a story? How is a subjective position negotiated versus being an object viewed from someone else's ‘null point?’” We also made attempts to locate the unfolding narrative process within a social structure that encourages some elaborations and productions of meaning and restricts others. This aspect of the interpretations is more speculative and requires a more global interpretation of the text than is offered by traditional CA.
Throughout this process we used and discarded various categories and schemes for describing the structure of interaction and particular conversational practices. According to CA convention, we attempted to ground our interpretations in the actual talk we were studying (Heritage, 1984; Psathas, 1995). We do not wish, however, to make exorbitant claims of objectivity or neutrality based on these procedures. The notion of “discovering” a “grounded theory” (Glaser & Strauss, 1967) by a systematic method of reading seems to us a means of concealing our own inevitable a priori interests and theories that are always guiding any inquiry. Although we engaged in many of the supposedly validity-producing procedures of qualitative research, journaling, reading against our interpretations, peer debriefing, we believe that our analysis of this text is both a product of discovery — we did not know what we would eventually say or believe about the text — and of mapping from our theoretical interests.1 For example, the metaphor of the “center” and “decentering” came to be a useful device to explicate the internal rules of the talk. It thus is both a category we “found” and related to our theoretical lens. Also, one of the conversational practices —“turn management to de-objectify”— refers to a general category of mapping in CA. However, we wished to discuss a specific use of turn management that occurred throughout the text.
In our analysis, each category or practice discussed is representative of a general theme or tactic that reoccurs throughout the text. We give examples of the text to demonstrate our claims. While we both have theoretical a priori beliefs that influence our sense making of the text, we attempted to keep our interpretations connected to the recorded talk. Again, we do not feel that our methods created a privileged explanation, and we hope that the presence of examples will allow readers to interpret for themselves and to contest our interpretations (Atkinson, Heath & Chenail, 1991).2
Analysis
We characterize the internally constituted rules that provide order to the text according to the notion of participant agendas. An agenda represents a pattern of what a speaker accomplishes in the interaction, and it may be unrelated to how a speaker accounts for her or his be havior. For instance someone may say, “I am not a racist but …” and then textually perform a racist context. Another person may claim to be a great listener and then proceed to interrupt constantly. Their effect on the context is not necessarily related to the content of their utterances. We interpreted the participants' agendas based not only on what their accounts said but also on how these accounts of things functioned in the discourse. Content of utterances or narratives is important, however, in that they are an attempt to move the story toward some effect. In addition, we attempt to look at some more global themes related to the social structure. These interpretations depart from the minimalist, functional view of CA and tend to be more global and thematic.
We found that the metaphor of decentering and demarginalizing effects within the unfolding text provided a structure to examine all three participant agendas. We use the term “practices” to describe rhetorical forms that accomplish particular agendas. In our analysis of the consultation, the decentering moves of the therapist draw from these various definitions.
***
The Couple's Agenda
Tom and Jane,3 a white, heterosexual, married couple had been in therapy for 2 years. Tom is in his forties and Jane is in her thirties. Both work full-time in professional careers. After a heart attack, Tom had originally presented for problems with anxiety. This had evolved into couples therapy. The couple noted that they were close to divorce at the time therapy commenced. They have one mutual child aged 5; Tom has several grown children that they raised together; and, from a previous relationship, Jane had a child who was killed in a car accident approximately 7 years prior to therapy.
As will be demonstrated, the couple sought to maintain changes and specific meaning systems developed with their current therapist. These meanings could be described as countering a “pursuer-distancer” cycle. Jane's role in assisting change had been defined within therapy as backing off and letting Tom be more in charge. Tom's role was to be more active and to “stand up” to Jane. They describe the turning point of their therapy as an episode where Tom began to feel he could disagree with Jane, and she began to respect him more since he stood up to her. Using the center/margin metaphor, Jane deploys her agenda by specifically decentering herself in the unfolding narrative. This includes taking on responsibility for the problem, supporting Tom's views, and actively holding herself back from determining too much of the unfolding narrative. Tom's agenda dovetails with Jane's: he seeks to augment his authorial presence in the relationship and in the therapy.
Exemplar 1 provides content and process information regarding the couple's agenda. Specific information regarding the therapist's management of these agendas will be explored in the following section. In Exemplar 1, Jane notes improvements in their life as a couple. Notice how Tom is centered in her account. She is learning from him; he is supporting her morals (lines 284–286). The therapist responds with a mutualizing metaphor: they are experiencing more of a “connection,” rather than one providing the experience for the other. As the therapist offers this new metaphor, Jane's backchannel comments (line 289) appear to indicate agreement. The summary, beginning in line 289, continues and reinforces this centering movement of Jane into an authorial position by using the second person pronoun repeatedly: “your part,”“you've been talking,”“two parts to your contribution.” These statements construct a clearing in the narrative for Jane to have an authorial presence by spotlighting her as protagonist. This engenders a turn transition (line 294): Jane breathes in, turns, and looks at Tom. She gives the turn to Tom who will reformulate the meanings that had begun to form. He gives an account of the couple's history where he has had to learn not to be afraid of Jane.
This exchange demonstrates the couple's agenda not only in the content of the account but also in the yielding of the turn. Jane (line 294) gives the turn to Tom who verbalizes a desire to bring the therapist “up to speed” and reformulate the version of the story issuing from the previous exchange. Rather than being more connected, which implies mutual change and responsibility, the new account centers on Tom's perception of his fears and the changes he made in feeling comfortable with being “angry” at Jane. In contrast, Jane's earlier comments were that she has learned to cue in on Tom and respect him.
Jane's act of giving up the turn, and Tom's act of receiving it, changes the narrator and influences who and what becomes central to the account. In both Jane's brief account and in Tom's longer one, Tom is the central actor. It is an account of his actions to regain control of couple's life. It demonstrates the struggle and direction of the couple's agenda, as performed throughout the talk in this text, to assist Tom's centeredness, and it manifests Tom's story as the active, struggling protagonist. Tom is achieving; Jane is restraining herself from being too present. Awkward and struggling speech patterns (e.g., lines 304–306), comparatively longer accounts, and detailed explanatory work constructs and reconstructs Tom's sense of effort in dealing with his conception of Jane. This work functions to construct a “dominating” Jane, as Tom performs his sense of being overwhelmed.
The story that the couple performs with the therapist does not emerge de novo, in the therapy room. The movement to centralize Tom in this couple mirrors dominant cultural stories about white, heterosexual relationships: women must be careful of being “too much” there, or present, and that the “natural position” of a woman is to be supportive. The above turn transition is representative of several in the text where Jane looks over at Tom, Tom reacts nonverbally, then Jane revises some aspect of her story. In contrast, Tom tends to look forward or at the therapist, while talking, and does not appear to see Jane's reaction to his talk.5 The links to possible cultural themes around gender in the local narrative may be interpreted from other exchanges:
The inertia of these cultural stories may inform Jane's careful and measured speech and willingness in the text to revise her story after glancing at Tom. To be part of a “team” requires her to move out of the authorial center. This type of marginalization is supported by the culture that frowns on a female partner being in charge of a heterosexual relationship.
This theme is demonstrated again in Exemplar 3 where Jane responds to an issue just raised by Tom, namely, that he did not feel there was enough romance in the relationship:
There are two unusually long (3 second) pauses (line 414 and line 621) where Jane appears to check in on Tom prior to responding to his discontent. After the first pause, she begins from a centered perspective, turning to Tom and requesting a clarification because her experience has been different. The end of the question however is softened. Tom refrains from answering and does not look at Jane. A long pause occurs, then Jane shifts, first agreeing with Tom's analysis —“naturally there was a time that there was no that the that the romance wasn't there” (line 621). Now, on second thought, just because “she feels it inside” (line 623) is not sufficient, and, according to the story formulated in the concurrent therapy, may even be problematic. She then defines her sense of disagreement as another example of her previous over-centeredness (line 624).
The Therapeutic Agenda
Hypotheses regarding the management of talk by the therapist crystallized around the themes of decentering and co-authoring practices. Consistent with a postmodern backdrop, we view the therapist's talk as constructing a sense of mutuality and multiple “centers” for understanding the presented content. The therapist attempts to privilege multiple subjectivities (“null points”) and to counter practices that centralize particular perspectives in ways that objectify or restrain narratives of agency of self or others. Thus, he was active in countering the agenda to marginalize Jane's perspective in the talk, as well as challenging the agenda to center Tom.
This decentering agenda includes the person of the therapist. That is, he attenuates the possibility of becoming a “centered” authorial presence as therapist. While concretely and directionally (from center to margin, and margin to center) affecting the text, the new story is not the therapist's, and his presence as an author is situated in a position among multiple perspectives. The therapist's talk functions to spotlight locations in the text, though the spotlighting action is executed with a minimum of centering of the therapist. In the structure and management of the talk, we identified five practices used to accomplish this agenda: (1) matching/self-disclosure, (2) reciprocal editing, (3) managing turns to de-objectify, (4) expansion questions, and (5) reversals. These practices exemplify ways in which talk was managed in order to effect the agendas discussed above. While not intended to provide an exhaustive inventory of possible practices for performing postmodern therapy, these practices illuminate forms of talk in this interview that achieved a decentering and demarginalizing agenda in both local and social narratives in therapy.
Using matching and self-disclosure: Matching and self-disclosure are common therapist behaviors in many therapy models. However, the therapist's use of these practices fits well within a decentering frame. In Exemplars 4 and 5, taken from the opening 5 minutes, the therapist's use of self establishes an ethic of participation. Rather than adopting the position of expert, with specialized knowledges that will illuminate the couple's situation, he emphasized common grounds on which they both participate. An important effect of this practice is a context for “subject-subject” relationships. The connotation of dealing with an expert is that the expert's “gaze” is more defining of the meaning of a situation. This gaze simultaneously “objectifies” others according to the standards of the gazer as well as providing a system of power relationships where a specific type of subjectivity is produced (Foucault, 1979). Hierarchy is manifest in the cultural field of the consultation where cultural stories define the knowledge status of counselor and therapist. This is especially problematic given the conference/theater setting of this interview. The therapist attenuates these somewhat through matching and self-disclosure practices.
The opening statement in line 13 begins with a statement that presents the therapist in a shared position in the narrative with the clients. He too knows how the “animals in the circus” must “feel,” implying a shared feeling in the current context. This is affirmed with an interpretive resource, “we're all in the same boat,” which enables the therapist to use the first-person plural pronoun, and to share with the clients a common position in the unfolding narrative. This occurs again in line 37 around expectations for warm weather. The therapist responds with personal comments about the weather rather than commenting on it impersonally (he does not say, “yes it sure is cold”). The management of the weather comment is typical of everyday social talk. This establishes a conversational frame of we are all visitors here in this place, making it new yet familiar territory for both clients and therapist.
This positioning establishes the therapist as an “outsider” to the conference and more like an insider to the couple perspective; he is in the same boat as the clients: he is a visitor from another country in a strange place. This may be contrasted to the hierarchical nature of their given roles: clients in therapy and a well-known consulting therapist (who received a flattering introduction by the couple's current therapist). Rather than reify the therapist-client hierarchy, the therapist responds in both segments from a personal perspective, not a professional one. A sense of “we-ness” or subject-subject relationship is deployed by “placing” the self or identity of the therapist in an equally privileged or disprivileged social space.
This practice, while a common one, is important in that it is part of a particular textual agenda. Matching and self-disclosure can occur without the agenda being a “decentering” one. A psychodynamic therapist may use a similar practice to establish rapport or make a particular point. A structural therapist may use a similar practice during a joining phase, and then proceed to establish a hierarchical relationship with a family. The practice itself, like the others we will discuss, must be considered in light of what type of context they tend to accomplish.
Reciprocal editing: Most therapists are familiar with reframing: an attempt to rearrange or alter the perspective on the apparent “facts” of a situation in order to engender a shift in attributions of meaning. Reframing identifies a technique used by a therapist to change the client's reality. Reciprocal editing, however, is a mutual process whereby the attributions evolving between all participants are open and fluid. In this session, it begins with the therapist offering a shift in meaning in response to a client account. This editing process is made reciprocal in the hesitancy and uncertainty of the delivery, and the use of a tag question to clarify or revise the therapist's revision. This creates an ongoing invitation for the clients to define and to evaluate the meanings attributed to accounts, experiences, and the unfolding story in session.
In each segment a revision or interpretation of the present narrative is suggested. “Responsible,” and “respect,” introduce subtle discriminations in the labeling process that may lead to alternative versions of the present story. The therapist's statements contain a variety of uncertainty maneuvers, including a rather long pause (line 322) and halting speech with repeats (line 337). He uses questions either at the end of the editing statement or throughout to make sure his interpretation of the previous talk is accurate to the client's sense of things. These maneuvers function to create space for the client to edit back, and they privilege or “center” the person's lived experience and expertise on the meanings in their own life. Rather than centering the therapist's interpretation and marginalizing the clients' perspective in the unfolding narrative, the therapist can be interventive while maintaining a less hierarchical (and thus more “decentered”) position.
In terms of the participant agendas, the first segment also addresses the marginalization of Jane and the possible cultural themes in the couple agenda. Culturally, the word “control” is especially loaded for females in the family context; they face the double bind of being assumed to have responsibility for family relationships, and yet they must not appear to be in a position of running things (Bograd, 1988). “Controlling,” as applied to females, is typically negative and associated with the culture's discomfort with female power. The therapist works to shift the meaning to one of “responsibility.” This term acknowledges the relational functions Jane likely was serving without the negative connotation. Responsibility also shifts the “meaning space” around Jane so as to provide validation for her efforts. Responsibility implies working hard for others, perhaps too hard, and is likely easier to move from that position than from having a problem with being “controlling.”
This movement from control to responsibility also represents a reversal as well as a form of decentering at both the levels of local and cultural story. Seeing women as “controlling” is representative of an androcentric (man-centered) perspective in constructing meaning. Women may internalize these definitions based on the central male view (Hartsock, 1983). In this exchange, the therapist's talk functions to challenge this cultural practice by shifting to a possible female perspective that is often marginalized. Jane first defined herself in relation to a centered male perspective (Tom's conception of her), as “controlling” rather than “responsible.” With the introduction of that frame, a clearing is constructed for a more indigenous female perspective. The reversal affects an androcentric cultural practice that is being produced at the local level by the immediate talk.
In Exemplar 7, the therapist again attempts to move the couple from their control vs. noncontrol perspective. Rather than presenting a metaphor that suggests either partner should be more or less central, he suggests that they are experiencing respect, and a co-privileging of perspectives. Neither partner's previous position is defined as marginal or problematic, and thus in need of movement or correction. This establishes the possibility of multiple centers of subjectivity versus a “fixed-sum game.” In addition, neither partner becomes the repository of a single enduring trait they must struggle against. The label “mutual respect” does not restrict either partner from being characterized as responsible or authoritative. These qualities are free to circulate in both partners.
Managing turns to de-objectify: The movement of conversation turns is a fundamental characteristic of talk. The act of being the speaker and having the floor influences the persistence and perspective of ideas and power relationships. The therapist generally tends to be nondirectional in allowing the couple to speak. Examples of deliberate moves to interrupt, shift, or take a turn for himself became apparant as we analyzed the text. In the following examples, the therapist manages turns (lines 139, 144) in order to give the floor to Jane, whose experience may be marginalized in Tom's account.
In Tom's account, the death of Jane's child (Tom's stepson) is embedded in that history as an explanatory element of Tom's frame of “how we got here.” Upon hearing the reference to Michael in the conversation, the therapist directs the turn away from Tom, giving it to Jane. He does not ask Tom for Michael's name, but turns to Jane, writing down Michael's name and the date of his death, and then looks at Jane again and nods to her. This action occurs within Tom's account, which he reestablishes in line 147. This movement by the therapist functions to highlight a “marginal” element in Tom's account, where the death of Michael is an interpretive resource to inform the frame of “problem history.” Michael is referred to as “her child,” without identity and without a proprietary qualifier of Tom's (my stepson). Being “her child,” Michael's death is situated in a more central position in Jane's possible story. The currently unspoken account may also include the extraordinary suffering of a parent who has survived the death of a child. The turn shift then functions to recenter Jane's experience as a Mother, which is marginal in Tom's current narrative. The validation provided by this move is strengthened by taking the time to write down Michael's name carefully (making somewhat of a show about getting it right) and the date of his death, the eye contact, and the nod.
Again we would note that all therapists manage turns in some fashion or another. The decision on how to do so creates a variety of outcomes. For this therapist, turn shifts that accomplished a de-objectification were consistent with other practices and the overall agenda. In the following example, the therapist again uses a turn shift to acknowledge a person who is being accounted for. In this instance, the discussion is about Al, the couples present therapist who is in the room moderating the panel.
This juncture provides a view of how the therapist manages the intersection of several factors of context: the conference, the current therapy, the couple's loyalty to the therapist, and the regular therapist's presence in the auditorium. The therapist's response (lines 163–165) about meeting Al in Atlanta achieves several effects. First it addresses and avoids a possible sense of elevation or objectification of himself. Al had evidently spoken with the couple about the opportunity to meet with the therapist whom Al admires (this was apparent in Al's introduction of him, for which the family was also present). The therapist, rather than overtly saying, “I'm no more special than you; we are equal,” reveals the time he first met Al, placing himself in a microcontext of ordinary person (not “distinguished” therapist) out in the world meeting new friends. He then crosses the “theatrical wall” at the conference, to include Al in the conversation. Rather than having Al be an object of a conversation he is witnessing, his presence, subject status, and sense of relationship with the couple are acknowledged and validated. Thus, this exchange not only attenuates the elevation of the therapist but also avoids the marginalizing of Al's presence in a “text” to which he is more witness than active participant. With these actions, the therapist invites Al's voice into the unfolding narrative.
Expansion questions: The therapist's position of centering client experience, or co-centering client and therapist experience, can also be seen in the use of expansion questions. Tomm (1987) describes these types of questions as “reflexive.” They attempt a context or perspective shift for the participant, and they function to expand the language resources restraining current definitions or perceptions of an issue. A story can be told from multiple perspectives and can lead to an infinite variety of links and turns that may affect the unfolding of meaning. In the following example, the therapist asks an expansion question that invites the couple to consider current changes, now relabeled as “respect,” and how these changes may be nested in their personal history.
This question represents a form of adjacency invitation. An “adjacency pair” is a conversational form whereby one speaker's action creates a specific structure within which the respondent is supposed to react. A common example involves a person extending a hand upon greeting another. The expected reaction is for the other to extend a hand to shake. In Exemplar 10, the therapist's adjacency invitation attempts to organize the clients' response into an identification of instances in their histories that support or contextualize current changes. It would also function to decenter the therapist from the present achievements. The changes achieved can be explained by indigenous resources in the client and her significant system. The question itself contains embedded within it an expanded or edited account of couple history garnered through earlier practices, especially reciprocal editing. The therapist will pursue this particular invitation over many turns. The narrative digresses and then is rewoven into the request for indigenous, historical perspectives into the conversation. In the following example, the historical perspective is that of Jane's deceased father.
Note how the therapist has maintained the expansion question begun in line 391 (Exemplar 10) through to the passage beginning in line 565. Not only is the invitation present to contextualize current changes through historical resources, but the apparent digression of the text between is recontextualized as part of the initial adjacency pair begun several minutes before. Note also the mutualizing wording in line 571 (“he'd be proud of both of you”), which contextualizes Jane and Tom as receiving validation from Jane's father, together in the same position, on the same team. Having had the adjacency pair completed, a new connection within the historical story is constructed. The couple is validated by a resource indigenous to their own life. This also has the effect of bringing Jane's father into the unfolding narrative, not only as a validator or source of recognition, but also as a co-author. The multiple authors of the unfolding story have gone beyond the original three participants in the conversation to include, at times, Al in the audience, a deceased father, and (later) more of the audience through the use of a reflecting team (Andersen, 1991).
Using reversals to decenter accounts: A reversal represents a practice that can powerfully expose marginalizing and centralizing effects in discourse. A reversal in poststructural theory represents an analysis that subverts or reverses a dominant narrative (Shumway, 1989). Finding an equal truth in an opposite account, or shifting to a marginalized perspective that undermines the current framing of an issue are forms of reversal. In the talk below, the therapist shifts to an objectified party's experience, and explores the issues from that perspective.
In the first segment, Jane has been discussing changes in her perceptions since being in therapy. Where we have seen Jane working hard to center Tom, we can also identify work in lines 374–383 to center Al, their therapist. Her successes are based on realizing the “truth” status of other's opinions (vs. her own). Again the agenda is a diminution of “too muchness” in regard to personal presence or control. Following her account, Tom makes a joke about Jane being a “Leo.” The therapist responds immediately by asking Tom what he is. The effects of this response are twofold. It equalizes Jane and Tom, given that his joke could easily be interpreted as coming from a center out to a Jane perspective, caricaturizing Jane and her experience. Through the question's reversal back to him, Tom experiences the same position vis-à-vis the narrative that he had just put Jane in by making his comment. Second, it achieves this without the therapist confronting Tom, and without centering the authorial presence of the therapist.
The concept of reversal as well as many other practices are elaborated in Exemplars 14–16. In Exemplar 14, Tom revisits the theme of desiring more romance. This excerpt follows Exemplar 3 (discussed above) where Jane begins to get angry that Tom doesn't recognize the current amount of romance, then changes her mind and decides that she has been in error at not recognizing Tom's perspective.
In this segment, Tom begins by setting an agenda whereby he feels that Jane should dress more attractively (according to Tom's perception) in order to rekindle romance. This reflects a social theme that tends to marginalize the experience of women. Smith (1990) discussed how discourses of femininity produce male-focused actions, whereas the discourses of masculinity also produces male-focused actions. That is, the gaze of masculinity organizes female behavior in ways that are not reciprocal. Rather than confronting Tom, which would entail separating himself from Tom (an objectifying and centering practice) and may encourage competition and hierarchy, the therapist employs a reversal. Tom's agenda comes from a “centered,” typically male perspective that neglects the influence of the perspective on an “other.” A reversal is constructed by shifting to the position of the “other,” thereby decentering the experience. Tom's agreement that he would dress for Jane as well enables a clearing for the therapist's comment focusing on mutuality of perspectives (line 637). The therapist continues the process of decentering from Tom's objectifying perspective by interrupting Tom and giving the turn to Jane.
Again, this is a rare instance of the therapist deliberately moving a turn, and thus pursuing an agenda of his own. The turn to Jane is an important part of the decentering movement begun with the reversal, otherwise the therapist would enter a space where he was talking for Jane rather than Jane's voice talking. Jane does concur that it is important for her to be “dressed for” as well. The narrative issuing from Tom's centered perspective that marginalized Jane's experience has been deconstructed, and Tom's accounting of the issue shifts:
In this sequence Tom mulls over this new vantage point or perspective on his own thoughts, even to the extent of using the third person regarding “a man” (line 673). This is a striking example of how this particular account has been deconstructed, not only at the immediate level where Tom is the authorial perspective to be dressed for; but also at the cultural level where dressing for men may be part of a cultural prescription for centering the experience of men.
DISCUSSION
According to an ethnomethodological perspective all social interactions have an organization that is created by the participants. Each member of the interaction is actively producing the context rather than obeying external rules. Through a microanalysis of the functions of talk, an interactional or textual structure can be inferred. This structure involved three agendas: Tom and Jane's management of the talk to center Tom textually, and the therapist's attempts to construct mutual and multiple centers of social construction. This includes addressing the possibility of a vantage point being over-centered such that “others” are seen as objects rather than subjects. The therapist also attempts to attenuate the possibility that his opinion becomes a centered vantage point that objectifies the clients.
This particular analysis offers a number of insights about how this therapist's take on postmodernism may be translated into practice. Rather than specific techniques, throughout the text the therapist is “decentering” on a number of levels. In the story that is being created with the participants (as if read like a text), “who” gets to be an author or a protagonist appears to be closely attended to. The therapist considers not just the content of an account, but the account as a political act. He makes efforts to distribute this authority when it congeals into a monovocal perspective. That is, the therapist attempts to make the meaning-making process democratic. The assessment of story-telling authority in the immediate discourse includes who is speaking, who is telling whose story, and how the act of accounting for reality may be affecting the shape of future discourse. Decentering practices require a delicate balancing act because these contributions to the discourse must include a reflexive element. The influence of the therapist in creating multiple centers and fluid perspectives is itself a political act. This postmodern therapist simultaneously attends to both decentering the client accounts as well as the overall construction of a therapeutic account in which the therapist is also an actor.6
The practices generated by a postmodern agenda are linked to cultural as well as local stories. The unfolding narratives constructed by people talking to one another, produce, maintain, and evolve the cultural stories or discourses that specify lives and interactions. The meaning of a story or of an account shifts according to who tells, where the telling occurs, when it is told, and how it is told. Particular meanings, however, cannot be created in any or all circumstances. The deconstruction or decentering of a local text requires consideration of the context within which this production is embedded, otherwise meanings simply arise from nowhere, and the production of meaning seems to be cooperative and harmonious. Accordingly, a well-conceived decentering movement can affect not only local perspectives and the emergence of marginalized ideas but also power relationships. The couple may be viewed as performing aspects of the culture, specifically hierarchical power relationships. The therapist does not assist Tom and Jane in liberating themselves from Jane's “dominance.” He nests this story in cultural discourse. The directionality this view provides can be seen in the choice of words for reciprocal editing and the deployment of reversals around culturally charged gender issues.
The practices discussed are not structurally unique to postmodern therapy. However, the therapeutic agenda contextualizes what these practices mean and what they accomplish. We feel that is not sufficient to make claims for what one is attempting to do in therapy. With the best of intentions, we may accomplish a variety of unanticipated effects in our talk with others. The notion of a textual agenda provides a means for evaluating therapy practices that consider the therapist in context. Intentions are seen as accounts that function in specific ways, and do not necessarily represent or correspond to an external reality. Haraway (1991) suggests a similar notion. She proposes that we treat knowledge as an agent, and evaluate it according to its effects — what it creates and restrains — rather than its internal consistency or truth status. This analysis similarly treats talk as an agent that must be evaluated by its consequences.
CONCLUSION
According to ethnomethodological theory, postmodern therapy is not and cannot be “non-interventive” or without an agenda (a claim not made by the therapist). Subjects continuously and actively produce their contexts. According to an analysis of this session, a postmodern therapy may differ from other models in its decentering agenda, rather than a “non-interventive” position. While we have noted five conversational practices that inform a postmodern therapy session, it is important not to view these practices as techniques for “doing” postmodern therapy. Techniques imply a centered therapist who is doing something to the clients. A postmodern intervention attempts to acknowledge and act from a participant status. The therapist is active, but activity is generated from an insider position to the interaction.
In our analysis and descriptions, we have relied upon methods of study from conversation, discourse, and narrative analysis. By their use we hope to have constructed an account that makes alternative and challenging readings of therapy talk available. This may be contrasted with attempts to reveal an essential dynamic of the therapy or of the participants' experience. In accord with post-positive research in both qualitative and quantitative efforts, we envision research as the production of knowledge that is always informed by theory, systems of method, and social context. Following Atkinson et al. (1991), we invite and embrace the social conversations that will assess the meanings, benefits, and dangers of this work.
