Much of the traditional empirical research on therapy prioritizes the study of variables formulated at high levels of abstraction: Active elements of therapy process are conceptualized either as some aggregate qualities of the therapist (e.g., warmth, genuineness, congruence, and alike), attributes of complex therapists’ initiated strategies (e.g., the quality of in vivo experiences, the impact of the homework assignments, the value of interpretation or insight, etc.), or as the result of some underlying, but not directly observable, summed relational dynamics such as transference/countertransference, quality of empathy, and so on. The data collected under the influence of this paradigm are numerical or qualitative measures of the observed or reported occurrence of these variables “as such.” By “as such” we mean that the process of how these variables were realized is most often either assumed to be homogeneous across observations (therefore of little research interest) or the interactive dialectical process between therapist and client that actualize such data are treated as antecedents of the variable of interest and ignored. In addition, research based on this perspective tends to yield descriptions of the therapy process as unidirectional: the observations are organized from the therapist’s perspective. The interactive responsive nature of the clinical reality, the dynamic social interactions through which new meanings and insights are realized and new ways of being in the world are discovered are often overlooked, or parked at the margins. Even when relational issues are the focus of investigation, the active elements are frequently conceptualized as therapist activated. For instance, Rogers’ notion of the empathy places it in the Therapist Offered Facilitative Conditions (TOFC) framework (Rogers, 1957). In his theory, as well as most of the discourse on “humanistic” therapies, descriptions of the therapy process are focused on what therapists have to offer, his or her personal qualities (e.g., genuineness, presence, etc. (Geller & Greenberg, 2002), and how therapists overcome obstacles and challenges presented by clients, rather than the explication of therapy as a dynamically evolving interactive event.

On the one hand, using abstract, conceptually anchored variables in therapy research offers the obvious advantage of generalizability and replicability of the results. What is observed is assumed to be an [imperfect] actualization of an ideal concept of the variable, thus the results can be abstracted from much of the contextual elements in which it was generated. In this framework the data are summed over many instantiations and become more portable and generalizable in contrast to observations made closer to the phenomenological level—which are more grounded and delimited in the circumstances that generated them. In these research designs, deviations between the actual, observed, realizations and the abstract definition of the variables are either ignored or assigned to the “error term” statistically. Alternatively, the differences between the abstract/conceptual version of the event and what is observed in practice are accounted for as the quality of the variable in the particular instance (e.g., quality/appropriateness of interpretation, level of empathic response, treatment fidelity, etc.). The conceptual/abstract variable stands for an aggregate or class of events or qualities, and the relationships among such aggregate classes can be accurately evaluated statistically.

On the other hand, what is gained through the process of abstraction and categorization of what is essentially an interactive process, are purchased at the cost of limiting what researchers can discover about what makes therapy effective. As Stiles (1988) pointed out, even if we can identify therapeutically active and beneficial variables, such as an efficacious strategy, or a positive relational stance, studying these events “as such” has limited clinical utility: In therapy, examining a strategy apart from the specific context in which it is used will tell you only a part of its therapeutic potential. Increasing the dosage of a “good thing” does not necessarily produce better results (Stiles, Honos-Web, & Surko, 1998). It is, rather, a matter of appropriate responsivity,Footnote 1 doing the “right thing” (strategy) at the right time, responding according to the client’s needs and resources (Ribeiro et al., 2014; Stiles, 2011; Stiles et al., 1998; Stiles & Horvath, 2017). Research treating interventions or relational elements as “pre-packaged goods” without carefully examining how they are developed, shaped interactively, and made to do therapeutic work in the local context, misses a vital piece which is necessary to get a fuller understanding of how therapy works.

Our research program is designed to complement these more traditional approaches of research by focusing on the exploration of the interactional, conversational praxis of therapy. We start from the basic premise that all forms of psychotherapies, regardless of the particular theoretical framework that underpins the treatment are, at the core, discursive. Psychotherapy inevitably involves some kind of engagement and interaction between a client(s) and a help provider. And this engagement is essentially dialogical in nature involving the negotiation of shared meanings, common goals, and ways to make progress toward these common goals. From this perspective, the differences between treatments that are identified as discursive therapies (e.g., narrative (White, 2007) and those that usually are not labeled as such (e.g., psychoanalysis) refer to the theoretical assumptions with respect to the mechanisms of change but, in each case, the process of therapy, what actually happens during treatment is, universally, a series of discursive engagements.

We use Conversation Analysis (CA) as a preferred methodological framework for our investigations. CA treats discourse as a form of social (inter)action in terms of how participants organize their vocal/verbal and bodily conduct (Peräkylä, 2008; Sidnell & Stivers, 2013), rather than the means of communicating intents pre-formed in our brains. Discursive practices, including therapy, are organized through action sequences and turn-taking resources. It is by way of this sequential organization of our words and grammatical selections and the use of prosody and bodily movements that permit us to make sense of the other, negotiate meanings, create social relationships, and “get things done” interactively. Using this approach, we feel, has enabled us to look at therapy processes as a specialized discursive praxis that unfolds, turn by turn, as the participants strive to develop new meanings, position themselves differently in their relationships, and become mutually shaped through each other’s actions.

The kind of detailed examination of social interaction afforded by the CA approach is, by its nature, necessarily context dependent. To provide a coherent focus for our research program, we chose to concentrate on aspects of therapy that are recognized as common elements shared by diverse treatments. One of the most obvious such important shared features is that the client and therapist need to develop some level of consensus about the specific aims of the therapy and agree on the things that can be done in therapy to make progress toward these goals. These foundational aspects of therapy—along with the personal bonds that the participants form—are collectively labeled as the therapeutic or working alliance. This concept has its origin in the psychodynamic literature (Greenson, 1990; Horvath & Luborsky, 1993; Zetzel, 1956) but was formally proposed as a universal feature in every kind of treatment by Edward Bordin (1979). Bordin suggested that the acts of developing and sustaining the alliance throughout therapy itself generate a substantive contribution to the healing process. Over the last 45 years, the concept of the alliance has generated a great deal of research, yet relatively little work has been done to closely examine how such relational alignments are realized and repaired in various therapy contexts interactively.

Over the last 15 years, we have pursued a programmatic inquiry using the tools and conceptual resources of CA to take a closer look at how the alliance may be discursively and interactionally accomplished. To do so, we have used an already established discursive-interactional vocabulary to examine how relational alignments and affiliation are maintained, sometimes stressed or ruptured and repaired in clinical situations (Muntigl & Horvath, 2014a). Affiliation and alignment are terms we borrow from CA to capture different types of collaboration between interactants (Stivers, Mondada, & Steensig, 2011). Although these terms are not meant to be “equivalents” to terms used in alliance research, they do, in our opinion, partly explicate the alliance from a discursive-interactional viewpoint. Affiliation refers to practices that are pro-social (e.g., agreeing, complying, etc.) or empathic, whereas alignment refers to cooperative actions that work to get a conversational sequence or activity underway (e.g., actions that support someone’s discourse role as storyteller; complying with a request to perform an activity). For example, we have examined how some therapeutic practices, such as making a verbal note of a client’s non-vocal conduct (e.g., facial expression, gesture, or bodily movement) and raising it as a salient topic of discussion, can work to bring about closer affiliation between therapists and clients, but also can create dramatic shifts in alignment toward the pursuit of other therapeutic-driven business (Muntigl & Horvath, 2014b). We have also explored how head movements such as nods work to re-equilibrate conversations, repair disaffiliation, and move the participants toward closer affiliation (Muntigl, Knight, Horvath, & Watkins, 2012). For this chapter, we examine a session of family therapy to illustrate a selected set of discursive processes through which stresses in the alliance may occur and how these stresses are subsequently ameliorated. In particular, we show how a therapist and family members work together and orient to and negotiate the alliance with the therapist and also re-negotiate relational allegiances between family members through sequential units of conversation.

Methods

Our research draws on data from therapy sessions with real clients. To illustrate our work, in this chapter we will analyze excerpts from a single session conducted by a master therapist (Dr. S. Minuchin).Footnote 2 The session was part of teaching the material Dr. Minuchin prepared for the workshop offered to practicing family therapists. Clients who were receiving service from the institution hosting the workshop were recruited and volunteered to receive a consultation from an “expert therapist.” Informed consent to video-tape the consultation and to use the material for training and research purposes was obtained from the client and also from the client’s regular therapist who was present at the consultation session.

The client “Suzanne”Footnote 3 was a single parent in her early 30s. She had two children: a boy 6 years of age (Kenny) and a girl “Marcy” who was 12 at the time of the interview. The boy was in a special day program (for behavioral difficulties) provided by the host agency. The mother has had long-term substance abuse problems and has been on social assistance for much of her life. She has stopped abusing substances for over a year and has been receiving family therapy (with the children) from a therapist provided by the host agency. The plan was for the family to come in as usual for their weekly appointment, but would receive a consultation from Dr. Minuchin instead of their regular session. Suzanne came with her daughter, but left her little boy with a neighbor who had a boy of similar age.

Present in the excerpts: “Suzanne” (adult client), her daughter “Marcy,” “Jenny” the client’s regular therapist (a psychologist with over 25 years of experience), and Dr. Minuchin. Initial seating arrangements in the consultation room were: Jenny on the left side of the room on a chair, Marcy and Dr. Minuchin on a couch in the center, Suzanne on the right side of the room sitting on a couch close to Minuchin, and directly opposite Jenny. The session was video recorded and subsequently transcribed using CA transcription conventions, including prosodic elements and significant non-verbal conduct (Hepburn & Bolden, 2013). Some of the transcribed excerpts are reproduced in this chapter, but we used both the video tapes and the transcriptions in our analysis.

Analysis

For our analysis, we have selected three interactional contexts that we feel are exemplary in demonstrating how alliances are forged in family therapy: (1) diffusing tension and building alliance, (2) strengthening “within family” alliancesFootnote 4, and (3) balancing support versus autonomy. For the first, we show the interactional practices through which the alliance between a therapist and client becomes locally ruptured and then repaired. In the second, we examine how a family therapist discursively works to build a productive allegiance between a mother and daughter, one in which both have a mutual understanding and agreement on family structural relations. And, finally, in the third, we explore how alliance building may run counter to constructing autonomy and authority and how a family therapist works with the clients to balance these therapeutic requirements through specific discursive practices.

Diffusing Tension and Building Alliance

When a family therapist’s utterance disaffiliates with a client (or vice versa),Footnote 5 it may create tension between these persons and may, momentarily, negatively impact on the degree of social rapport between them. Further, utterances may only indirectly work to disaffiliate, alongside the more central action that the utterance is performing; that is, although an utterance may appear to be functioning primarily as a question, there may also be other kinds of discursive work (e.g., blaming or complaining) generated by the action being carried out. This kind of multi-functionality of utterances may call upon respondents to deal with a number of different issues at the same time. For example, a therapist may need to deal with the tension brought about by the disaffiliation at different levels, such as between therapist-client but perhaps also between family members or even between therapists—if more than one therapist is present in the room. In Extract 1, we show how a therapist’s utterance creates tension at these different levels, and, further, we show how Minuchin works at re-building relationships between the different participants including the second therapist present.

Extract 1

figure a

This extract begins with Jenny, the client’s regular therapist, directing a question to the mother, Suzanne (lines 01–03). Through this question, she makes the son’s (Kenny) absence a salient topic to be explored in the course of ensuing talk. Jenny’s utterance also conveys other actions that create a line of disaffiliation between the participants. To begin, there is an aspect of blame directed at the mother. There was a prior understanding that the mother would bring both children to the session, but she arrived only with the daughter. Thus, Suzanne may be viewed as having failed to honor this agreement. Further, Kenny’s absence may be seen as reflecting poorly on the client’s regular therapist Jenny because she did not deliver the family she promised to have there for Minuchin to work with. Thus, by making it clear that there was an agreement that Travis would be present, Jenny’s question may be functioning to defend herself and save face vis-à-vis Minuchin. We may also note that the son’s absence is something that is relevantly reportable, in the first instance, by the parent. By orienting to the institutional relevance of the situation in this way (Heritage, 2004), Jenny’s talk implies that Suzanne is obligated to inform the other therapists and, by not doing so, the mother’s inaction may be viewed as a “misconduct” (Drew, 1998). The blame component of Jenny’s talk also performs additional discursive work that promotes further disaffiliation: It puts added pressure on the mother to provide a satisfactory explanation of the son’s absence. Various turn features illustrate this: “<le:tting> me know” implies that Suzanne is accountable to Jenny; the juxtapositioning of “Kenny was planning to come” with “he’s not here” in lines 02–05 creates an implication that Kenny was intending to join them and, thus, his absence constitutes a form of breach.

Jenny’s request for information strongly solicits a response or answer (Stivers & Rossano, 2010); that is, it sets up the expectation that Suzanne will respond to the relevant features of the prior action that positioned her as accountable for having committed a breach. Common responses to blame include denial, admission, account, or counter-blame (Buttny, 1993), but, as Buttny has pointed out, response options may be considerably expanded in multi-party participation frameworks. The long delay or silence in line 06 seems to index Suzanne’s difficulty in responding, and her confirmatory “yea:h.” (line 07) merely acknowledges, but does not answer, Jenny’s question. Moreover, her extended response in lines 09–11 seems to stray further from the topic of the son’s absence. Although she begins her turn with what might appear as an account in progress (“um I was-”), she quickly self-repairs (Schegloff, Jefferson, & Sacks, 1977) by first drawing attention to the necessity of disclosing something to the therapists (“I have to tell you”) and then by using her current emotional state (i.e., her anxiety about coming to see Minuchin) as a discursive resource (and excuse) to shift topic. As well, note that as the mother completes her turn, she directs her gaze at Minuchin, thus making an appeal for Minuchin to take up the next turn. In this way, Suzanne may not only be relieved of the burden of having to answer and account for a certain “misconduct,” but it gives another conversational participant a chance to mitigate the mounting anxiety and relationship stress brought on by (1) Jenny’s disaffiliative request for an account concerning Kenny’s absence and (2) Suzanne’s disaffiliative response in which she fails to answer and, instead, implicitly beseeches Minuchin to take up a conversational turn at talk. In this way, Suzanne’s utterance functions to disalign with Jenny’s initial interactional project of uncovering the “reasons” behind Travis’ absence and instead opens up a new line of activity that solicits Minuchin’s participation.

Minuchin subsequently complies with Suzanne’s implicit non-verbal request by responding to her disclosure of feeling nervous. He does so by first asserting his reciprocating feelings of nervousness (“so am ↑I:.”) and, thereafter, by validating their shared emotion (“so that’s good. w- we are nervous to↑gether.”). Minuchin’s response momentarily shifts the topic to “nervousness” and away from “Kenny’s absence,” thus creating mutual affiliation/alliance around the topic of their shared nervousness. In line 16, Minuchin also shakes hands with Suzanne. Suzanne responds with affiliative laughter (line 17) that claims acceptance of the positive stance realized in Minuchin’s prior turn (Schegloff, 2007). Minuchin’s attempts at repairing the alliance appear to have been met with some success. In this extract, it is shown how Minuchin uses a disaffiliative episode as an opportunity to form a therapeutic system that helps to build a positive alliance. In Structural Family Therapy (SFT) terms, Minuchin accommodates to the family by professing to also be nervous and works to “join” the family by not only empathizing with but also sharing their “distress.”

Later in the conversation, line 18, Suzanne answers Jenny’s initial question and thus explicitly orients to issues of blame and accountability by accounting for Kenny’s absence. In doing so, she offers confirmation that she breached her parental obligations, but at the same time justifies her conduct by claiming inability (“I don’t think I woulda been able to °get him here today”), referencing her anxiety (“woke up in a really anxious mood”) and displaying uncertainty (“>I didn’t know which way te go this morning so,<”). These displays of uncertainty, uneasiness, and low agency have potential negative implications that create a self-deprecating stance: She is not a good mother and the “proof” is that she was unable to bring Kenny to therapy. Research in CA on everyday conversation has shown that self-deprecations are generally followed by two different response types: The first is affiliative, working to strengthen social relations, and consists of disagreeing with the self-deprecation or offering praise, whereas the second is disaffiliative and consists of agreeing with other’s negative assessment of self (Pomerantz, 1984).

In his response from lines 25–44, Minuchin performs detailed moral and affiliative work by drawing from a range of interactional resources that orient toward the alliance between himself and Jenny, but also between the therapists (Minuchin and Jenny) and Suzanne. Beginning in line 25, Minuchin directs his talk to the other therapist (“you ↑know Jenny…”), thus momentarily changing the participation framework of the interaction by orienting both physically and verbally to Jenny and away from Suzanne and by making Suzanne a type of by-stander to the conversation (Goffman, 1981), someone who is no longer a ratified participant in the conversation and can only listen. His subsequent formulation of “I think that … actually,” lines 25–26, seems to foretell an upcoming difference of opinion; in fact, the sequential placement of “actually” implies that Minuchin is about to provide a different account of the events surrounding Travis’ absence. This shift in the participation framework performs alliance building work in two important ways: First, it allows the therapists to momentarily “join in alliance or coalition” with Jenny (Minuchin, 1974, p. 148), thus repairing the relational damage created by “interrupting” Jenny’s original line of inquiry and by recognizing and promoting Jenny as the relevant person to affirm the new framework presented. Second, as will be shown later in the sequence, it positions the therapists as being “in agreement” as to how Suzanne positively dealt with the situation, thus establishing re-affiliation with the mother.

In line 26, Minuchin briefly returns to the “original” participation framework in which Suzanne is a ratified participant by requesting permission to use her first name (“may I call you by your first name?”). Through this move, he works to re-establish an alliance with Suzanne. Subsequently, in line 30 onward, he again resumes his coalition with Jenny by making eye contact with and directing his talk at her. While doing so, he now completes his alternative interpretation of events, as indexed by the term “actually,” and then does affiliative work by praising Suzanne’s decision not to bring Kenny (“>I ↑think< (0.9) actually Suzanne, (.) did something very nice. …”). Thereafter, he orients to and highlights Suzanne’s authoritative role of mother/caregiver (“>↑she decided< what is, (0.7) good for Kenny.”; “you know, (0.3) instead (.) of responding, (0.7) to your needs or my needs, (0.9) ↑she responded to ↑Kenny’s needs.=). Note also that Minuchin emphasizes and strengthens his positive assessment through prosodic resources of stress and rising pitch and that Jenny weakly affiliates by offering her acknowledgment (lines 15 and 19). Thus, Minuchin’s discursive practices provide Suzanne with an opportunity to re-consider her position vis-à-vis Minuchin’s and to consider accepting the agentive and authoritative position that Minuchin is offering her. This strategy also allows Minuchin to perform a face-threatening act (i.e., disagreement with Jenny) in a face-saving and thus affiliative manner: He is actually telling another therapist about how positive the mother’s actions were, contrary to what the mother (Suzanne) may have thought. Also, by forming a coalition of alliance between himself and Jenny through the altered participation framework, Minuchin not only helps to smooth over what might be considered a difference in views between the therapists but also gives Jenny the opportunity to make a display of shifting perspectives and thus affiliating with this new agentive position for the mother. As can be recalled, from Jenny’s initial perspective, Suzanne had undergone a breach by not bringing Travis. But from Minuchin’s viewpoint, Suzanne was demonstrating authority, an action that deserves praise.

Strengthening “Within Family” Allegiances

One of the initial contexts for family therapy involves a disalignment in perspectives regarding family role relationships. Whereas the parents may show difficulty in adopting certain age and stage-appropriate “executive” roles vis-à-vis their children, the children, in turn, may be uncertain about which roles they may assume and how these may work to complement parental roles. To achieve good allegiances between family members, it becomes necessary that clarity and endorsement of respective roles be realized. One of the main therapeutic objectives in SFT is achieving mutual understanding and agreement on family roles. Consider Extract 2, in which Minuchin initiates a dialogue with the mother and daughter to gain consensus on role relationships and tasks. Just previously, Minuchin wondered whether the family situation sometimes led Marcy to conduct herself in ways more typical of an 18 years old (i.e., a grown-up) and, moreover, that Marcy’s resistance to parental authority may be related to that. This was then followed by Suzanne reflecting on the difficult life she had growing up and if that bore any relation to Marcy’s life and her being 12 years old.

Extract 2: [14:55]

figure b

At the beginning of this extract, Suzanne makes numerous displays of uncertainty (“I don’t know.”; “↑I’m not making any sense.”) and non-verbal displays of discomfort such as shifting around in her seat and placing her hands to her head. From the daughter’s perspective, this may be seen as a confirmation of “weakness” in which the mother appears uncertain, as someone who is unable to confidently appropriate a position of epistemic authority. It is likely for this reason that Minuchin immediately aligns with the mother by showing strong endorsement of the mother’s epistemic status and her ability to make perfect sense to others (“you make a lot of sense to me:.”). Minuchin also strongly affiliates with Suzanne by using intensifiers such as “rilly” and “a lot of” and by touching her arm during a brief moment of distress and uncertainty. He then repeats his endorsement of Suzanne’s ability to make sense and then asks Marcy whether she is of the same opinion. Through this move, Minuchin works to gain Marcy’s endorsement of her mother as someone with entitlements to knowledge and experience and as someone whose talk is transparent and logical. When Marcy voices agreement in line 12, Minuchin then in lines 14–15 leverages her response in order to draw Marcy’s attention to the implied paradox of having parental authority, while at the same time not having rights to make rules for your children. Minuchin casts this viewpoint from the mother’s perspective (“but ↑mo:m doesn’t feel frequently that…”), which works to downgrade his own epistemic position (i.e., it is the mother that claims this, not Minuchin), but he also cedes epistemic rights to Marcy by giving her the opportunity to respond to the mother’s feelings (“is that true?”). Minuchin’s use of “frequently”—see also line 23—also works as a shield against the possible inference that Minuchin is being critical of Suzanne’s performance as mother; that is, although Suzanne frequently does not enact her parental authority, she does at times do so. In this way, Minuchin works to create a balance between having to discuss the mother’s difficulty in adopting a parental role with Marcy, while at the same time not leaving the impression that the mother is unable to exert authority. By way of response, Marcy makes partial concessions to the mother’s viewpoint (“I think she has the right to make rules for me,=>but…”), but then continues by voicing disagreement about the content of the rules (“I jus< don’t like the rules she makes.”). Thus, by first forming a strong alignment with the mother to support her role as someone who has legitimate parental authority and then getting Marcy to affiliate with this position, Minuchin has taken first steps in achieving a new allegiance between the family members. This is one in which the mother’s actions begin to “make sense” from the reference point of the daughter.

Later, in lines 22–28, Minuchin again tries to solicit the daughter’s agreement on what appears to be a dysfunctional structural hierarchy between the parent and child. He prefaces his turn with an evidential expression (“what I: ↑ (0.5) hea:r”) that displays his lesser access and knowledge about what Marcy’s mother is actually feeling. Thus, the ensuing claim that the mother needs to apologize to Marcy when acting like a parental authority is put on record as based on hearsay and is thus presented as a possibility that seeks confirmation from Marcy. Note that Minuchin uses the term “frequently,” which again—as in line 14—works subtly to endorse the mother’s parental role as someone who is able to enact authority. Also, Minuchin’s choice of the word “need” as in (“she needs (.) to, ↑apologize…”) further suggests that the mother may be facing certain “familial” obstacles preventing her from enacting control (e.g., she may feel she needs to apologize in order to maintain an affiliative mother-daughter relationship) and, by implication, changing the present mother-daughter patterns of communication may facilitate an improved way of relating. This epistemically downgraded turn design allows Marcy to not only confirm the veracity of this claim, but to also reflect and elaborate on the reasons why this may be so. Although Marcy does not immediately respond, she does, after a couple of confirmation-seeking prompts, provide verbal and non-verbal affiliation. After having gained Marcy’s confirmation, Minuchin proceeds to probe into the reasons why the mother may have trouble taking up authority by asking for Marcy’s opinion (“an what do you think.”). What follows is a concession in which Marcy takes some responsibility for the family dynamics (lines 34–37); that is, she concedes that she sometimes may be at fault (rather than her mother, thus absolving the mother of the need to apologize) but is unable to apologize to her mother. With this concession, Minuchin has managed to move the family a bit closer to achieving mutual understanding on the problem and to forming the preconditions for forging a different allegiance in which the mother and daughter may begin to construct their relationship in a more productive way.

From line 38 onward, Minuchin begins to summarize the family’s dilemma, framing it as an issue of joint responsibility reinforced by the repeated use of “both of you” (lines 38–39): Marcy needs a parental figure who does not apologize for exercising her entitled authority, and Suzanne does not feel entitled to take up this authoritative position. Then, in line 45, Minuchin frames the dilemma as a puzzle (“it’s- it’s an interesting thing”), which does a range of epistemic work: It implies that Minuchin does not have special insight into why this problem occurs, and consequently, he declines responsibility to deliver a solution; it suggests that further reflection and exploration may allow the family to “solve the puzzle” and create a more positive relationship; and it operates as a fishing device (Pomerantz, 1980), because it targets the family members’ personal epistemic domains to which they may display greater rights and access.

Balancing Support Versus Autonomy

It is not uncommon for clients to experience and display emotional distress during a session. These displays have therapeutic significance not only because they provide access into the client’s emotional experiences, but because they provide opportunities for affiliation and empathy and thus the potential for forging a strong alliance between the participants. There are cases in family therapy, however, where displays of upset may create certain therapeutic dilemmas: On the one hand, distress calls for support from the group, but this support may also undermine the upset person’s autonomy, thus making them appear “needy” and unable to take control of the situation. In Extract 3, it is shown how Minuchin is able to balance the needs of support versus autonomy in a context where Suzanne becomes emotionally distressed. Although Minuchin offers Suzanne support, he is careful not to impinge on her autonomy, making her look weak in the eyes of her daughter, and to resist attempts by others to undercut the mother’s authority.

Extract 3: [17:00]

figure c

At the beginning of this extract, Suzanne is discussing the difficulties she is having with Marcy (lines 01–07). She finds it hard “to put my foot down” and that “it is such a ↑fi:ght.” She then, in line 07, reveals her emotional state in these situations as one of overwhelming fear. Following a prompt by Minuchin in line 10, in which he attempts to elicit more talk about the fear (for elicitation practices, see Muntigl & Hadic Zabala, 2008), Suzanne complies by naming “the ↑fear of reJEC↑tion” and then begins to verbalize what Marcy may say to her in these contexts but cuts it off (line 12). She does, however, provide Minuchin with non-verbal access of what Marcy bodily performs through a reenactment (Sidnell, 2006), in which Suzanne mimics the facial and head movements during which Marcy may be rejecting her. After having conveyed her emotional assessment of typical situations involving her daughter, Minuchin proceeds to utter a formulation (Antaki, 2008), that provides the upshot of Suzanne’s fear of rejection, “so you- you need her to love you euh?”, and that underscores how important it is for the mother and daughter to maintain a close, affectionate relationship. It is at this point that Suzanne begins to shows signs of distress: She produces a pronounced sigh (line 16), then begins to sniff (line 23), which is followed in line 26 by an apology that is delivered in a tremulous voice (Hepburn & Potter, 2007) and, finally, another apology with tremulous voice followed by her moving off camera (line 33). During this time, Minuchin repeatedly points out the significance of Suzanne’s emotional self-disclosure to Marcy—“you’re very important to your mom.” (line 20); “it is very important that you show, love an ( ) an approve.” (lines 22 & 25)—and later, in line 30, attempts to solicit her feelings concerning the important role she plays in her mom’s life.

Still showing signs of deep distress, Suzanne concedes in line 34 that her prior self-disclosure was difficult and, in would seem, painful (“°snih° ~it was really ↑ha:rd for me to say that.~”). Jenny then produces an overlapping affiliative move of offering Suzanne some tissues (line 35), which is followed by Minuchin’s attempt to get Suzanne to return to her seat (line 37) and Jenny’s overlapping utterance that repeats her offer, but this time in full. Following this, in line 39, Marcy reaches for the tissue box in an attempt to hand them over to Suzanne. Minuchin is then quick to reject and block Marcy’s offer, demanding that Suzanne take the tissues herself (“let her take it. (0.4) °let her take it.°”). From a Structural Family Therapy perspective, the unfolding sequence is an “enactment” of the family’s problematic tendency to reverse mother/daughter roles under stressful conditions: In offering the tissue box to her mother, Suzanne is placed in a role of dependency in which her daughter provides for her. This role relationship is what Minuchin has been trying to reverse through his interactions with the family and that would explain why he resists Marcy’s attempt so vigorously. In doing so, Minuchin implies that Suzanne is able to fend for herself, that she can recover from distressing situations. Thus, although Marcy’s gesture on the surface does have an affiliative quality, it does so at the expense of undermining Suzanne’s position as an agentic and autonomous parental authority.

Conclusions

The analyses above provide examples of the ways we approach analyzing therapy process using CA. Examining therapy as an interactive phenomena unfolding in language can help to illuminate the discursive elements that dynamically realize relational structures in therapy. For example, the analyses allow us to follow, turn by turn, how a potentially disruptive interaction can be worked through to establish a pivotal alignment with the parent and, at the same time, challenge her traditional self-critical defensive response and generate creative novel alternatives to explore. CA methods enable us to identify the specific conversational resources used and the way these resources are sequenced and timed to create an opportunity in the discourse for the client to move from the position of “failure” to “success,” from powerlessness to agentic identity. The first excerpt also provides an opportunity to observe how the use of timing, phrasing, emphases, and direction of turns in the conversation can realize multiple therapeutic objectives: Prevent potential disaffiliation between the two therapists, foreclose the danger of the parent defensively disenfranchising herself (which is her previous, pathogenic, pattern), and at the same time establish a preferred relational position with both the other therapist (Jenny) and the client.

The detailed “microscopic” attention to conversational elements also enables us to track how “appropriate responsiveness,” doing the right thing at the right time (Stiles & Horvath, 2017), is realized in a clinical situation: In the second example, the therapist again has to deal with the mother’s self-deprecating stance, but this time he leverages the strong alliance developed earlier in the interview with the daughter (Marcy) to re-align the relationship between mother and child. Asking Marcy to align with Minuchin’s assertion of the “right moral order” in the family, in some sense, undermines her own rebellious position with her mother. The analysis provides us with a better understanding of how phrasing and timing creates a “safe place” for Marcy to explore the dynamics between her and mum in a non-defensive way. In SFT the goal is to frame the difficulties the system encounters as a shared struggle to achieve functional roles within the family. In Excerpt #2 we were able to explore an example of the discursive work that realizes this goal: The therapist refers to the ongoing conflict as an “interesting thing.” The positioning of the issue as “interesting” has a potentially positive connotation which is juxtapositioned with the use of the term “problem” that “you” (plural) have. The use of “problem” eliminates culpability, and at the same time “you” (plural) re-emphasizes the theme of mutuality, and suggeststs that the challenge the family faces is a relational issue. The CA analysis draws attention to how this shift from blame to the need to fulfill both mothers’ and child’s needs is realized sequentially and interactively The sequential examination of the communicative turns allows to explicate the nuanced ways the therapist and the clients work discursively to shape new alignments and search for a “new” vision of what is happening between mother and child. In SFT terms Minuchin is re-structuring the family, introducing a hierarchical moral order with parents exercising executive functions and children benefiting from the protection and age-appropriate autonomy such structure provides. We believe that our CA analysis provides the critical window on the dynamic responsivity involved in the implementation of his therapeutic objectives. Importantly, this perspective provides the conceptual framework and analytic tools to explicate how the therapeutic process involves attentive work on relational alignments not only between therapist and clients but between clients themselves.

As we noted in the beginning of this essay, our overall goal is to gain a clearer understanding of how various aspects of the alliance are built, maintained, and made effective in therapy. Examining therapy discourse sequentially in fine-grained detail has provided us with a clearer view of how the relational aspects of treatment are developed interactively, how these alliances partner with various therapeutic objectives and realize forward movement in treatment. The extracts we provided were drawn from a family therapy session. Exploring data from a multi-person context has provided particularly rich opportunities to analyze the way discourse works to not only realize therapeutic objectives between therapist and clients individually, but also re-shape allegiances through selecting participants and structuring the flow of conversation. However, the analytic methods and perspectives we demonstrated are equally applicable and useful to explicate the discursive processes that are at the core of individual psychotherapy irrespective of the theoretical orientation guiding the process (Buchholtz & Kächele, 2017). We believe that, from a broader perspective, research on the essentially discursive, interactive nature of psychotherapy at this detailed level provides a needed compliment to quantitative psychotherapy research.