Goal Setting in Mental Health Rehabilitation: References to Competence and Interest as Resources for Negotiating Goals

Goal setting is at the heart of mental health rehabilitation, but its joint negotiation by clinicians and clients has proven to be a challenging endeavor. This paper investigates goal setting decision-making in the context of Clubhouse Communities: non-profit organizations designed to pave the way for the recovery of individuals diagnosed with mental illnesses. Using the method of conversation analysis, we demonstrate how clinicians make and account for proposals to involve clients in the discussion as more equal partners. In these accounts, clinicians highlight the client’s potential in terms of either competence or interest. Clients, in turn, resist clinicians’ proposals by invoking the opposing factor: when clinicians highlight clients’ competence, clients appeal to their lack of interest and vice versa. In this way, clients are able to reject clinicians’ goal-proposals without disagreeing with the rationalizations of their competence or interest. By contrast, jointly formulated decisions are best reached when clinicians focus the talk on the characteristics of the desired activity rather than on the characteristics of the client. In so doing, clients are able to claim personal ownership of the goal.


Introduction
Rehabilitation is ultimately concerned with changing behavior. It aims to optimize functioning and reduce disability with a specific set of interventions (WHO, 2021). In general, rehabilitation is highly personcentered, meaning that the interventions and the approach selected for each individual depends on their goals and preferences (WHO, 2021). Thus, the identification and setting of goals is at the very heart of rehabilitation. Such goals can be defined as ''the intended outcome of a specific set of interventions'' (Wade, 2009). Often the interventions and activities used in rehabilitation are very ordinary -activities that individuals would perform in their everyday lives, such as cooking, taking a bus to the supermarket or walking to a mailbox. What makes these activities rehabilitation is their goal-orientation: their intended future-oriented results (Wade, 2009). This paper focuses on mental-health rehabilitation, where the principles of goal setting resonate with the so-called Recovery Model (Anthony, 2007), which has recently become one of the guiding principles in the provision of mental-health services.
Whereas traditional psychiatric treatment focuses on patients' symptoms and functioning, the recovery approach focuses on a non-linear process of continual growth that builds the resilience of people with mental illness (Bensing, 2000;Goossensen, Zijlstra & Koopmanschap, 2007;Goss et al., 2008). Recovery can be viewed as an overarching philosophy that encompasses the notions of self-determination, self-management, personal growth, empowerment, and choice (Framework for Recovery-Oriented Practice, 2011). The Royal College of Psychiatrists define the concept of recovery as follows: ''For many people, recovery is the process of developing a new sense of self, purpose in life and hope. It is a journey for the individual and those close to them in rebuilding a satisfying life. Central to the theme of recovery is resilience which allows for individual strengths and coping skills to surface, in spite of adversity''. (Royal College of Psychiatrists, 2012, see Roberts & Boardman, 2013: 403). The recovery approach involves enabling and supporting individuals with mental illness to take control over their lives. Goal setting is seen as an integral part of this process. When goals are personally meaningful, clients report that they facilitate motivation and assist the recovery process (Ades, 2003).
In order for goals to be personally meaningful, the rehabilitation handbooks and other professional guidelines recommend the presence of a number of attributes. These guidelines emphasize that goals must be desired and ''owned'' by the client, that is, set by clients according to their own interests and preferences (e.g., Maybery, Reupert & Goodyear, 2015;Wade, 2009). In addition, goals should be set at an appropriate level of challenge. This means that they are realistic but sufficiently challenging; i.e., they require activities that clients are, at least partially, unable to perform in the present situation (e.g., Struhkamp, 2004;Wade, 2009). To achieve these ideals, it is noted that a client should participate actively and work collaboratively with mental health professionals (Maybery, Reupert, & Goodyear, 2015).
From this perspective, decision-making in goal setting can be viewed as a complex and challenging interactional situation whose success depends on the interaction between professionals and clients (Schoeb, 2014).
This study takes a conversation analytic approach to goal-setting interaction and investigates the detailed practices professionals and clients use when deciding the client's goals. On the one hand, discussing the client's goals is a matter of epistemic authority and epistemic rights (Heritage & Raymond, 2005) -that is, who knows and possesses the right to express what the best possible goal is for the client and how access has been gained to that knowledge. Establishing decisions, however, also involves varying degrees of deontic authority or deontic rights among the participants (Stevanovic, 2013a;2018) -that is, the power to determine what will be done. In goal-setting interactions, participants may be assumed to orient to the need to distribute deontic rights as equally as possible, even if their epistemic rights involve various asymmetries. Decision-making is typically launched by a participant making a proposal for some future action or event, which implies a claim of the right to suggest a decision. The recipient's subsequent treatment of the proposal validates the first speaker's deontic rights, which occurs most effectively through an accepting response to the proposal, thereby paving the way for a genuinely joint decision. However, various responses other than these are also possible. In addition to explicit rejections, there is a multitude of different types of ''non-accepting'' responses, which allow the participants to mutually negotiate the outcome of the sequence. Typically, such responses somehow hamper decision-making: they entirely circumvent the relevance of the proposal for an actual decision (nondecision) or at least postpone its emergence (Stevanovic, 2012). Thus, proposals and the varied possibilities of their uptake are means through which participant negotiate not only the content of the decisions but also the degree of their ''jointness''.
Prior conversation analytic research on goal setting in health care has noted that professionals often balance between cooperation and authority when making goal-decisions. On one hand, professionals attempt to elicit clients' preferences, but, on the other, they also seek to determine goals that are suitable for the given institutional context and judged as advantageous and achievable in the rehabilitation process (Barnard, Cruice & Playford 2010;Davis 1986). Clients' wishes rarely translate straightforwardly into joint, written goals (Barnard et al., 2010). Asking clients to state their goals is also problematic, as they may not know what could be considered an appropriate goal in a given context (Schoeb, Staffoni, Parry & Pilnick, 2014). As a consequence, clients' participation is often reduced to answering professionals' questions and responding to their proposals (Lenzen et al. 2018;Parry 2004a). Although, in this respect, goal setting is dominated by professionals, clients have nonetheless been noted to possess the means to influence the course of interaction. Most typically, they can resist the goals proposed by professionals (Barnard et al., 2010;Lenzen et al., 2018;Murdoch et al., 2020). This resistance is typically displayed passively, for instance, by providing a minimal or nonresponse to the professional's proposal (Barnard et al., 2010;Murdoch et al., 2020). As the client's agreement is generally expected, even minimal responses may be powerful in generating further dialogue and postponing the final decision (Barnard et al., 2010;Stivers, 2008).
Prior research has not, however, investigated the realization of the recovery approach ideals in the interactional practices associated with the clients' personal goal setting. Thus, to better understand good quality interactions, we focus here on the participants' interactional resources for setting goals. We investigate these resources in Clubhouse Communities, which are non-profit organizations seeking to aid the recovery of individuals with mental illnesses. In this context, the relationship between professionals and clients is understood to be one between equals. Goal setting is thus a co-operative endeavor in which decision-making power is shared equally.

Supporting Individual Paths to Recovery in the Clubhouse Community
The Clubhouse model is a non-governmental psychosocial rehabilitation model offering communitybased rehabilitation to individuals with mental illnesses. The model has its roots in peer-support groups of people discharged from psychiatric hospitals in New York in the early 1940s (Anderson, 1999). Since 1950s Clubhouses have involved both client members and professional staff. The current Clubhouse Standards state that each Clubhouse should have enough staff to engage with the client members, yet few enough to make carrying out the responsibilities of the house impossible without member involvement (Clubhouse International, 2021). Accordingly, the Clubhouse staff requires specific kind of professional knowledge: staff members need to know how to facilitate client members' involvement while placing themselves equal with the clients. In this article, we focus specifically on the role of the professional staff as they decide on the client member's goals.
The model has gained worldwide recognition for its effectiveness in assisting those with serious mental illness in their recovery process (Pernice-Duca, Markman & Chateauver, 2013). The model involves local community centers that offer a sense of belonging, opportunities for social relationships and support in obtaining employment, education, and housing (Hänninen, 2012). In Clubhouses, the activities of the community are built around a work-ordered day, which is planned and implemented together with clients and staff members (Hänninen, 2012). The work-ordered day consists of daily activities, such as grocery shopping, staff recruitment, editing the members' magazine, and administration -all activities that relate to the maintenance and development of the Clubhouse community (Clubhouse International, 2021). One central goal of Clubhouse activities is to advance clients' return to the labor market (McKay et al., 2018). In order to develop clients' work competence, Clubhouses arrange rehabilitation groups and transitional employment where clients can, with the support of staff and peers, take steps toward competitive employment (Pirttimaa & Saloviita, 2009).
The organization of communal life at the Clubhouse is informed by the so-called recovery approach, which focuses on enhancing self-determination among clients and control over their own lives (e.g., Anthony, 2007;Bensing, 2000;Goossensen et al., 2007). There is no one definition of what recovery means, but it is often explained as a subjective experience in which a person overcomes difficulties to the extent that they regain a sense of control over their life (Deegan, 1988). This means that individuals may still exhibit symptoms of the illness from which they are recovering (Davidson & Roe, 2007). Thus, recovery can be characterized as a ''journey'' or a process that encompasses five superordinate categories: Connectedness, Hope and optimism about the future, Identity, Meaning in life, and Empowerment (Leamy et al., 2011). In addition, dealing with difficulties and struggles in life is an important component (Stuart et al., 2017). These main components related to the recovery process have been identified in the Clubhouse structure (Pernice-Duca et al., 2013). Although Clubhouses offer community-based rehabilitation, they aim to support their members in their individual paths to recovery. This individual support is offered in several ways, one being goal-setting discussions. Some Clubhouses offer programs that focus on goal setting in specific subjects, such as the ''To Your Health'' program, which aims to increase behavior that promotes members' health (Camann, 2001). In some other houses, such as the Finnish Clubhouses from which our data is drawn, goal setting is more freely structured. Goals are discussed individually when client members and staff decide that they require review, typically a few times per year. The aim is to identify how the activities in which members participate (or wish to participate) in or outside the Clubhouse support their process of recovery. The starting point is that client members define their own goals in the presence of professionals, who facilitate the process in an equal, co-operative manner. Thus, the staff needs to have interactional skills to facilitate this process.
The aim of this paper is to investigate the realization of recovery approach ideals during processes associated with the clients' personal goal setting. We focus specifically on the way staff members facilitate the goal setting process in and through their goalrelated proposals. Our analysis is guided by the following research questions: (1) How do staff members make goal-related proposals in the turn-by-turn sequential unfolding of interaction? (2) how do clients (Clubhouse members) accept or resist these proposals?

Data and Methods
The data consist of six video-recordings of actual goalsetting discussions collected in 2017-2020 in three different Clubhouses in Finland. The lengths of the encounters vary from 17 to 75 min and comprise approximately 175 min of interaction. The data were collected as part of a larger research project Social inclusion, interaction and mental health (VOIS) that investigated decision-making in mental health related problems. In one of the sub-projects, we recorded interactions in various decision-making contexts in Finnish Clubhouses (e.g., ''small scale'' decisions in everyday meetings and ''high scale'' decisions on selecting members to Transitional Employment places). As a part of this data collection three Clubhouses video-recorded a small set of goal-setting interactions, as this setting was considered a highly relevant context for decision-making.
The data come from six different dyads: five staff members with six different clients. In addition, one session includes another staff member who had just begun work at the Clubhouse and was present to observe how to conduct a goal-setting discussion. The staff members were all trained in social work and their work experience varied from six months to several years. All clients had already attended the Clubhouse for a long period of time and had previously participated in at least one goal-discussion. Thus, in each discussion, the participants discussed and evaluated their prior goals and set new ones. The data were collected with one video-camera placed in the corner of the room. One of the Clubhouse members acted as a research assistant and was responsible for recording the meetings. The researchers were not present in the discussions and data about the participating clients' personal background were not collected. The contextual information about the clients that is provided in the results section is drawn from the video-recordings. In the discussion, participants sat either side by side on a sofa or opposite one another at a table. In five of the discussions, a staff member wrote down the goals (either by pen or on a computer), while, in one, the goals were written down by the client.

Ethics
The study was conducted in accordance with the ethical principles stated in the Declaration of Helsinki. For those parts of the larger research project that involved collection of diagnosis-based personal data, we obtained ethical approval from the Ethics Committee of Helsinki University Central Hospital. For the qualitative, interaction-based parts of the project, a university-based ethical review was unnecessary according to the criteria set by the Finnish National Board on Research Integrity (https://tenk.fi/en/ethicalreview/ethical-review-finland). For this reason, research ethics approval was obtained from the Southern Finland Clubhouse Association (date of decision: Sept. 19, 2016), and research permits were granted by the board of directors at the relevant Clubhouses.
Prospective, informed, written consent was obtained from all research participants. Participants' capacity to consent to research was not assessed with any competence assessments tools. Although some of the participants presumably had mental health diagnoses, there was no reason to question their capacity to make decisions regarding their own research participation. The participants had no legally authorized representatives and were able to make their own choices. Still, to consider the possible vulnerability of the participants, we wanted to make sure that the informed consent process was truly informed. Before signing a consent, the participants were informed about the aims of the study, their rights as participants, and about the use of the data. This was done by means of an informing letter and in personal encounters with the researchers who were conducting the study. We used everyday language and plain text in our explanations of the study, and the participants had time to think about their decisions. The participants also had the possibility to withdraw their consent at any point in the study. The anonymity of the participants was carefully ensured by altering all identifying details in the text to avoid any research-related harm to the participants.

Analysis
The data were analyzed by means of institutional conversation analysis (CA) (e.g., Arminen, 2005, Heritage & Clayman, 2010. Conversation analysts inductively investigate recordings of naturally occurring interactions to uncover the reoccurring practices of interaction through which the meanings of social actions are produced. According to the CA view, every turn of talk performs social actions (such as proposing, accounting, agreeing, or resisting). The main idea is to investigate the function of these actions in a given moment of social interaction (Schegloff, 2007). Institutional CA builds on this basic view and explains how the sequences of social actions contribute to the performance of the goals of the institution at hand (Arminen, 2005;Heritage & Clayman, 2010).
In the analytic procedure of this study, the interaction during the goal-discussions was first transcribed according to CA conventions (Schegloff, 2007;see Appendix). Next, the video-recordings were watched several times and all segments of interaction in which a staff member proposed a goal were identified. We found 32 goal-proposals in our data (proposals that were not related to goal-decisions were excluded, for instance, proposals that suggested a time for a followup meeting). In each case, we included the proposal, the turn of talk preceding the proposal, and the response to it. After identifying cases, we began work with the collection of identified data segments in a data-driven way, probing the categories and patterns identified in a single data segment and comparing them against every new segment of data. Finally, focusing on staff members' ways of proposing and clients' ways of responding to staff members' proposals, we conducted a more specific analysis on the interactional resources staff members use for accounting for their goal-proposals, on one hand, and resources clients use for agreeing to and rejecting them, on the other. The data extracts presented in this paper are drawn from across our data set on the basis of their capacity to demonstrate, in a clear and accessible way, the interactional resources employed to produce and respond to goal-proposals.
Results: Focusing the Talk on Clients' Competence and Interest or Qualifications of the Desired Activity?
The goal-setting discussions were typically organized around staff members' proposals, inviting clients' views and agreement on the suggested courses of action. Clients, on the other hand, were reluctant to accept staff members' proposals. Their responses could be characterized as passive resistance, involving unmarked acknowledgements; i.e., rather than overt rejection of the proposal, they involved avoidance of commitment to it as an aim they would pursue in the future (Heritage & Stefi, 1992;Stevanovic, 2012). To involve the clients in the discussion as more equal partners and minimize this resistance, staff members rationalized their proposals by providing accounts (Houtkoop, 1990). In these accounts, they highlighted clients' potential in terms of either (1) competence (the client should pursue the goal because they are capable of achieving it) or (2) interest (the client should pursue the goal because they are interested in it). When, in their following turns of talk, clients resisted staff members' proposals, they invoked the opposite factor: when staff members produced an account of clients' competence, clients appealed to their lack of interest. When staff members accounted for the rehabilitant's interest, clients rejected the proposal by appealing to their lack of competence. In this way, clients were able to reject staff members' goal-proposals without disagreeing with their rationalizations. However, goal-decisions were best reached when staff members focused the talk on the characteristics of a desired activity, rather than on the characteristics of a client. In so doing, they invited clients to specify what would need to change regarding the activity, thereby involving them in the goal-setting activity.
In this section, we start by demonstrating, with two data examples, how staff members use competence and interest to account for their goal-proposals. After that, we move on to describe clients' responses to these proposals and show, continuing with the same data examples, how the opposite factor is invoked for resisting proposals. Lastly, we present data in which the talk is focused on a desired activity instead of the client's characteristics.

Clients' Competence and Interest when Accounting for Goal-Proposals
In our data, staff members actively made proposals, i.e., turns-at-talk in which they named a course of action, suggesting that it could be realized, while conveying that the action was contingent upon the client's acceptance (Houtkoop, 1987). These proposals were designed in various ways, but the most typical format involved a conditional that named a course of action: ''The first goal could be that this situation would continue as it is now.'' ''Should we put that the main goal is recovery, and I think the regular participation at the Club would also be a goal.'' ''I think it could be useful to train the social skills, if you could try to work with other members.'' ''How about if we put more social interaction.'' These proposals were followed by an account that justified the proposed action (why it should be considered a goal). In this manner, the staff members oriented to the need to establish deontic symmetrythat is, even if they made goal-proposals for their clients, they did not possess the right unilaterally to determine the client's goals. These accounts could be defined as ''the social reasoning that people go through to make sense of their words and (perhaps) impose that sense on other people'' (Antaki, 1994: 1). Such accounts not only reveal the sense of the speaker's reasoning but also the need to convince the listener that something is as the speaker claims it to be (Antaki, 1994). When accounting for their proposals, staff members highlighted clients' potential either in terms of competence or interest: ''You can cook very well. Everybody always compliments your cooking.'' ''You've been able to find friends -and that has been because of some joint activities.'' ''Because I think that your social interactions are increased, which is excellent.'' ''As you've been so extremely interested in those Transitional Employment places and this work at [the name of the place].
To provide a more nuanced picture of the staff members proposals and their accounts, we will next present two longer data extracts and their detailed analysis. The first extract provides an example in which a staff member invokes the client's competence after making a goal proposal. Before the extract occurs, the participants have discussed the client's prior work experience. The client has described his study background and explained the types of work he has performed, stating that his aim is a full-time position in the competitive labor market. In the first line, he topicalizes his trouble in working-life.
Extract 1A (CL = Client, SM = Staff member). The client's turn in lines 1-2 is a problem disclosure that enables an affiliation or attempt to solve the problem (Jefferson, 1988;Jefferson & Lee, 1981;Lindström & Sorjonen, 2013). The staff member orients herself to the latter option by suggesting a Clubhouse created transitional employment (TE) program as an option for the client's employment (lines 3). The program allows clients to test their workrelated skills by working outside the Clubhouse community for a short period of time. However, the proposal in not in line with the client's goals, as he has previously expressed his aim of gaining a full-time salaried position. In her proposal, the staff member first topicalizes the TE-program through the interrogative (''how about'') (see Korpela, 2007) and then makes a negatively formatted declarative (''you haven't been yet''), suggesting TE as one possible goal on his way to employment. The client responds only to the latter part of the proposal, confirming that he has not previously participated in the TE-program (line 4). He also states that this is something that has been proposed to him ''a lot'' (line 6). In this way, he demonstrates that he treats the staff member's prior turn as a proposal. As selection for the TE-program is only open to clients who are capable of working outside the community, displaying himself as a person to whom TE has been a long-term option highlights his capabilities in the community. This is something that the staff member targets when accounting for her proposal.
First, the staff member explicitly topicalizes the client's skills and evaluates them in a positive way (''you've got such strong skills''). Second, she uses extreme-case-formulations (''such strong skills,'' ''so incredibly skilled with all those computers''), which have been shown to be a way to legitimize claims in interaction (Pomerantz, 1986). Third, she refers to her own observations at the Clubhouse. The staff member first makes a more general observation about the client's evident skill with computers (lines 9-10). Then, she further justifies her account by highlighting a specific situation at the Clubhouse (Clayman, 2007): she describes how the other Clubhouse members ''hang on the client's sleeve'' to receive assistance with computer-related problems. When the client minimally agrees (line 16), the staff member further compliments the client by showing how his competence appears in the community, thereby supporting his further participation (Weiste et al., 2021).
In sum, to account for her proposal, the staff member portrays the client as a competent member of the community whose skills are needed and highly appreciated. In this way, she can demonstrate that the client is more than capable of applying for the TE program.
Similar to competence, staff members also appealed to clients' interest to justify their goal proposals. Extract 2A, below, provides one such example, taken from a different dyad. Prior to Extract 2A, the staff member has proposed transitional employment as a goal, but the client has resisted. In the first lines (1-5), the staff member proposes that they attempt to find a new goal from the activities available in the Clubhouse community. In the first line, the staff member's proposal is prefaced with reference to his thought (''I was thinking that''), providing the client with more open-ended opportunities to respond (Stevanovic, 2013b). He designs his proposal as a conditional question (''would there be'') and proposes that the goal be some Clubhouse activity in which the client could participate. He also summarizes the activities where the client has already participated (''No Limits Club,'' lines 4-5; ''Clubhouse meetings and executive board,'' lines 15-16), highlighting her activity. When the client confirms the staff member's summary (''yes,'' line 17), the staff member invites talk on the client's interest in these activities. He first asks if they are something the client is interested in (line 18), inviting a yes/no response. Then, he pauses and reverses the polarity (''isn't it interesting?''). By stating his own opinion and adding a new substantial element to what is interesting about the meetings (having the power to influence), the staff member strongly invites the client's confirmation and an elaborated response (Weiste & Peräkylä, 2013).
In sum, in the extracts above, the staff members made goal-proposals by naming a course of action for the client to pursue in the future and following it with an account that justified the proposal. In their accounts, the staff highlighted the clients' potential either in terms of competence or interest. Next, we move on to investigate the clients' responses to the staff members' proposals.

Clients' Competence and Interest when Resisting Goal-Proposals
Prior research has noted that arriving at a joint decision requires three components: access to the proposed subject, agreement, and commitment (Stevanovic, 2012). In our data, the clients generally first agreed with the staff members' proposals. In their next turns of talks, however, they resisted them by invoking the opposite factor to that appealed to by the member of staff in their proposals: when the staff member highlighted the client's competence, the client appealed to their lack of interest. Conversely, when the staff member emphasized the client's interest, the client rejected the proposal by appealing to their lack of competence. In this way, they displayed their lack of commitment to the staff member's proposal, leading to a non-decision. These responses involved turn designs such as: ''Yes of course but obviously something can come along and I'm not capable of doing it.'' ''Yes but I'm also quite unsure about that Transitional Employment, I don't know if I'm capable of it.'' ''I most definitely would like to work with others, I even feel that I should, but then I just feel that I cannot, I'm like 'go away'.'' I think I do but then again I don't know if that is something I would be interested in pursuing right now.'' In order to give a more nuanced picture, we will next analyze in detail the clients' responses to the staff members' proposals presented above (Extracts 1A and 2A). Extract 1B is the direct continuation of Extract 1A. In that extract, the staff member proposed transitional employment and accounted for her proposal by appealing to the client's superior skills with computers to demonstrate his capability. In line 20, the client responds, invoking his lack of interest.
Extract 1B (continuation). In line 18, the staff member explicitly refers to the client's competence to highlight his capabilities in the Community. The client first remains silent (line 19) and then responds with the particle ''nii,'' which displays recognition of and familiarity with the content of the co-participant's talk (Sorjonen 2001)that is, he claims access to the staff member's proposal. Next, however, he continues with the conjunction particle ''mut'' (''but''), which expresses a divergent option (Hakulinen et al., 2004(Hakulinen et al., : 1098 and refers to his lack of interest. The client's claim is accompanied by laughter, thus implying that displaying a lack of interest is, in some respects, a delicate activity (Haakana, 2008). He also hesitates and displays his lack of interest as only partial (''always,'' line 21) and as something he is unsure about (''I don't know,'' line 20). The staff member responds with a minimal acknowledgement (''mm,'' line 22) and the client clarifies that his lack of interest is specifically targeted towards ''maintenance tasks'' (line 23). The staff member agrees, displaying now some recognition of the client's view (''niinpa¨,'' line 24). The client further justifies his claim by referring to the maintenance tasks and describing them, including laugh particles in his talk, as ''irritating'' (line 25). Thus, he could be considered to be passively resisting the proposal. He does not explicitly reject it, but he avoids focusing on its justifications and thus committing to it. The staff member acknowledges the client's claim (''mm-m'') and agrees with the client, showing that she has no such competence and can only imagine the client's irritation (line 26). In this way, the staff member indexes the client's superior epistemic authority to determine his interest (Heritage & Raymond, 2005) and transitional employment is no longer discussed as an option.
To conclude, by referring to his lack of interest, the client was able to reject the staff member's proposal on transitional employment without disagreeing with the staff member's justifications on his competence. In so doing, the client was able to present himself as someone who would be competent, but yet not interested, in pursuing the proposed goal.
Next, we provide an example of clients' responses to staff members' proposals appealing to their interest. In these cases, the clients rejected the proposal by referring to their lack of competence to pursue such a goal. The following data example is a continuation of Extract 2A, in which the staff member proposed a Clubhouse activity goal and highlighted the client's interest in actively participating in meetings. In Extract 2B, the client first confirms her interest and then rejects the proposal by referring to her lack of competence.
Extract 2B (continuation). In line 21, the client confirms the staff member's suggestion, stating that she has considered it herself. She also upgrades her assessment from the staff member's evaluation ''interesting'' (line 18) to ''really interesting'' (line 21). By referring to her thought process and upgrading her assessment, she shows that her views on the meetings are independent from those of the staff member (Heritage & Raymond, 2005). While still agreeing with the staff member's evaluation, she advances her primary epistemic rights in the evaluation of what she finds interesting (Heritage & Raymond, 2005). Next, after a short gap in line 24, she continues with the conjunction particle ''mut'' (''but'') and changes the direction (Hakulinen et al., 2004(Hakulinen et al., : 1098. She describes her difficult emotional experience, ''terribly intense tension,'' when participating in the meetings (lines 24-25). She also describes the consequences of those feelings: she does not dare to say anything and then feels ashamed for remaining quiet (lines 27-28). Thus, the client passively resists the proposal: she does not explicitly reject it but avoids committing to it (Stevanovic, 2012). The staff member affiliates with the client's account with the particle ''nii'' (line 29; see Sorjonen, 2001), and the client continues to describe her troubles in more detail (not shown in the extract). At this point, rather than abandoning his proposal and submitting to the client's epistemic right to know her own capabilities (as we saw in Extract 1B), the staff member pursues his prior proposal. In line 42, he expresses an opinion which contradicts that of the client and states that being associated with new Clubhouse activities would be a ''step forward'' in the client's rehabilitation process (lines 42-43).
To conclude, in the extracts above, the clients passively resisted the proposals by agreeing with them in principle but refraining from committing to them (Stevanovic, 2012). This was achieved by focusing the talk on the opposite factor to that invoked by the staff members when accounting for their proposals: when staff members invoked the clients' competence, the clients appealed to their lack of interest and vice versa. In this way, the clients were able to reject the staff members' goal-proposals and still present themselves as competent and/or motivated individuals. Thus, accounting for the proposals seemed to hamper the progress of the core activity (see Svennevig & Djordjilovic, 2015), i.e., a joint decision on the client's goal, leading to a non-decision. In our data, decisionmaking best proceeded when staff members focused the talk on the characteristics of a desired activity, rather than competence of the client.

From clients' competence and interest to the characteristics of a desired activity
When proposing goals for the client, staff members sometimes focused on characterizing an activity in which the client had been noted to express interest, at least on some level. Rather than inviting talk on the client's competence (Extract 1) or interest (Extract 2) in the proposed activity, they focused on specifying the activity in question. Similar to the previous two cases, staff members oriented to the need to establish deontic symmetry, but, in this case, they shared access to the grounds of the proposal. This type of goal proposal did not trigger the client's resistance, and the participants continued their discussion of the goal. Although these activities were not goals as such, focusing on them helped the clients themselves to formulate the goal. Extract 3 provides an example.
Extract 3 occurs during the same goal discussion as Extract 2. In Extract 2, the staff member proposed that the client participate in some activity at the Clubhouse and referred to her interest in participating in the meetings. The client resisted the proposal by referring to her trouble and ''inability'' to adopt such an active role. Here, the staff member returns to the same subject and suggests, from a slightly different angle, that client ponder ''work assignment meetings.'' Extract 3 (CL = Client, SM = Staff member).
In lines 1-7, the staff member designs his proposal using a conditional question (''what if we put''), marking the plan as contingent on the client's approval (Stevanovic, 2013b). He pauses and names ''work assignments'' as the subject of the proposal (line 1). Then he specifies the referent to be ''joint work assignment meetings,'' (line 2) in which Clubhouse staff and members allocate daily tasks. Next, he requests the client's engagement in joint decisionmaking about the proposed plan: he invites the client to ponder what should occur in those meetings to allow them to run more smoothly and how the client could participate in them more actively by expressing her own preferences. Thus, compared to Extract 2, the staff member focuses on a specific activity at the Clubhouse (work assignment meetings) and invites the client to ponder that activity from her own perspective (lines 3-5). Although he initiates his turn with a proposal and topicalizes the activity, he leaves its specifications open and invites the client to qualify what would need to change regarding the activity in question.
The client responds in line 6, referring to her own thought process and beginning to ponder her own perspective. During a gap in line 7, the staff member remains silent and allows the client time to think. In lines 8-11, she makes a suggestion: she would consider beforehand what tasks she would like to perform so that she could react more quickly in the meeting. The staff member orients to the client's suggestion as a goal proposal and moves on to form a decision on the goal. He asks the client to write down her suggestion (let's put it) and repeats part of it, establishing a joint decision on the goal. The client begins to write, and they move on to another subject.
To conclude, when the staff member focused the talk on the desired activity and invited the client to qualify what needed to change regarding that activity, the client herself proposed specific solutions that could be written down as a goal. In this way, the decisions were best reached and jointly formulated.

Discussion
This paper has demonstrated the ways staff members facilitate the goal setting process in and through their goal proposals. To involve clients in the discussion as more equal partners, staff members rationalized their proposals by providing accounts that highlighted the client's potential in terms of either competence or interest. By focusing on clients' competence and interest, they were also able to avoid directly articulating their inabilities and troubles in their everyday lives. As Sacks (1992) has noted, ''normal'' behaviors and experiences in everyday routines and lives are socially controlled moral concerns. Thus, addressing clients' inability to manage their everyday lives directly as an issue requiring change might constitute a face-threatening act (Goffman, 1955). Rather, as goals are inevitably set for future situations requiring actions that clients are currently unable to perform (Struhkamp, 2004;Wade, 2009) reassure them that they possess the ability to attain the goal.
Our analysis also demonstrated that clients passively resisted staff members' proposals by agreeing with them in principle but refraining from committing to them (Stevanovic, 2012). This was achieved by focusing the talk on the opposite factor to that invoked by staff members when accounting for their proposals: when the staff member invoked the client's competence, the client appealed to their lack of interest (Extracts 1 A and B). When the staff member invoked the client's interest, the client rejected the proposal by appealing to their lack of competence (Extracts 2 A and B). In this way, clients were able to reject staff members' goal-proposals and still present themselves as competent and/or motivated. From the perspective of the core activity, i.e., a joint decision on the client's goal, accounting for proposals hampered the progress of interaction (see Svennevig & Djordjilovic, 2015), leading to a non-decision. In our data, decisions were best reached and jointly formulated when staff members focused the talk on the characteristics of a desired activity rather than on the characteristics of the client (see Extract 3). Although these activities were not goals as such, focusing on them helped the clients themselves to formulate the goal. This finding is supported in numerous clinical handbooks and guidelines that emphasize the need to search for clients' own interests and preferences when setting a goal (e.g., Maybery, Reupert & Goodyear, 2015;Wade, 2009).
Inasmuch as staff members refer to clients' own interests and preferences as a basis for suggesting to the client specific plans for future action, they can be seen as orienting to the need to establish deontic symmetry between themselves and the client. As shown by Peräkylä (1998), when professionals produce statements about the specific characteristics of their clients, it is crucial for them to incorporate references to publicly observable and accessible evidence for their conclusions. In so doing, they treat themselves as accountable for what they say and offer their clients the possibility to correct those accounts if needed. In a decision-making context such as the goalsetting meetings investigated in this study, such accountability is an essential feature of shared responsibility for the emerging decision. Both participants must reach a common understanding on the epistemic basis of the decision. It is at this point, however, that complexities emerge. When making claims about clients' interests and preferences, staff members approach their clients as unique individuals, which is consistent with the notion of personalization (Hunter & Richie, 2007). However, in so doing, staff members inevitably place themselves in an epistemically subordinate position in relation to their clients, who have direct epistemic access to their own inner life (Weiste, Voutilainen & Peräkylä, 2016). Intriguingly, epistemic subordination is not only associated with preferences and interests but also with competence. Even if a staff member is able to refer to publicly observable evidence of a client's apparent competence, the client may still seek to refute the staff member's conclusion by referring to their inner fears and anxieties as evidence of actual incompetence. Whether or not such client resistance prevails, it nevertheless hampers the progress of the decisionmaking activity.
Professional recommendations, such as those used in the recovery approach, emphasize each client's individual path to rebuilding a satisfying life, where individual strengths and control over one's own rehabilitation process are central (Anthony, 2007; Royal College of Psychiatrists, 2012, see Roberts & Boardman, 2013). This also contains the idea that the goals of rehabilitation are based on personal but realistic starting points. In a collaborative professional-client relationship, naming and making visible staff members' observations about clients' competence and interest is seen as a way to help clients recognize their own life situation and strengths, which form the basis for goal setting (Francis, 2014). At the same time, it is the staff member's professional responsibility to consider what is possible both in the particular institutional context and in terms of the client's competence . However, as our results demonstrate (Extracts 1-2), connecting the proposal for a possible goal to a rationalization highlighting the client's competence or interest creates an interactional challenge to the progress of decisionmaking. On the other hand, as Extract 3 shows, clients' epistemic authority is retained when staff members propose an activity that is possible in the particular institutional context but offer clients the opportunity to position themselves vis-à-vis the activity by naming the required actions. In this way, the proposed activity is ''context-suitable,'' but ownership of the goal is promoted by the client personally. In the future, more research is needed on how these epistemic and deontic relations are different when the goal-decisions are initiated by clients.
Staff members' focus on their clients' competence and interest, and avoidance of explicitly articulating their incompetence in goal-setting situations might stem from the recovery approach in mental health rehabilitation, which highlights continual growth and recovery instead of difficulties and symptom reduction. Nevertheless, according to Stuart et al. (2017), the recovery approach, which disregards incompetence and struggle, may end up marginalizing and blaming individuals instead of empowering them. However, the tendency to avoid discussing clients' failure to accomplish mundane activities is not characteristic only of mental health rehabilitation. For instance, in her study of interaction between physiotherapists and patients recovering from strokes, Parry (2004b) noted that therapists avoided verbally exposing the shortcomings of their patients' performance and rather suggested actions without verbally identifying problems. Parry (2004b) also noted that when patients' lack of ability was discussed, it was mitigated or ''cushioned'' by constructing it as something that could be improved with therapy. Similar ''cushioning'' occurs in Extract 2b, where the client verbally exposes her inability and the staff member highlights that it is something the client can improve in the Clubhouse. Although Parry (2004b) demonstrated how both therapists and clients prefer to minimize the exposure of incompetence and emphasize competence, she also noted that patients' talk about their inabilities served an empowering interactional function: verbal exposure of inability indicated that patients were sufficiently competent to recognize their shortcomings. In this way, talk about inabilities became a resource for normalness construction.
Our analysis nevertheless has certain limitations. The small number of goal-setting discussions (n = 6) in our data means that the interactional practices described in this article may well fail to represent all the ways in which goal setting is facilitated in Clubhouse interaction, let alone in other rehabilitation contexts. However, in conversation analysis, the adequacy of the sample size is not determined based on the number of participants but on the appropriateness of the data for the unique mode of inquiry (O'reilly & Parker, 2013). The dataset was rich enough for us to be able to identify the reoccurring interactional pattern (staff members' proposals and their accounts, n = 32) across the dataset. To control the validity of our qualitative analysis, we also sought evidence for our analytic claims from how the participants in interaction interpreted the meaning of the preceding talk (Peräkylä, 2011). By investigating the next turns of talk (clients' responses to staff members' proposals) we were able to reveal how previous talk was understood by the participants themselves (Sacks et al., 1974). In this way, we have been able to describe in detail some of the epistemic struggles that clinicians may easily face when facilitating deontically symmetrical goal-setting discussions, and we thus provide a more nuanced understanding of the interactional practices required for high quality goal-setting.
Funding Open Access funding provided by University of Helsinki including Helsinki University Central Hospital. Research for this study was supported by the Academy of Finland (Grant No: 307630) and the University of Helsinki.
Data availability The authors confirm that the data supporting the findings of this study are available within the article (the anonymized data extracts). The whole collection of cases is available on request from the corresponding author in Finnish. The data are not publicly available due to privacy/ethical restrictions.

Declarations
Conflict of interest The authors declare no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.