Article found at articles.com
PRACTICING AS A POSTMODERN SUPERVISOR

In this article, aspects of postmodern supervision are explored that highlight the blurring of boundaries that occurs between the multiple roles performed by supervisors. An approach to supervision is detailed that shows how a supervisor who navigates between roles can assist supervisees in constructing identities congruent with the stories they tell about themselves as successful therapists. Six of the many possible roles that supervisors play are illustrated through dialogue taken from a group supervision session. In the last part of this article, concepts that inform a postmodern approach to supervision, borrowed from the literature on narrative and constructionist therapy, are discussed.

Traditionally, what supervisors model through their supervision has been intended to fit with what supervisees hope to accomplish through their clinical work. As Yingling (2000) explains, "I try to create a relationship context where systemic interventions can be isomorphically experienced" (p. 37). Although systems thinking and postmodern principles of family therapy supervision may diverge, there is still merit to the argument that the supervisor succeeds only to the extent that good practice is modeled through the process of the supervision.

In this article, I examine aspects of a postmodern approach to supervision, specifically the blurring of boundaries between the multiple roles supervisors are perceived to perform. I explore how, as a supervisor, I might navigate my way through multiple roles, offering supervisees opportunities to construct identities through our relationship that are congruent with the stories that they prefer to tell about themselves as therapists-stories with which they describe themselves as successful therapeutically and engaged with those with whom they work. This process is similar to the one postmodern-oriented therapists seek to construct with their clients. I accomplish this goal of supervision best when the process I use as a supervisor moves me fluidly through the multiple roles I perform as the supervisor, creating concurrent self-constructions of myself that are helpful to supervisees in their training. I make the effort to help supervisees experience themselves much like clients experience themselves, co-constructed by the process of therapy. Thus, my goal during supervision is to offer the best experience possible for supervisees to experience preferred identity conclusions, ways in which they wish to be known to themselves and others as therapists. In this postmodern approach I am never just "the supervisor," a singular identity determined by my role, but instead I am co-constructed in multiple ways through interaction with the supervisee, depending on what the supervisee wants or needs.

I begin with a description of six of the many possible roles that supervisors might play and argue for understanding each as both contextually and temporally relevant. An illustration of dialogue taken from a supervision session illustrates the process of the supervisor navigating between these six identities as the supervisee is invited to construct his or her preferred identity story as a therapist. In the last part of this article, I explore concepts which might inform a postmodern supervision, borrowed from the literature on narrative and constructionist therapies.

I prefer in my work to use the term "postmodern" as opposed to "narrative" or "constructionist" in order to avoid the totalizing effect of association with only one emerging branch of therapeutic practice (for example, that of the Dulwich Centre in Australia; White, 1997). Broadly speaking, postmodern theory is a collection of interpretations made about the world that are constantly changing. Postmodernists, a titular oxymoron, abstain from metanarratives, that is, grand explanations for how the world ought to be (Best & Kellner, 1997). Despite their reluctance to put forth grand theory, those who theorize about postmodern epistemology most often focus attention not on the individual, but on the process by which our world comes to be known through the language we use, the power dynamics of who controls how language is used, and the stories that we collectively tell (Berger & Luckmann, 1966; Denzin, 2001; Newman & Holzman, 1997). These stories have previously been called "narratives"-stories told over time about our lives embedded within the social discourses that influence and shape them (Freedman & Combs, 1996). In this portrait of a chaotic world in which the way we know ourselves is through stories, negotiated with others, in which cultural pluralism is celebrated (each culture having its own truth claims), there is tremendous potential to argue that our identity is fluid, changing to match the demands made upon us. It is an optimistic point of view. With fluidity in the ways we know ourselves and are known by others comes the possibility to find unique solutions that fit well within the discursive spaces in which we live and interact, solutions that address the problems that challenge our sense of well-being. Such a postmodern philosophy finds expression in a number of clinical forms, including the work of White and his colleagues in Adelaide, Australia, and elsewhere (Fox, Tench, & Marie, 2002; White, 1997, 2000; Winslade, 2002), that of Newman and Hozman (1997) in community psychiatric settings, and my own work and that of others who are concerned with high-risk children and their families and the narratives of survival that they author and sustain (Madsen, 2000; Smith & Nylund, 1997; Ungar, 2001, 2002, 2004a, 2004b).

More specifically, in the field of family therapy supervision, postmodernism has been discussed by Wieling et al. (2001), White (1997), Gardner, Bobele, and Biever (2002), Fox et al. (2002), and Winslade (2002), among others. Not surprisingly, this postmodern practice has been no more evaluated than have most other approaches to supervision. According to Storm (2000; Storm, Todd, Sprenkle, & Morgan, 2001) there is a surprising lack of research that has demonstrated the effectiveness of supervision of any kind. In the absence of hard evidence, I present anecdotal support for the usefulness of this postmodern approach in this article. Although it is beyond the scope of this article to discuss at length, efforts to evaluate this approach will require a methodology that fits paradigmatically, relying on the supervisees' co-constructions of the supervision process and includes them as coresearchers (for related work, see Flemons, Green & Rambo, 1996; Gaddis, 2004).

The Flexible Supervisor
As supervisors, the identity constructions we assume are experienced as different roles in relation to those we supervise. Each of the roles that we perform reflects our diversity. Each is not only an expression of ourselves as professionals, but also who we are as individuals. Our identity as supervisors is, of course, also an expression of our different cultures, genders, races, ethnicity, sexual orientations, and abilities. Starting with such a plurality of possible selves, when we encounter supervisees we have much to draw on and much to account for. Most supervisors already acknowledge these contested terrains of diversity and are transparent in their practice (Fine & Turner, 2002). As a postmodern supervisor, this diversity is also embraced as just another way in which the supervisor accentuates aspects of his or her identity in order to participate with supervisees in a co-construction of the supervisees as competent in their practice. When accomplished well, a supervisor who is fluid in his or her selection of an identity as supervisor opens up possibilities for the therapist (and supervisor) to play many different roles, both when practising therapy and during supervision. Although supervisor-supervisee is one possible way to describe the roles enacted through the relationship, multiple and intersecting roles are to be encouraged if the problems that supervisees encounter doing therapy are to be overcome. In this way, I must also reflect on how I have constructed my identity as a supervisor, seek challenges to that identity when it is unhelpful to the supervisee, and finally define an identity for myself that I can perform in a way that fits well with what the supervisee requires. In the process, boundaries between the many possible selves I might define become necessarily blurred.

Six Role Constructions
The boundaries between the roles we perform are arbitrarily constructed, defining us more in the minds of others than in ways meaningful to ourselves. Boundaries do not exist as objective fact, as any parent, partner, employee, or child knows. At any one moment we may perform one or more of these roles, and define that role as encompassing elements of the others. Postmodern understandings of the self as socially constructed mean that there is no essential self nor defined roles for the self to perform (Best & Kellner, 1997; Gardner, Biever, & Bobele, 2002). Instead, there are only behaviors, invested with names that we collectively use to categorize those behaviors. I am a "father," for example, when I do things that, according to others in my culture, define me as a father. So it is in supervision. Observable in my practice as a supervisor, there are six roles (and likely more) that I perform, each invested socially with particular meaning and expectations by myself, my colleagues,, and those I supervise. These roles include: supporter (to the supervisee), supervisor, case consultant, trainer/teacher, colleague, and advocate (for both the client and/or supervisee). Each role is slightly different:

* The supporter is the role I assume when I support the supervisee emotionally, helping him or her to explore the meaning of the experience of therapy on a personal level. Within the bounds of the code of ethics, this role means that supervision explores emotional hurdles faced by a supervisee that relate to issues beyond his or her immediate work, without necessarily being so intrusive as to lead to the dual relationship of the supervisor becoming the supervisee's therapist.

* The role of supervisor is played when I join with the supervisee to help make him or her the best clinician he or she can be, drawing on the talents and abilities unique to that person. As a supervisor (as opposed to supporter) my focus is on the person in his or her role as therapist, rather than the therapist as a person beyond the role he or she plays in the clinical setting.

* The case consultant offers advice on best practice options, clearly articulating the practice expertise of the supervisor. The case consultant does exactly as the name implies, comments on the situation of the client and what the client may need rather than the nature of the therapist's work with the client.

* The trainer or teacher does more than consult: He or she instructs the supervisee in how to do an intervention, rehearsing techniques or coaching live performances of therapy. The trainer or teacher shares not only his or her own expertise, but also the collective knowledge of the field of practice in which he or she works.

* The colleague shares clinical responsibilities with the supervisee for an individual or family, working together collaboratively to fulfill a mutual mandate to help. In this way supervisee and supervisor become more peer-like in their interactions.

* The advocate either encourages the supervisee to take action, or him- or herself as supervisor becomes involved in direct action to garner the resources necessary to ensure the well-being of either the supervisee or the client with whom the supervisee is working. Being the advocate means either extending practice beyond the realm of the supervision session or encouraging the supervisee him- or herself to become more active in helping clients to negotiate for resources and representation.
In a postmodern supervision, these roles necessarily blur together. Supervision models for supervisees have multiple identity constructions that can be experienced simultaneously. As in supervision, during therapy, multiple identity constructions are experienced by the supervisee who must negotiate each with those he or she is helping. Thus, the process mirrors that which supervisees are encouraged to use with their clients. In this way, supervision is a bid to deconstruct a narrow definition of the therapist/supervisor and encourage a more fluid practice based on postmodern theory that encourages people to see themselves as simultaneously multiply determined and participants in a process of self-definition. Thus Storm and her colleagues (Storm, Todd, Sprenkle, & Morgan, 2002) have noted a trend among supervisors to "have wide latitude to focus on the interface of their supervisees' personal and professional lives" (p. 232).

A postmodern turn in the field of family therapy and greater sensitivity to the intersectionality between the personal and professional is leading to ever-lengthening lists of the roles that therapists play. This same trend is emerging slowly in the supervision of family therapists. Todd and Storm (2002), for example, distinguish between training, consultation, and supervision as roles that supervisors fulfill. To them, supervision is a multidimensional process in which a "qualified therapist monitors professional development and socialization of a partially qualified clinician" (p. 2). Multiple roles, they say, are "frequently intermingled" (p. 3) with the impression given by those who have examined supervision "that these sharp distinctions are often obscured in activities that are labelled 'supervision'" (Storm et al., 2001, p. 229). A postmodern analysis of supervisor-supervisee relationships should go further, to understand all three roles that Todd and Storm identify (and many others) as aspects of a fluid process of co-construction. Supervision is the sharing of wisdom just as it is the collecting of others' expertise, while also intervening and building relationships that further the goals of good clinical practice. However, supervisees appear to contract with me to address their (1) lack of technique, leading me to assume the role of consultant and trainer; or (2) lack of ability to use the techniques that I share, which casts me in the role of supervisor who offers a reflective process for supervisees to examine critically what they are doing and even why. Alternately, in helping supervisees to make use of their knowledge, I am also expected to model concurrently the roles of therapist, advocate, and colleague, even as I seek to perform the more limited role of just being their supervisor.

Supervision, which navigates the spaces between the roles as much as each role itself, allows those being supervised to experience the synergy between helping, playing the professional, being concerned, and living meaningfully, all aspects of what is arguably the therapeutic process that we explore through supervision. Furthermore, it is an artificial distinction to see boundaries between roles when each is so culture bound. Supervisees have not been very good at perceiving their supervisors as unidimensional guides performing their role beyond culturally defined expectations. For example, in a study of 160 supervisees, Anderson, Rigazio-DiGilio, Cochran-Schlossberg, and Meredith (2000) found that those interviewed appreciated different aspects of supervision, training, and consultation, but each argued that all three were important to their growth as therapists.

CASE EXAMPLE
In the following example, the supervisee says she becomes stuck during her work with a 5-year-old boy and his family during the boy's treatment for sexual abuse. She works on a long-term protection team at a local Child and Family Services Agency. A dynamic woman in her late twenties, the therapist appears to center her expert knowledge about the boy and his experience, only to become bewildered when he refuses to engage in treatment in ways that she had expected. Her problem, she says, is being "stumped," the therapeutic process stuck, even though the boy seems content to visit the therapist and play. He, interestingly, does not appear to be stuck. Supervision, taking place in a group setting, sought to consider alternate ways in which the boy may be constructing the experience of his "abuse," the sources of his feelings of trauma, and his experience of the therapeutic relationship. As a new story for the boy emerges, the therapist reconsiders her role as a clinician with so young a child, having to respect his expertise regarding his life as she would any other older child or adult. Through the multiple roles I assume as the supervisor, I seek to move our conversation toward consideration of multiple realities and the child's own competing "truth claim" (Anderson & Goolishian, 1992). In the process, I am not only the supervisor eliciting the supervisee's experience, but also an advocate for the boy (whom I want to be heard), a teacher as I express ideas that would decenter the therapist from the therapy, supporter to the supervisee briefly as we explore her reaction to the child's abuse, and finally case consultant when I advise on a course of treatment.

Supervisor: Okay.

Therapist: At age 4, he disclosed to his mom that his dad's pee tastes like salt [she laughs a little nervously]. So there had been some pretty hard-core sexual abuse for about a year as that was the time when the dad got access. After the divorce, the dad got access to the boy unsupervised. So after this got disclosed, they had a CPS [Child Protection Services] investigation, and the evidence they got against this person was just amazing. He was already in jail for an assault and he was simply held in jail. Okay, so that's all said and done. Mom works part-time. Her mother helps to look after the boy. When this boy disclosed it was all pretty intense [she laughs lightly] for a 4-year-old especially [she emphasizes these last words]. But he never talked about it again in terms of the sexual assault. Not with anyone. But he talks all the time about his father. About how much he misses him and wants to be with him. Apparently it has been found that his father worked at him for a long time, told him that single mothers are bad, that people who aren't married will go to jail, really brainwashing the kid, making him hate his mother. So this child will talk about his dad with his mom but it's all about "I hate you because you won't let me see dad." He's so tortured. You can just see the torture and anxiety on this little fella's face. He says he just hates everyone who won't let him see his dad.

Group Member #1: What else do you know about him?

Therapist: He has slight delays, he has trouble achieving his milestones. But it's not that bad considering everything that's happened to him. Mom says it's all gotten worse since the sexual abuse. And mom has gone through a really difficult time with lots of guilt over what happened. She's real depressed. She's so sad, sometimes she really down, and then when he comes into play therapy he's really down with what's going on. The family has had a lot of problems with this, as the sexual assault was so bad [she emphasizes these words]. There was anal penetration as well, so it was like so bad.

Supervisor: And how did this all come out then?

Therapist: He was just sitting at the kitchen table and it all came out, like he just started talking. He was able to talk about this real easily at the time.

Supervisor: You said you were stumped. What's making you stumped?
Therapist: Well in most cases, with most children, the stuff we do, a lot of it is, when children have been assaulted their development has been interrupted, their development has been arrested, they may not want to play anymore, they want to take care of mom, or in a family violence situation they just stop playing. They're slightly parentified. So we bring the child in and just play, play and see what happens. We want to make the child feel better, to get things back on track. So I've been seeing him for about 6 months. He gets extremely agitated whenever I try to direct the process. Anything about his feelings. He won't talk. He is very repetitive, he only wants to do one thing. He's very, very specific about what he wants to do when he sees me. He's fascinated by robots and machines that can destroy things-like bulldozers, things like that. He's obsessed with those things. He's very specific in what he'll play with. So I've tried to make my strategy fit with whatever he wants to do. Only once, though, has he talked about his dad, about how at his dad's he had a truck like the one we have at the agency or something. But any other questions, no go. Like I have to try not to be so directive 'cause I was trained as a child protection worker. But one time, he did say he wasn't allowed to go over to his dad's place anymore because they used to touch each others private parts allll the time [she draws this phrase out], like allllll the time, every day, and that was the closest thing to a disclosure I've ever gotten. And then he became very agitated again. But that's like that for many of the kids we work with, but this little guy has so stumped me because he gets so frustrated with me, but he keeps wanting to come and see me. He tells his teacher when he is going to come and see me, "I'm going to go see [the Therapist] today," and we meet in the same place everyday, and I know his mother is really great and doing some good stuff with him. Talk about resilience! He's got it!

Group Member #2: How do you know he gets frustrated with you?

Therapist: That's a really good question. He clenches his fists, holds his breath, and he said, oh he's so cute [she pauses and smiles, then sighs and continues], he lashed out at me. I can't remember exactly what he said, but I think it was, "You don't know how to play by the rules. You don't know anything about what I want" [she breathes out heavily, then laughs lightly at what the boy has said] and he's like only 5. It kind of left me a bit stumped. But he redirected really quickly. But then I said, "You can teach me." And he was like, paused, and then said sure, and then he began to show me a bunch of stuff. I don't know, I'm really stumped.

Group Member #1: What does he do with the robots?

Therapist: Well that's really good. There's these robots and there's good robots and bad robots, and there's this robot that regulates everybody. So he gets thrown into the fire pit if your robot is attacked by another robot and is disabled, your robot gets shoved into the pit and discarded. And if one of the robots does something bad, whichever robot the boy is holding becomes extremely agitated and then the head robot has to regulate everything, to make it better. Then he gets frustrated. There's a lot of symbolism there, and his mom says that's all good, but I'm feeling I'm stuck in this trap. What have I done for this child? What do I have to offer? Other kids I'd know what to offer, where to go. But what is it about this kid that's gotten me so stumped?

Supervisor: I'm curious about something, something I need to clarify. The trauma, what is the trauma in this boy's case?

Therapist: The sexual assault.

Supervisor: Yes, I heard that. But in this case, the trauma, you've said nothing to make me believe that. I'm sorry, but you've said nothing that's led me to think that the anal penetration was traumatic for him. Did he say that was a traumatic event [Advocate role]?

Therapist: No he said that he thinks that's normal behaviour for a dad.
Supervisor: So, sorry, I'm getting to something, so the trauma, well there's been no trauma before the system intervened. Before people like us tried to help [Teacher role].

Therapist: Well, no. Oh my God, yes that's so true. The system intervened a year ago. He was being interviewed at the agency and they interviewed mom first and then mom skipped across to the police station, because it was right across the way, and just grandma was there when he came out of the interview room. And dad had said bad things would happen to mom if he said anything. So he tells, and then mom is gone. I guess he was inconsolable. And the fear and trauma he went through when he came out of that interview room and mom wasn't there.

Supervisor: When we started this conversation, talking about the trauma of the abuse, and there was this amazing evidence before the courts, and you used phrases like "brainwashing this kid" in reference to what the father says about the mother. And I'm trying to reconcile that what you told me about how he is tortured with anxiety, but at no point did you say that he had negative feelings about the abuse. It's all been about the system responding to the abuse. Unless I missed it.

Therapist: No you've got it. You're right.

Supervisor: So are you doing therapy on the systems' response to the abuse or the abuse?

Therapist: I don't know, I'm not sure. [At this point another group member interjects and we explore together what the boy's mother may have been feeling throughout this experience of disclosure and investigation.]

Group Member #2: How is the boy doing now?

Therapist: He has sexualized behaviour. He's aggressive with the other children. He masturbates in public. She's concerned about that. That's not unusual in cases like this, but mom's really upset about that. He has nightmares. Wetting the bed.

Supervisor: Is the question, then, are we all more comfortable talking about the abuse done by the father? [Supporter role] And I know this is a bit weird to say, but are we more comfortable talking about that than the system's abuse of the boy?

Therapist: The system's abuse?

Supervisor: Yes the system's abuse of the boy. And I don't know if I have permission here, we don't have a very long working relationship. But I got something else that I want to give back to you, if that is okay with you?

Therapist: Yes, for sure.

Supervisor: Well you said about the boy, you talked about the abuse, then you said, "He's so cute." And he said, "You don't know what I want," and you said laughing, "He's only 5." And I'm sort of wondering about that. Has he entered a world here where people are not hearing him, not because we don't mean to, but if we step back a bit, we aren't seeing what is really happening? You said you were stumped, and I'm wondering if that's because there is no contract? [Supervisor role] The boy's contract may be extremely different, a contract different from the one of people who see him as a cute innocent victim who lacks the precociousness to express himself. I'm wondering if we, and that includes me, really see him as a full-fledged victim able to express himself [Colleague role]?

Therapist: Hmm, hmm.

Supervisor: Rather than a full participant who can make sense of this event in his own way, which to hear him speak about it, he's saying he's frustrated about not seeing dad. He's really frustrating the professionals around him, because he's saying he has a really different story about what happened than those professionals. And what about the masturbation, for example. Does he find that fun?

Therapist: Oh yeah, of course.

Supervisor: And people have asked him about that, and he's said he likes it.

Therapist: Oh yeah.

Supervisor: You see, that is quite different isn't it?

Group Member #3: I find it interesting what you are saying, because we see the exactly same clientele, in terms of the contract you make with a 5-year-old or 15-year-old. And they're being brought into therapy and half of them don't know why they're there. And we're really dealing with the family's anxiety or worry, but it is just a fundamental of therapy, but you have to be mindful of that, that it's the parents you're contracting with often, not the kids.

Supervisor: Well you said about the boy, you talked about the abuse, then you said, "He's so cute." And he said, "You don't know what I want," and you said laughing, "He's only 5." And I'm sort of wondering about that. Has he entered a world here where people are not hearing him, not because we don't mean to, but if we step back a bit, we aren't seeing what is really happening? You said you were stumped, and I'm wondering if that's because there is no contract? [Supervisor role] The boy's contract may be extremely different, a contract different from the one of people who see him as a cute innocent victim who lacks the precociousness to express himself. I'm wondering if we, and that includes me, really see him as a full-fledged victim able to express himself [Colleague role]?

Therapist: Hmm, hmm.

Supervisor: Rather than a full participant who can make sense of this event in his own way, which to hear him speak about it, he's saying he's frustrated about not seeing dad. He's really frustrating the professionals around him, because he's saying he has a really different story about what happened than those professionals. And what about the masturbation, for example. Does he find that fun?

Therapist: Oh yeah, of course.

Supervisor: And people have asked him about that, and he's said he likes it.

Therapist: Oh yeah.

Supervisor: You see, that is quite different isn't it?

Group Member #3: I find it interesting what you are saying, because we see the exactly same clientele, in terms of the contract you make with a 5-year-old or 15-year-old. And they're being brought into therapy and half of them don't know why they're there. And we're really dealing with the family's anxiety or worry, but it is just a fundamental of therapy, but you have to be mindful of that, that it's the parents you're contracting with often, not the kids.

Clearly, the therapist in this example holds certain values about the boy and has the best interests of the child at heart. Her being stumped, however, is about her construction of the boy as a less-than-equal participant in the therapeutic process. Through supervision, what becomes clear is that there is no contract with the boy that meets his needs. Supervision seeks to challenge the therapist's construction as the knowing adult, the expert on trauma, and the conceptual baggage that both identities bring. At the same time, as the supervisor, I try to elicit from the therapist another story of what role she could assume with the boy: a concerned playmate?; a surrogate parent?; a trustworthy adult? Each of these roles may be valued by the boy more than "therapist" to his unnamed experience of abuse.

In trying to help the therapist to overcome her feelings of being stumped, I move between the different roles I perform as a supervisor. Looking back over the passage, I see moments when I am the therapist's (and group's) supporter ("Is the question, then, are we all more comfortable talking about the abuse done by the father?"), supervisor ("You said you were stumped. What's making you stumped?"), case consultant ("So he's found something from you that he needs, a place to play"), teacher ("There's been no trauma before the system intervened. Before people like us tried to help"), colleague ("I'm wondering if we, and that includes me, really see him as a full-fledged victim able to express himself?"), and advocate for the boy and therapist ("Did he say that was a traumatic event?").
This approach appears to have worked. The therapist eventually owns her part in silencing the boy's understanding of all the good things that therapy is providing him. She agrees with what I suggest as her colleague and supervisor-and the boy's advocate-that the boy is getting what he needs from her. She says, "yes" several times during our conversation in regard to my interventions. But she also says, "I'm so worried," making it clear that this is her problem, not the boy's. She reported leaving our supervision far less stumped, and in subsequent follow-up sessions said that she had remained well-engaged with the boy, but feeling less like she was failing as his therapist. She could see her own blurred boundaries and roles that she performed as the boy's therapist, just as during supervision, I too had assumed multiple roles as her supervisor. Being guided by postmodern concepts during this work helped me to perceive the multiple and intersecting identities that the therapist had assumed in her work. I might call these, from my perspective, "parent," "playmate," "savior," "protector," or "expert" (in the sense of trying to be more knowledgeable than the boy about what he needed). Each of these self-constructions had undermined her preferred identity as a competent "therapist" even though the evidence was there all along that she was doing wonderful work. Using our interaction, and interaction with the group, supervision became a process of negotiating with the therapist for new identity conclusions that were more congruent with how she wanted to know herself and be known by her colleagues and clients. And yet, at no time do I completely refute any of the multiple roles she assumes. Her desire to protect the boy is supported, but this identity, through our conversation, is reconstructed in a way that helps the therapist to see herself as successful at protecting the boy by providing a safe place for him to engage with a caring, consistent adult. Thus, the therapist, like me as the supervisor, can sustain multiple self-definitions that snap together well in the service of clients (or supervisees).
Admittedly, trying to partialize each role as I have done here is an artificially constraining exercise, as it is the context in which I say what I say and the reaction by others that help to determine if my intervention positions me in one role or the other. Said authoritatively, a phrase like "I wonder . . ." can be interpreted as me trying to teach others what to think or a collaborative musing with colleagues that sincerely includes me, beyond my role as supervisor, fully in the conversation as an equally confused learner searching for answers.

Concepts Relevant to Practising as a Postmodern Supervisor
Although any list of concepts that purport to delineate a postmodern approach risks leading to the illusion of foundational thought, I have found that several ideas help to orient aspects of my practice as detailed above (for related work see Gardner et al., 2002; White, 2000; Wieling et al., 2001). Combined, these constructs help me to remain decentered, conscious of my subjectivity, and grounded in my culture and context. I list here five of these ideas for consideration by other supervisors seeking to integrate a postmodern orientation into their work.

1)    The meaning of expert. A postmodern understanding of supervision contributes to an acknowledgement that meaning is always in the process of being negotiated. As Gardner et al. (2002) explain: "Clients, therapists, supervisees, and supervisors are always in the process of constructing new meanings about themselves and those with whom they are in conversation" (p. 218). This naturally leads to some confusion over the relative positioning of knowledge and practice experience as two different aspects of supervision. When is it appropriate for the supervisor to cloak him- or herself in the vestments of expertise? When is it appropriate to share past work experience, or to play the expert roles of trainer or consultant? I believe postmodernism provides a theoretically sound alternative to positioning the therapist as expert. As Gardner et al. write: "Many postmodern supervisors strive to make use of experience and knowledge gained from training, education, and clinical experience while at the same time appreciating and nurturing the uniqueness of supervisees' experiences and knowledge" (p. 218). Supervisors might therefore choose to decenter themselves from the process of supervision, bracketing their expertise as just one way of knowing the world. However, supervisees are just as likely to want to center the therapist, to hear about his or her experiences, to engage the supervisor in everything from consultation to advocacy. Even as I pull off my cloak of privilege and power, another may be offered. This tension is clearly shown in the case illustration above. I am positioned in the session as the one in charge, and yet I question through my interventions the way we construct expert knowledge based on age and the supposed authority that comes with it.
It is a difficult balance to strike. When is it appropriate to assume the role of expert, and how can this be done without compromising the broader goal of supervision, that being to elicit the supervisee's wisdom? Evaluation is one forum in which the supervisor must center him- or herself. As Sessions (2002) explains in relation to supervision of social work students, there is a need to ensure that therapeutic treatment is done in a way that is appropriate to the needs of those receiving service, even if this means that the supervisor must advise the student on his or her suitability to practice. Arguably, the field of marriage and family therapy supervision is striving to flatten the hierarchy in relationships. However, there remains a gap between good intentions and actual practices of supervisors, with a preference for some hierarchy to remain in relationships between supervisors and supervisees (Storm et al., 2001).

2)    Cultural sensitivity. In practice, the postmodern supervisor demonstrates sensitivity to the uniqueness of each person being supervised. In the case of the therapist and the little boy discussed above, what the boy has accepted as normal in his particular context, is evidently disturbing to the therapist. Until the therapist sees that the boy's experience of the world is different than her own, she remains stumped. As is common in therapeutic work and supervision, each person brings a special culture. This same trend toward cultural sensitivity is growing among all theoretical orientations to supervision (Storm et al., 2001). Perceived this way, one might turn to Ng's (2003) work on family therapy around the world to see that practices must change to suit families, just as they must change to suit therapists. Ng succeeds in drawing out several principles that are important to attend to when working across cultures including understanding how culture influences what we perceive, starting with the key members of the family appropriate to each culture and using the therapist as a cultural mediator. But with cultural sensitivity comes its own challenges. Supervisors may not be comfortable asserting what they know in any forum in which it could be perceived that their way of practising is "the" way to practice. As Lappin and Hardy (2002) explain: "The personal comfort supervisors derived from knowing that they saw the world the same way as their supervisees has been replaced by a nervous focus on respecting difference to the point that muted unease can win out over confident supervisory declaratives" (p. 41). Clearly, with the need for more contextually sensitive supervision there also comes the challenge to assert what we know as senior clinicians in ways that do not subordinate the truth claims of those with whom we work. The terrain may be difficult to navigate, but my experience is that bracketing off my knowledge, as one truth, merely makes this sharing less oppressive and even welcome. In the above case illustration, I alternate between the authoritative role of supervisor and one more collegial in tone. Instead of making categorical claims, I preface comments with statements like: "I find that what I do fits for me personally, but it may not for you" or "This has been my experience, what has been yours?"

(3) The questions we ask elicit stories about problems and their solutions. The pattern of questioning that we use as supervisors should elicit the stories of supervisees in ways that allow those stories to be invested with power. It is not that, as the supervisor, I have a better story, rather simply a different story that explains how clinical work can be done. My expectation is that supervisees entertain my questions as part of a process that leads them to discover the best practice they can perform. As Stewart (2000) notes in his discussion of postmodern questions in supervision: "We encourage supervisees to use a lot of ideas gained from years of experience by others, but we also encourage them to travel lightly-to not let any one set of ideas or practices define the therapeutic moment" (p. 8).

As we always remain outsiders to the therapeutic process of others, there is the challenge to make our questions less instructive, eliciting others' knowledge rather than parroting our own. This balance is not typical among those who have expert knowledge in our western culture. However, as supervisors, we still carry the responsibility to oversee ethical and reasonable practice. Goldenthal (2000), for example, worries that he pushes his supervisees to do things that might be in conflict with the mandates of their agencies, but feels obliged in the interest of clients to instruct supervisees to act appropriately. Goldenthal's role as advocate, social engineer, and all-around troublemaker collides with the real-world constraints of those he supervises. In such instances, where is the appropriate line to be drawn between professional practice and more overtly political action? Should we, for example, push supervisees to politicize their work when we perceive the need for advocacy, rather than therapy in its more traditional form? Does my attempt to advocate for a small boy and his need just to "play" offer too much of a challenge to the agency structure in which the therapist works, which demands that she show evidence of therapeutic progress, which means the boy deal specifically with the "trauma" of his sexual abuse?

A postmodern approach to this problem acknowledges the multiple roles of the helper and the need, perhaps, to maintain balance. It is not, however, an argument for relativism. Although it remains in most cases the supervisee's role to evaluate the best course of intervention, in instances in which supervisors suspect gross neglect or ethical malpractice-instances in which even postmodernists would agree that principles held as "universal by consent" (Leonard, 1997) are being violated-supervisors are compelled to exert their power over the process of the supervisee's work.

(4) Relationships as vehicles to identity constructions. Anderson et al.'s (2000) study of 160 supervisees at the University of Connecticut revealed four broad dimensions of supervision valued by the study's participants: openness in the supervisory environment; respect, support, and encouragement; opportunities for personal growth; and conceptual and technical guidance. Such a list seems to demand varied definitions of the role of supervisor-definitions far broader than any one title might provide. Indeed, such findings demonstrate that the relationship between the supervisee and the supervisor is a negotiation for definition of each other's roles. Relationships provide the forum in which we define both ourselves and our experiences in positive or negative ways through the language we have and the power that we can exert over what that language comes to mean. What, we may ask, was the "trauma" the boy experienced? Was it the sexual abuse or the subsequent intervention by those trying to help him? In my work with the therapist, I perceived clearly a role for myself in helping her to consider how she was co-constructing the child as a disempowered individual, while still offering her a new co-construction of herself as having acted as a good therapist.

(5) Transparency. In supervision, I strive to make all my roles transparent. When I am performing one or the other, I might say to those I supervise, "This is why I have chosen to ask you these questions. What is your experience of what I am doing?" My own training as a reflective practitioner (Schön, 1987) is indicative of the transparency of those who have supervised me throughout my career who themselves offered such reflections. Becoming the reflective practitioner that Schön extols demands this capacity on my part as the supervisor to model transparency, enabling understanding of my own position vis-à-vis those with whom I work and the process in which I engage in as helper. Thus, this article, and similar disclosures of my process, have been shared with those I supervise.

CONCLUDING THOUGHTS
The principles listed above have support from postmodern therapists and supervisors who strive to deconstruct the meaning of their roles as supervisors and helpers. In particular, such a postmodern orientation may most benefit those practicing as supervisors outside academic contexts (Storm et al., 2001). To the extent that we are successful in adopting this orientation to intervention, we will be able to navigate our way through the various ways in which we perform as supervisors. As the case illustration demonstrated earlier, as a supervisor I move fluidly between roles, constructing my identity in any number of ways during a supervision session. Although I may assume any of a number of roles, I still fundamentally agree with Crocket (2002) and others who argue that we must strive for a supervision that breaks from patterns of domination, which centers me as the supervisor without the transparency or willingness to deconstruct my privileged position. This is a far better place from which to develop narratives of success in others. Interestingly, supervisees may be less than enthusiastic about a supervisor who does not perform as the expert. Crocket notes, following her work with one woman whom she partners with in supervision, that she sees "clearly the importance of carefully preparing counsellors to engage in supervision, not as supervised persons ... but as counsellors who invite a colleague to stand alongside them, to partner them, as they review their work. This is a supervision that does not produce counsellors at the margins, but that offers them speaking positions as they story their desires for supervision and for their work as counsellors" (p. 20).

Given that we simultaneously wear multiple hats, this postmodern positioning of supervisors as other than experts does not mean we do not lead. In fact, Winslade (2002) shows us that teaching, as a component of supervision, is still a responsibility. As he says:

I have a responsibility to introduce students to fields of knowledge and experience and stories of life that they may not have come across. This is best thought of as introducing a new person into a conversation that is well under way. It is necessary for someone who is new to a conversation to listen to what the others are talking about and to look for places where they can join in. (p. 37)

Perhaps it is possible, after all, to be both centered and decentered during the same supervision session, incorporating roles as diverse as supervisor and teacher. Ultimately, that may be the contribution of postmodern thinking to supervision, the critical perspective it brings to understanding that our roles are fluid, constructed, and based on negotiations for power. A supervisor who helps another therapist to find his or her own multiple stories, while performing the supervisor's own preferred identity story, is modeling a process of compassionate and reflective clinical intervention.

REFERENCES
Anderson, H., & Goolishian, H. (1992). The client is the expert: A not-knowing approach to therapy. In S. McNamee & K. J. Gergen (Eds.), Therapy as social construction (pp. 25-39). Thousand Oaks, CA: Sage.
Anderson, S. A., Rigazio-DiGilio, S. A., Cochran-Schlossberg, M., & Meredith, S. (2000). Four dimensions deemed essential to quality supervision: Theoretical, personal, and empirical account. In Readings in family therapy supervision: Selected articles from the AAMFT Supervision Bulletin (pp. 64-66). Washington, DC: AAMFT.
Berger, P., & Luckmann, T. (1966). The social construction of reality: A treatise in the sociology of knowledge. London: Penguin Books.
Best, S., & Kellner, D. (1997). The postmodern turn. New York: Guilford Press.
Crocket, K. (2002). Introducing counsellors to collaborative supervision. International Journal of Narrative Therapy and Community Work, 4, 19-24.
Denzin, N. K. (2001). Interpretive interactionism (2nd ed.). Thousand Oaks, CA: Sage.
Fine, M., & Turner, J. (2002). Collaborative supervision: Minding the power. In T. C. Todd & C. L. Storm (Eds.), The complete systemic supervisor: Context, philosophy, and pragmatics (pp. 229-240). New York: Authors Choice Press.
Flemons, D. G., Green, S. K., & Rambo, A. H. (1996). Evaluating therapists' practices in a postmodern world: A discussion and a scheme. Family Process, 35, 43-56.
Fox, H., Tench, C., & Marie (2002). Outsider-witness practices and group supervision, international Journal of Narrative Therapy and Community Work, 4, 25-32.
Freedman, J., & Combs, G. (1996). Narrative therapy: The social construction of preferred realities. New York: W. W. Norton.
Gaddis, S. (2004). Re-positioning traditional research: Centring clients' accounts in the construction of professional therapy knowledges. International Journal of Narrative Therapy and Community Work, 2, 37-48.
Gardner, G. T, Bobele, M., & Biever, J. L. (2002). Postmodern models of family therapy supervision. In T. C. Todd & C. L. Storm (Eds.), The complete systemic supervisor: Context, philosophy, and pragmatics (pp. 217-228). New York: Authors Choice Press.
Goldenthal, P. (2000). A matter of balance: Challenging and supporting supervisees. In Readings in family therapy supervision: Selected articles from the AAMFT Supervision Bulletin (pp. 61-62). Washington, DC: AAMFT.
Lappin, J., & Hardy, K. V. (2002). Keeping context in view: The heart of supervision. In T. C. Todd & C. L. Storm (Eds.), The complete systemic supervisor: Context, philosophy, and pragmatics (pp. 41-58). New York: Authors Choice Press
Leonard, P. (1997). Postmodern welfare: Reconstructing an emancipatory project. Thousand Oaks, CA: Sage.
Madsen, W. C. (1999). Collaborative therapy with multi-stressed families: From old problems to new futures. New York: Guildford Press.
Minuchin, S., & Storm, C. (2000). Supervisors as social engineers: Creating family therapy-friendly organizations: An interview with Salvador Minuchin. In Readings in family therapy supervision: Selected articles from the AAMFT Supervision Bulletin (pp. 59-60). Washington, DC: AAMFT.
Newman, F., & Holzman, L. (1997). The end of knowing: A new developmental way of learning. New York: Routledge.
Ng, K. S. (2003). Toward a global view of family therapy development. In K. S. Ng (Ed.), Global perspectives in family therapy (pp. 1-15). New York: Brunner-Routledge.
Protinsky, H. (2002). Dismounting the tiger: Using tape in supervision. In T. C. Todd & C. L. Storm (Eds.), The complete systemic supervisor: Context, philosophy, and pragmatics (pp. 298-307). New York: Authors Choice Press.
Schön, D. A. (1987). Educating the reflective practitioner. San Francisco: Jossey-Bass.
Sessions, P. (2002). The art of teaching. International Journal of Narrative Therapy and Community Work, 4, 62-66.
Sexton, T. L., Ridley, C. R., & Kleiner, A. J. (2004). Beyond common factors: Multilevel-process models of therapeutic change in marriage and family therapy. Journal of Marital and Family Therapy, 30, 131-150.
Sprenkle, D. H., & Blow, A. J. (2004). Common factors and our sacred models. Journal of Marital and Family Therapy, 30, 113-129.
Stewart, K. (2000). Postmodern questions. Readings in family therapy supervision: Selected articles from the AAMFT Supervision Bulletin (p. 8). Washington, DC: AAMFT.
Storm, C. (2000). The standard of care for supervision: Finding our way through the fog. Readings in family therapy supervision: Selected articles from the AAMFT Supervision Bulletin (pp. 49-51). Washington, DC: AAMFT.
Storm, C. L., & Sprenkle, D. H. (2002). The ivory tower and the institute: Supervision in educational programs. In T. C. Todd & C. L. Storm (Eds.), The complete systemic supervisor: Context, philosophy, and pragmatics (pp. 93-111). New York: Authors Choice Press.
Storm, C. L., Todd, T. C., Sprenkle, D. H., & Morgan, M. M. (2001). Gaps between MFT supervision assumptions and common practices: Suggested best practices. Journal of Marital and Family Therapy, 27, 227-239.
Todd, T. C. (2002). Privately contracted supervision. In T. C. Todd & C. L. Storm (Eds.), The complete systemic supervisor: Context, philosophy, and pragmatics (pp. 125-134). New York: Authors Choice Press.
Todd, T. C. & Storm, C. L. (2002). Thoughts on the evolution of MFT supervision. In T. C. Todd & C. L. Storm (Eds.), The complete systemic supervisor: Context, philosophy, and pragmatics (pp. 1-16). New York: Authors Choice Press.
Ungar, M. (2001). Constructing narratives of resilience with high-risk youth. Journal of Systemic Therapies, 20(2), 58-73.
Ungar, M. (2002). Playing at being bad: The hidden resilience of troubled teens. Lawrencetown Beach, NS: Pottersfield Press.
Ungar, M. (2004a). A constructionist discourse on resilience: Multiple contexts, multiple realities among at-risk children and youth. Youth and Society, 35, 341-365.
Ungar, M. (2004b). Nurturing hidden resilience in troubled youth. Toronto: University of Toronto Press.
White, M. (1997). Narratives of therapists' lives. Adelaide, South Australia: Dulwich Centre.
White, M. (2000). Reflections on narrative practice: Essays and interviews. Adelaide, Australia: Dulwich Centre.
White, M. (2002). Journey metaphors. The International Journal of Narrative Therapy and Community Work, 4, 12-18.
White, M. (March, 2004). Narrative therapy: New modalities of practice. Workshop presentation, Truro, Nova Scotia, Canada.
Wieling, E., Negretti, M. A., Stokes, S., Kimball, T., Bateman, C., & Bryan, L. (2001). Postmodernism in marriage and family therapy training: Doctoral students' understanding and experiences. Journal of Marital and Family Therapy, 27, 527-534.
Winslade, J. (2002). Storying professional identity. International Journal of Narrative Therapy and Community Work, 4, 33-38.
Yingling, L. C. (2000). What is a systemic orientation-really? In Readings in family therapy supervision: Selected articles from the AAMFT Supervision Bulletin (pp. 36-38). Washington, DC: AAMFT.
Michael Ungar
Dalhousie University
Michael Ungar, PhD, School of Social Work, Dalhousie University.
The author wishes to thank Lou Costanzo for his helpful comments on a draft of this article and his thoughtful mentorship.
Correspondence concerning this article may be addressed to Michael Ungar, PhD, School of Social Work, Dalhousie University, 6414 Coburg Road, Halifax, Nova Scotia, Canada, B3H 2A7; E-mail: Michael.ungar@dal.ca
Copyright American Association for Marriage and Family Therapy Jan 2006
Provided by ProQuest Information and Learning Company. All rights Reserved