Agency and purpose in narrative therapy: questioning the postmodern rejection of metanarrative.

Postmodern ideas have led to increased interest in narrative in the domains of both theology and psychotherapy. Narrative theology allows the concept of an intrinsic human telos, a divinely created purpose which constitutes the perfection of a human life. Socially constructed understandings of the institution of therapy, however, and postmodern ideological commitments regarding the nature of metanarratives, make it difficult for narrative therapists to consider this theological possibility. This paper will examine the concept of agency in narrative psychology and therapy, and the moral questions that exist at the boundaries of these disciplines, exploring the possible clinical significance of including a teleological assumption in their theoretical core.

**********

Abroad array of scholars and historians, both inside and outside the church, assert that the United States and much of the Western world has entered a postmodern epoch (e.g. Gergen, 1991; Grenz, 1993; Smith, 1989). Since some characteristics of modernity and its associated ideals remain, it is uncertain how widely the term "postmodern" may be applied to the contemporary world (e.g. Harvey, 1990). Nevertheless, there is general agreement that social and technological changes of the 20th century, themselves born of the modern temper, paradoxically subverted many key modernist assumptions, particularly those regarding the value of instrumental reason.

The turn to postmodern concepts can be easily traced in the field of family therapy (e.g., Held, 1995; Lee, 2001, 2002). Early theories of family therapy developed in the 1960s and 1970s were based on the mechanistic metaphors of cybernetics, and tended to place the power to create change squarely in the hands of therapists. As the experts in the counseling relationship, therapists held the responsibility to diagnose and treat the problems brought by clients. It mattered little if families understood why the sometimes arcane interventions seemed to work. The therapist was paid to make change happen, and did so in what for that time were strikingly unconventional ways.

In the 1980s, once family therapy had established itself as a legitimate discipline, several lines of criticism arose. Feminist scholars (e.g., Luepnitz, 1988), for example, decried the implicit paternalism of approaches that reestablished power hierarchies within families without considering the disparities experienced by women. An extensive debate about the nature of therapist power led many to question accepted practices. Were therapists too instrumental in their approach to families? Did their assumption of professional expertise undermine the family's own competence? Therapists needed new models that would embody a more humble approach and a greater respect for the families under their care.

Therapists began to propose new understandings of their role. Harlene Anderson and Harry Goolishian (1992), for example, recommended a "not-knowing" approach in which therapeutic expertise was replaced with a more circumspect attitude:

The not-knowing position entails a general attitude or stance in which
the therapist's actions communicate an abundant, genuine curiosity ... a
need to know more about what has been said, rather than convey
preconceived opinions and expectations about the client, the problem, or
what must be changed. (p. 29)

The therapist's role shifts from that of expert change-agent to conversation partner and consultant. This far more respectful stance brings new possibilities to light through an egalitarian process of therapeutic dialogue.

INCREDULITY TOWARD METANARRATIVES

In the realm of psychotherapy, postmodern ideas have led to an emphasis on the role of narrative in human understanding and self-perception. With this emphasis comes an ideological rejection of "metanarratives," the broader understandings embraced by a culture that form an inherited context of meaning for the more particular stories that individuals and families within the culture may tell about themselves. This intellectual shift, and the concomitant changes in the understanding of the therapist-client relationship described above, draws sustenance from the writing of postmodern critics Jean-Francois Lyotard (1979/1984) and Michel Foucault (1980), and the related perspective of social constructionism (e.g. Gergen, 1994, 1999). Harlene Anderson (1997), for example, summarizes her understanding of the intellectual content of postmodernism:

Postmodern thought ... primarily represents a broad challenge to and a
cultural shift away from fixed metanarratives, privileged discourses,
and universal truths; away from objective reality; away from language as
representational; and away from the scientific criterion of knowledge as
objective and fixed. (pp. 35-36)

Here are echoes of Lyotard's (1979/1984) definition of the postmodern temper as "incredulity toward metanarratives" (p. xxiv), while the term "privileged discourses" reflects the work of Foucault, particularly as interpreted through the writings of Michael White (White & Epston, 1990). Similarly, narrative therapists Jill Freedman and Gene Combs (1996) insist that their way of practicing therapy is not about technique, but adopting an attitudinal stance that is congruent with a postmodern, narrative, and social constructionist worldview. This stance is defined by four key ideas:

1. Realities are socially constructed.

2. Realities are constituted through language.

3. Realities are organized and maintained through narrative.

4. There are no essential truths. (p. 22)

Let us leave aside, for the moment, the ontological question of what is meant here by "reality" (q.v. Held, 1995), and focus instead on the more phenomenological and practical use to which these ideas are put. The "realities" that are of primary clinical interest are the self-perceptions clients bring to therapy. Narrative therapists maintain that clients are frequently oppressed by "problem-saturated" stories of their lives and relationships (White & Epston, 1990, p. 16). These self-understandings may have been learned from one's family of origin, where parental stories carry authoritative weight because of a child's dependency (Parry & Doan, 1994). Therapists, too, can perpetuate problem-saturated descriptions through the use of diagnostic language (Gergen, 1999).

Such descriptions (e.g., "I am depressed," "I never do anything right") are held to be socially constructed and maintained. To use a Wittgensteinian metaphor, both family life and the process of therapy, like all human interactions, take place in the context of self-sustaining language games. For the postmodern therapist, Lyotard's "incredulity toward metanarratives" translates into "questioning how we accept certain conditions, situations, and events as 'realities' in our lives ... [and] to 'unmask' these 'realities' so that we develop a clearer sense of what are the discourses (narratives) of influence" (Zimmerman & Dickerson, 1996, p. 75). To do so means separating clients from the influence of the problem-saturated stories they have learned to tell about themselves.

The best-known strategy by which this is accomplished is called "externalizing" (White & Epston, 1990). Clients view their problems as troublesome but inextricable personal characteristics. In externalizing, clients are taught to give their problems names and relate to them as if they were separate characters in their life stories. A client who says "I'm depressed," for example, may label the problem as "Depression" and begin to explore her relationship to it. In particular, therapists help the client notice any events that constitute exceptions to the dominant narrative: "Has there ever been a time in which you have refused to let Depression defeat you?" Such "unique outcomes" (White & Epston, 1990) become the basis for a collaborative process of developing a new and less oppressive storyline.

This brief overview of narrative therapy, of course, cannot do full justice to the nuances of the craft nor the differences between its practitioners. Held (1995) has argued that the postmodern and narrative turn in family therapy can be seen as an attempt to resolve a practical dilemma, a conflict between the content and process levels of therapy. On the one hand, the content of therapy must be individualized, meaning that therapists must respect the unique histories and utterances of those in treatment. A client is not simply one more opportunity for the application of the therapist's expert knowledge. On the other hand, therapists also need generalizable procedures for the process of therapy that extend across clients. Without these, the profession itself has little substance or reason for being. Therapists differ in how they balance these concerns. The not-knowing approach of Anderson and Goolishian represents the first part of the dilemma; the development of teachable techniques like externalizing represents the second. Postmodern therapists attempt to advocate both a stance and a corresponding set of practices, in such a way that the latter does not lead back to a notion of therapist privilege.

For present purposes, though, the summary is sufficient to point toward the essentially moral questions of personal agency that are the focus of this paper. The purpose of the narrative therapy literature is not to create a full philosophical or theological anthropology, but to explore how narrative concepts can help people in a therapeutic context. But how does the practical and linguistic domain of psychotherapy constrain what can be said? The question is particularly important at the boundaries of therapeutic discourse, where matters of ethics and theology reside.

This article will focus on two such boundary questions. First, how is human agency understood in the narrative therapy literature? Second, how would a teleological perspective change or complete this understanding of agency, and what would this imply clinically? Following a line of reasoning suggested by Murphy and Ellis (1996), the thesis of this paper is that narrative approaches to psychotherapy, which assist clients in reconstructing their life stories, are incomplete because they raise moral questions that require the resources of ethics and theology. It will be argued that narrative therapy operates with a restricted notion of human agency, because its ideological commitment to postmodernism makes it difficult to explore the possibility of an intrinsic human telos or purpose. Let us begin, however, with an examination of basic assumptions that both narrative therapists and narrative theologians might be able to hold in common.

THE CORE OF A NARRATIVE PROGRAM IN PSYCHOLOGY

Christian philosopher Nancey Murphy (1990) follows Imre Lakatos' terminology in his account of scientific rationality. Lakatos distinguishes between progressive and degenerating research programs. A research program consists of a "hard core" of theories protected by a belt of auxiliary hypotheses and lower-level theories that can be modified in the face of anomalous findings. A degenerating research program is one in which modifications of the auxiliary belt may preserve the core theories but add nothing new, while the modifications in a progressive program increase its empirical content and predictive power (Murphy, 1990).

In a similar vein, Murphy and Ellis (1996) suggest that ethical theories have a hard core which constitutes a "unifying vision of the good for humankind" (p. 115). Others have suggested that theories of psychotherapy also have assumptions at their core which create an ethical framework for how therapists perceive the goals of treatment (e.g. Browning, 1987; Fancher, 1995). What assumptions might we identify at the core of a narrative theory or program?

One fundamental assumption of a narrative program is that human behavior cannot be fully understood without reference to cognition and the construction of meaning. The so-called Cognitive Revolution of the late 1950s, a counter-reaction to behaviorism, was originally intended "to discover and to describe formally the meanings that human beings created out of their encounters with the world, and then to propose hypotheses about what meaning-making processes were implicated" (Bruner, 1990, p. 2). At first, the computational metaphors of information-processing that dominated the new cognitive science were functionally not far different from the stimulus-response conceptions of behavioral psychology, for they still avoided what seemed to be overly subjective notions like agency and intentionality. In an increasingly postmodern era, however, such positivist conceptualizations are beginning to give way to the more constructionist orientation of a meaning-centered psychology.

Understanding cognition in terms of meaning and evaluation opens a place for a limited form of teleology in the account of human nature. Psychologist Joseph Rychlak (1994, 2000) argues that from the 16th century onward telic explanation was increasingly unfashionable. The commitment in science to empiricist and reductionist principles edged out the search for what Aristotle termed final causes, which seemed reminiscent of the impositions of ecclesiastical conceptions that scientists wished to escape (Osmond, 1994). Scientific explanation was to emphasize efficient causality, which continues to be the rule in psychology; in such a climate, telic explanations may be perceived as a form of "spiritualism" (Howard, 1987).

Not all telic explanation, however, is intrinsically religious. Rychlak (1994) makes a helpful distinction between human teleology and deity teleology. The latter entails a specifically theological claim of the operation of divine purpose and intention. Rychlak himself remains theologically neutral. His cognitively-oriented Logical Learning Theory (LLT) presupposes instead a human teleology, which suggests "that people behave for the sake of reasons, purposes, and intentions rather than solely in response to the impulsions of efficient causation" (p. 8). Human behavior cannot be understood solely in terms of passive responses to antecedent events; people cognitively evaluate situations and make decisions that are oriented toward specific goals.

Rychlak coins the term telosponsivity to capture this final-cause account of the relationship between behavior and cognition. As a psychological construct, its rival is not deity teleology, but mechanism and an exclusive bias toward efficient cause explanation. Human teleology and deity teleology are only competing types of final cause explanation if one rules out larger questions of theology and human destiny in an a priori fashion.

The first core assumption, then, is that humans must organize experience meaningfully, which entails a telic relationship between cognition and behavior. A second assumption is that this cognitive organization is implicitly narrative in form. Humans appear to be biologically predisposed (suggesting Aristotle's categories of formal and material cause) to form narrative meaning structures (Newberg & D'Aquili, 2002). Indeed, some evidence suggests a neurological link between the inability to tell and/or understand stories (dysnarrativia) and handicaps in the development of a sense of selfhood (Bruner, 2002).

Meaning structures, in turn, are shaped by participation in the symbolic systems of a given culture (Bruner, 1990). Cultures possess what Bruner calls a "folk psychology" in which everyday explanations of behavior, especially deviations from social norms, are given in narrative form: "Folk psychology is about human agents doing things on the basis of their beliefs and desires, striving for goals, meeting obstacles which they best or which best them, all of this extended over time" (p. 42-43). Narrative accounts of human behavior, in other words, have an intrinsically (human) teleological structure.

Research in other fields of psychology also supports this understanding. In discourse psychology, for example, Graesser, Olde, and Klettke (2002) have shown that readers automatically construct "situation models" when attempting to comprehend textual material:

The situation model is a microworld that includes a spatial setting,
agents in pursuit of goals, and causal chains of events that unfold
chronologically. These microworlds are systematically constructed
through mechanisms that are widely accepted in contemporary theories in
cognitive science. (p. 258)

There are, in other words, preexisting expectations about wording and syntax, plot structures, and the like, and violations of these expectations in the text will affect comprehension, reading speed, and recall. Graesser and his colleagues observe that despite its relative complexity, narrative is easier to understand and remember than expository text. Researchers, however, "have not yet provided a satisfying answer to the question of why narrative has such a privileged status in the cognitive system" (p. 240). Attempts at explanation may come from different quarters than discourse psychology, such as the brain science of Newberg and D'Aquili (2002) already cited above. In the meantime, the preeminence of narrative in human understanding and meaning-making remains a core assumption.

These are essentially anthropological assertions, which in themselves are compatible with both narrative theology and narrative therapy. By its very nature, however, the social institution of psychotherapy includes additional commitments that reshape how narrative assumptions function within it. For example, therapy usually occurs in an office setting where money is exchanged for counseling services. This demonstrates a realistic disparity of power in the reciprocally defined social roles of therapist and client. But given the changes described earlier in how this relationship has been understood in recent decades, postmodern and narrative therapists wish to embody more egalitarian forms of practice. They are reluctant to impose meaning structures on their clients, including the identification of a preexisting telos or goal. In turn, this ideological commitment shapes a particular understanding of human agency.

HUMAN AGENCY IN NARRATIVE THERAPY

In a 1987 interview, Rychlak described agency as an ability to "influence things over and above your biological inheritance and your sociocultural setting" (Howard, 1987, pp. 14-15). What is quintessentially human is "the capacity to set the course of nature" (p. 15). Agency, in other words, entails not only goal-directed intentionality, but the effectiveness of that intentionality over against other constraints and causal factors.

As we have seen, for narrative therapists, such constraints include inherited but problematic discourses, metanarrative frames of understanding that restrict client freedom. Zimmerman and Dickerson (1996) provide an interesting albeit brief discussion of human agency. They write:

The question of "agency in a person" disconnected from contexts of
meaning and relationship makes little sense to us. The question that
fits better for us is this: What marginalized experiences speak of an
alternative truth?... In this sense, "agency" means that each of us can
play a part in choosing what to let guide us and in what contexts ...
This means working with clients to notice previously unnoticed aspects
of their experience, occurring in either the landscape of action or the
landscape of consciousness, that reflect personal agency in a preferred
direction. (pp. 93-94)

The therapist's task is to ask questions that produce a sense of agency in the client. Agency with regard to the "landscape of action" means that clients perceive their lives as a historical sequence of events in which their own decisions and actions have causal efficacy: they can do something that makes a difference. Helping clients achieve agency with respect to the "landscape of consciousness" or the "landscape of identity" means locating their actions with respect to their goals and intentions: not only can they make a difference, but their actions can be directed intentionally toward a particular desire or value.

Such an understanding of agency, of course, is not unique to narrative therapy. Snyder (Snyder, Harris, et al., 1991; Snyder, Sympson, et al., 1996), for example, has developed a self-report measure of hope, a construct comprised of two cognitive components dubbed agency and pathways:

First, we hypothesize that hope is fueled by the perception of
successful agency related to goals. The agency component refers to a
sense of successful determination in meeting goals in the past, present,
and future. Second, we hypothesize that hope is influenced by the
perceived availability of successful pathways related to goals. The
pathways component refers to a sense of being able to generate
successful plans to meet goals. (Snyder et al., 1991, p. 570)

In some studies, the Hope Scale has been utilized to operationalize a resilience factor. Horton and Wallander (2001), for example, examined mothers caring for children with chronic physical conditions, and found that hope buffered the relationship between the severity of caregiving demands and symptomatic distress. Such findings are of theoretical and clinical interest, for they demonstrate in a tangible way the importance of a cognitive sense of effective goal-directed agency.

As suggested earlier in the discussion of folk and discourse psychology, this kind of agentive intentionality is intrinsic to well-formed narratives. Gergen (1999), for example, states that Western standards of narrative structure require that stories come to a "point":

An acceptable story must first establish a goal, an event to be
explained, a state to be reached or avoided, or more informally, a
"point." This point is typically saturated with value; it is understood
to be desirable or undesirable. (p. 69)

Elements of the story are chosen and ordered in such a way as to provide some explanatory justification as to why a particular endpoint was or was not achieved. Our very understanding of ourselves and others is set in such narrative frames:

[W]e are often called upon to "tell our story," to recount our past, to
identify where we have been and where we are going.... To make sense of
our lives, we typically posit some kind of endpoint or goal ("how I came
to be X," "achieve Y," or "believe in Z"). (p. 70)

Thus, the positive content of a narrative concept of agency assumes cognition about and action toward a desired endpoint, extended over time. But is this merely a psychological concept, limited to, in Rychlak's terminology, human teleology? Is there also a transcendent teleological dimension to agency?

From the perspective of ethicist Alasdair MacIntyre, this is the central moral question for modernism, and a fortiori, postmodernism. At one level, he views human action as telic, meaning that it can be described "as flowing intelligibly from a human agent's intentions, motives, passions, and purposes" (MacIntyre, 1984, p. 209). At a broader level, however, MacIntyre takes the Aristotelian view that a human goal-oriented teleology derives its moral logic from a more transcendent understanding of human nature.

In his widely read book After Virtue, MacIntyre (1984) observes that the contemporary world is in a state of moral disarray and fragmented traditions. He argues that this is the inevitable result of the ill-conceived Enlightenment project of grounding morality in abstract reason, which rejected a teleological view of human nature. From an Aristotelian perspective, it is the concept of telos that makes sense of the moral life, which views humans as moving through the present to their true end. Without this, ethics becomes incoherent. What remain are rules without context, an abstract set of prohibitions that do not lead toward some final goal of perfection.

Without teleology, it becomes impossible to answer the question: What is it to live well? What is the nature of a truly good life? As Murphy and Ellis (1996) have written:

to complete the task of the social sciences, it is necessary to have an
answer to the question of the ultimate meaning of human life, or to use
a less ambiguous term, of the final purpose or telos of human life....
[I]f we had an answer to the question what is the telos of human life,
we would be able to judge among life stories and conclude that certain
characteristics, practices, achievements, constitute a good life, or at
least that some ways of life or characteristics are better than others.
(pp. 87, 105)

One may argue, of course, that various schools of psychotherapeutic practice have in fact operated according to some more or less articulate vision of what constitutes a good life. Browning (1987), for example, argues that theories like psychoanalysis, humanistic psychology, and Skinnerian behaviorism all involve "instances of religio-ethical thinking" that qualify them as "practical moral philosophies" (p. 8) for the ordering of the interior life. It is precisely this element, however, that the postmodern commitments of narrative therapy seek to eliminate.

As suggested earlier, narrative therapists hold a limited goal-oriented view of agency, which is its positive content. There is also, however, a negative content: agency is not only intentional movement toward preferred goals, it is a movement away from or out of subjugating discourses. This is the relevance of the technique of externalization, which is

helpful in the interruption of the habitual reading and performance of
[problem-saturated] stories. As persons become separated from their
stories, they are able to experience a sense of personal agency; as they
break from their performance of their stories, they experience a
capacity to intervene in their own lives and relationships. (White &
Epston, 1990, p. 16)

Narrative practitioners recognize that there are many potential sources for subjugation through the social construction of identity: the race, gender, and class politics of society; the restrictive stories of one's family-of-origin; even the theoretical biases of therapists themselves.

For this reason, narrative therapists give a privileged status to client preferences. Much of therapy is devoted to questions which elicit a client's preferred view of self, which includes "how they would like to behave, how they would like to see themselves, and how they would like to be seen by others" (Eron & Lund, 1996, p. 44). Therapist and clients together explore and articulate the client's own positive goals, but always against the negative background of subjugating influences:

Since we co-construct alternative stories a bit at a time from
experiences that do not fit with dominant, problematic stories, it is
important that therapists check frequently to be sure that the direction
or meaning of these experiences is preferred to that of problematic
stories. This point may seem academic but we, at least, are not always
right about what different people actually prefer. (Freedman & Combs,
1996, p. 129)

Two observations are appropriate at this point. First, the privileging of client preferences stands as an example of what MacIntyre (1984) called the doctrine of emotivism, in which "all evaluative judgments and more specifically all moral judgments are nothing but expressions of preference, expressions of attitude or feeling, insofar as they are moral or evaluative in character" (pp. 11-12). It was Nietzsche (1887/1996) who drew the conclusion that the failure of the attempt to justify morality on rational grounds meant that morality was nothing more than the expression of a will to power. The Foucaultian version of this intellectual heritage makes narrative therapists leery of imposing their own judgments, leaving only emotivist criteria for evaluating the direction of a client's life.

The second observation is an ironic one. Though narrative therapists seek to help clients escape the dominance of problem-saturated stories, there is a larger sense in which the language game of psychotherapy itself presupposes problem-saturated discourse. Clients enter therapy because they perceive themselves as needing the help that a therapist can provide, and this assumption in part defines the reciprocal roles of the social institution of psychotherapy. Narrative therapists may eschew the instrumental power tactics of early family therapy, but they still bring a general understanding of the goals of therapy that assumes some problematic element. This preunderstanding is the postmodern and constructionist stance that narrative therapists take toward the stories that clients have inherited and internalized.

There are therefore two logical levels at which we might speak of problem discourse. One is the client's understanding of his/her problems, the latter is a metadiscourse whereby therapists problematize the former. But must problem-saturated stories be a problem? Is it possible that problem-saturated stories at the level of client discourse need not be problematic at the metadiscourse level? To answer this question requires a brief reconsideration of a key tenet of postmodern therapy: the incredulity toward metanarratives.

METANARRATIVES REVISITED: HOW INCREDULOUS SHOULD WE BE?

Earlier, we noted Lyotard's (1979/1984) dictum that postmodernism entails "incredulity toward metanarratives," a phrase which appears in the introduction to a slim monograph entitled The Postmodern Condition: A Report on Knowledge. His concern there was with the epistemological question of how advanced societies legitimate knowledge when previously taken-for-granted metanarratives have already lost their authority.

The term "metanarrative" needs some qualification. Lyotard (1979/1984) had in mind the grand and comprehensive stories that anchor a society's large scale institutions. The grand story that the people of a nation must have knowledge to know true freedom, for example, legitimates institutionalizing the pursuit of knowledge. In practice, however, there is no simple dichotomous distinction between narratives and metanarratives, but layers of stories that are logically meta to other layers. I may tell a story about an episode that happened at work today. The story may point to a larger narrative context, namely, my ongoing struggles at this job. This in turn may take meaning from how I understand my work history, the events of my life, and even the question of whether or not life itself has meaning.

Does incredulity entail some form of suspicion toward all of these meaning levels? The answer would seem to be this: since all such stories of any level of generality are socially constructed, they are not ultimate. In other words, stories do not so much represent an objective truth as the will of some group expressed through language. Narrative therapists cannot grant authority to any metanarrative, lest they compound the oppression under which clients already live with their problem-saturated stories.

Some therapists go further, writing as if the elimination of metanarrative authority were already an accomplished fact: "In the demise of all grand narratives, we now live in a world in which personal narratives essentially stand alone as the means by which we pull together the text of our own lives" (Parry & Doan, 1994, p. 25). In this vacuum, therapists must help clients to understand that "there are no other yardsticks of stories or persons against which to measure the legitimacy of their own stories" (p. 27).

In such a view, a transcendent teleology is impossible; only emotivism remains. This creates, however, an intrinsic conceptual conundrum. If metanarratives are to be questioned as subjugating discourses, linguistic expressions of the will to power, on what basis can we treat the narrative actualization of client preferences as a moral good to be achieved through the therapeutic process? Are these somehow not expressions of the will to power? At the extreme, we may end up following Nietzsche into a nihilistic form of moral heroism: if power is all there is, use it unflinchingly and don't apologize. But if this is the terminus of the argument, how could one denigrate the instrumentality of cybernetic therapists?

The easy criticism, of course, is that the postmodern interpretation of history is itself a metanarrative which self-referentially and incoherently legitimates the rejection of metanarratives (e.g. Middleton & Walsh, 1995). As philosopher Peter Kreeft (1999) has noted, there is great irony in the fact that so-called "absolutists" admit that some moral truths may be relative, while moral relativists are absolute in their relativism.

Beyond this general point, however, we should observe that narrative therapists may not be quite ready to abandon the idea of a moral good. Metanarrative discourse is not entirely absent from the therapy process, and may even serve a legitimating function. Consider the following case, drawn from a narrative therapy textbook:

[A] therapist from a training program was seeing a black man in his
mid-30s who had come in because of a pervasive experience of depression.
He reported that he had been "wrestling with this demon" since he was a
teenager and that it "spoke to him of his identity." The supervising
therapist had some idea that the depression might well have been an
effect of the oppressive practices of racism, so when she went into the
therapy room at a prearranged break to consult with the therapist, she
raised some questions about how the therapist might talk with the man
about the effects of racism.... When she left the room and went back
behind the one-way mirror, the man's first remark was, "You mean
depression might not be about my identity?" He went on to say that
previously he had no hope, but now he could see hope in his future.
(Zimmerman & Dickerson, 1996, p. 74, italics in original)

The supervisor's intervention reflects the postmodern use of a hermeneutic of race, gender, and class oppression. Clinically, the intervention succeeds in both externalizing and transforming the problem, which gives the client a hopeful sense of agency that did not previously exist. This is consistent with the idea that externalizing "opens [a] space for therapists to consider and ask about the larger sociopolitical contexts and discourses that support problems" (Freedman & Combs, 1996, p. 283).

The question, however, is this: what metanarrative assumptions does the supervisor presuppose? Neither the therapist nor the client had raised the issue of racism. The supervisor keeps to the form of non-hierarchical and collaborative practice, and "situates" (Freedman & Combs, 1996) her comments appropriately as personal curiosity rather than the authoritative edicts of a professional expert. Nevertheless, in order to break the hold of the dominant narrative of depression, the supervisor introduces what for her is an authoritative metanarrative, thereby relocating the client's local story in the larger frame of racism.

Did this supervisor view her reframing of the problem as merely potentially helpful, or somehow "right"? It seems unlikely that the suggestion of racism was a neutral clinical hypothesis; the reality of oppression is a moral evil in postmodern thought, just as freedom from oppression is a moral good.

There are two issues here. One is the accuracy of her intuition. It is quite possible and even likely that the client had internalized the language games of a dominant white majority that led him to view his difficulties as strictly personal as opposed to social in origin. The other issue is the matter of her implicit moral commitment: racism is wrong. It is her commitment to postmodern ideology that leads her to privilege the metanarrative of racism in the first place, whatever incredulity she may attempt to maintain. The implications are that (a) some metanarratives can have a privileged status; (b) therapists can introduce these into the therapeutic dialogue even if they don't arise directly from client preferences; and (c) this can produce a positive clinical effect.

Suspiciousness toward the negative effects of metanarrative can be liberating, but this stance cannot and should not be absolute. The above case study demonstrates that, in point of fact, it is not absolute in practice. One might inveigh against the insidious power of subjugating discourses, but it is illusory to think that such freedom can ever be ultimate, because metanarratives are inescapable, and we must cede authority to any metanarrative to which we ascribe positive truth-value. Participation in human society is participation in language and thus in socially constructed forms of ordering. Rychlak (1994), for example, argues that social norms be understood "as predications held in common by a people rather than as some kind of supraindividual (efficiently caused) 'force in the sky' that patterns or shapes behavior in a mediational fashion" (p. 290). He continues:

According to LLT, a norm is a predication affirmed in common rather than
an impulsion from the group to the individual ... LLT appreciates the
tremendous role that social forces play in human experience, but these
forces obtain their influence from the fact that the person affirms
their meaning and intends that they be enacted. (pp. 296-297)

Similarly, Gergen (1999) suggests that there are limits to the kind of critique suggested by Foucault's work:

To reject all that Foucault might call "disciplining" or "ordering"
would be to erase virtually all that we value.... Rather than wholesale
rejection of all forms of ordering, we might adopt a posture of
differentiating appraisal.... In what ways does a tradition sustain that
which we hold to be good, in what ways does it fail? How could it be
otherwise? (p. 41)

To their credit, White and Epston (1990) do in fact recognize this limitation: "Foucault does not propose any alternative ideology ... Neither does he suggest that it is possible to 'deny' knowledge, that is, to act apart from and experience the world from outside the mediating effects of knowledge and discursive practices" (pp. 26-27). But then, if we are to follow Gergen's suggestions, on what basis do we engage in this "differentiating appraisal"? Absent teleology, how do we judge the ability of a tradition to maintain what is good?

It is beyond the scope of this paper to answer such questions in full; the purpose has been to open a space for teleological reflection in narrative practice. A postmodern stance views truth as local and socially constructed, but relies upon assertions that must themselves function as grand narratives. We cannot completely escape the constitutive power of discourses, as if it were possible to stand outside language and the human community. To problematize metanarratives requires an evaluative judgment that must rely on some other criterion, and this too is open to the same discursive critique. The way is open, therefore, to ask if it is possible that there are positive metanarrative commitments, particularly those regarding the telos of human life?

IS THERE A CLINICAL PLACE FOR TELEOLOGY AND THEOLOGY?

Incredulity, of course, does not necessarily entail an automatic rejection of the authority of any and all metanarratives. Inasmuch as narrative therapists encourage clients to articulate their preferred values and goals, metanarratives and their teleological assumptions are bound to emerge. Using emotivist criteria at least, therapists will lend support to metanarrative constructions that clients do not experience as oppressive.

But earlier, the question was raised: must problem-saturated stories be a problem? Even postmodern therapists make a metanarrative assumption: that clients' own problem stories should be taken as problematic at the therapists' level of discourse as well. This is not necessarily incorrect, but hopefully, this essay has succeeded in suggesting that there may be alternatives.

In particular, we might ask: is it possible that in some cases the path to wholeness lies not in the freedom from subjugating discourses, but further subjugation to true discourses, even ones that I do not in some sense "prefer"? Certainly the ascetic tradition within Christian spirituality would agree; the very notion of spiritual discipline requires submitting one's bodily existence to practices oriented toward a divinely given telos (Willard, 1988). Would narrative therapy with religious clients recognize the authority and possible value of their religious traditions and practices? If such a client expressed a religiously-oriented telos as a preferred outcome, would or should the therapist support this direction? If a client with strong religious background prefers to reject her tradition, would there ever be a place for a therapist to suggest a movement in the opposite direction, on the premise that the pursuit of freedom from her tradition is not necessarily in her best interest?

Research on the role of spirituality in well-being raises interesting questions in this regard. Froma Walsh (1998), for example, cites religion and its transcendent narratives as a positive aspect of family resilience in the face of adversity; such metanarratives provide ways of making meaning out of difficult circumstances. Summarizing the relevance of the research on spirituality, Walsh writes:

We clinicians may need to listen for what has not yet been expressed,
and to show comfort and respect in exploring the spiritual domain as it
is experienced. It is useful to draw out multiple stories and
dimensions, deconstructing harmful aspects while encouraging new
possibilities. For instance, some family members may hold an old
childhood image of an all-powerful God as harsh and punitive, often
connected to experiences of shame and helplessness when confronted with
autocratic authority figures. We can invite stories of other spiritual
experiences that have been positive, or even moments of inner peace,
communion, and nurturance. These can be drawn upon and expanded. (p. 73)

Note that Walsh's clinical recommendations fit the practice principles of narrative therapy: deconstruct the harmful while listening for unique outcomes, upon which positive alternatives can be constructed. Religion and spirituality do not represent inherently oppressive metanarratives, though religious symbolism may be experienced this way because of its association with other psychological experiences. But the example also highlights the presumed boundaries of psychotherapeutic practice, for the scenario raises essentially theological questions about God's nature and our relationship to him. Victims of the "harsh and punitive" God of childhood suffer the effects not only of psychological maltreatment but bad theology.

These are matters not only of theological ethics but professional ethics. Therapists, of course, are not theologians, even though theological assumptions may slip in unawares. At the very least, even the postmodern therapist need not be limited to exploring unique experiential outcomes (e.g. "moments of inner peace"). It would be within the bounds of ethical practice to inquire after unique theological elements in a client's story. Was the client ever taught anything different about the nature of God? Where did s/he see evidence that this alternative view might be valid? How does s/he make sense of these discrepant teachings? The therapist might then refer the client to a trusted pastoral colleague, perhaps even one in the same tradition in which the client was raised, who can help sort out the theological issues.

This serves once more to raise the question of how the norms of professional practice create a discursive context which constrains theological (or teleological) discussion. As we have seen, the boundaries are unclear. By what authority does a supervisor raise the issue of racism, or consider it to be a moral evil? Racism is not judged to be wrong simply by dint of its psychological effects on a person's life. It is a violation of a fundamental human right--a notion that requires a view of agency that transcends client preferences.

If this is so, then a narrative understanding of agency would benefit from a more explicit consideration of telos, which in turn will require the resources of ethics and theology. For Christians, this should come as no surprise. As White (2003) has written: "Our faith is rooted in not the present but the future. It is based on not our single story but the wider narrative of the cosmic saga" (p. 133). The Bible is the "dwelling place of the imagination" (Lindbeck, 1989, p. 43), where biblical stories reveal a world in which God is active in human history. Just as the portrayal of human life in a well-written novel impacts our own life stories, so too do biblical narratives help readers to imagine new possibilities (Barton, 2000).

This includes the ability to situate one's own life in a transcendent narrative context. Brueggemann (1986), reflecting on the notion of "calling" in the life of the prophet Jeremiah, writes that the substance of a call assumes the particularity of God who has a purpose for the life of this individual:

Such a sense of call in our time is profoundly countercultural, because
the primary ideological voices of our time are the voices of autonomy:
to do one's own thing, self-actualization, self-assertion,
self-fulfillment. The ideology of our time is to propose that one can
live an "uncalled life," one not referred to any purpose beyond one's
self. (p. 19, emphasis added)

Calling comes from without, not within. Biblical narratives help put our lives in context--in teleological context--and the biblical witness suggests that this enables us to bear suffering through the encouragement of hope (e.g. Rom 5:1-5; Phil 3:4b-14).

History gives examples of the oppressive use of biblical authority, and therapists are right to be wary in this regard. Allowing teleological assumptions into the therapeutic process guarantees nothing about the appropriateness of their use. Christians who wish to open a path for dialogue must begin with a theological conviction that may provide common ground with those committed to a postmodern stance: the Bible is a profoundly liberating metanarrative, revealing a God who has compassion for those who suffer and are oppressed. In the words of Middleton and Walsh (1995), the biblical metanarrative is "counterideological" and "antitotalizing" (p. 87; see also Schneiders, 1989), constantly undermining its domestication into institutions of power.

Narrative therapy, with its limited teleological conception of agency, provides an important challenge to treatment models based on the notion of efficient cause. By breaking the hold of problem-saturated stories and subjugating discourses, it opens a place for clients to articulate and strive toward preferred goals. But in the face of the postmodern erosion of moral traditions, therapists need to reflect on the possibility of a transcendent human telos. As narrative theologian George Lindbeck (1989) has stated: "Postmodern men and women do not easily survive in what for them is a cold and empty universe clothed only in the abstract jargon of sociological and psychological reason" (p. 53). Brueggemann (2000) describes the historical challenge thus:

There was a time, a very long time, when the assumption of God
completely dominated Western imagination ... In the seventeenth century,
it was hard, courageous work to imagine-consequently reimagine-the world
without God. And now, into the twenty-first century, in the face of
Enlightenment autonomy issuing in autonomous power and autonomous
knowledge, it is hard, courageous work to imagine-consequently
reimagine-the world with God. (pp. 1, 2)

Christian therapists must think creatively beyond the limits of their professional paradigms to reimagine the world as a place where God desires to move suffering and imperfect humans toward their created end.

REFERENCES

Anderson, H. (1997). Conversation, language, and possibilities: A postmodern approach to therapy. New York: Basic.

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). London: Sage.

Barton, J. (2000). Disclosing human possibilities: Revelation and biblical stories. In G. Sauter & J. Barton (Eds.), Revelation and story: Narrative theology and the centrality of story (pp. 53-60). Aldershot, England: Ashgate.

Browning, D. (1987). Religious thought and the modern psychologies. Philadelphia: Fortress.

Brueggemann, W. (1986). Hopeful imagination. Philadelphia: Fortress.

Brueggemann, W. (2000). Deep memory, exuberant hope. Minneapolis, MN: Fortress.

Bruner, J. (1990). Acts of meaning. Cambridge, MA: Harvard.

Bruner, J. (2002). Making stories: Law, literature, life. Cambridge, MA: Harvard.

Eron, J. B., & Lund, T. W. (1996). Narrative solutions in brief therapy. New York: Guilford.

Fancher, R. T. (1995). Cultures of healing: Correcting the image of American health care. New York: W. H. Freeman.

Foucault, M. (1980). Power/knowledge: Selected interviews & other writings, 1972-1977 (C. Gordon, Ed.). New York: Pantheon.

Freedman, J., & Combs, G. (1996). Narrative therapy: The social construction of preferred realities. New York: Norton.

Gergen, K. J. (1991). The saturated self. New York: Basic.

Gergen, K. J. (1994). Realities and relationships: Soundings in social construction. Cambridge, MA: Harvard.

Gergen, K. J. (1999). An invitation to social construction. London: Sage.

Graesser, A. C., Olde, B., & Klettke, B. (2002). How does the mind construct and represent stories? In M. C.

Green, J. J. Strange, & T. C. Brock (Eds.), Narrative impact: Social and cognitive foundations (pp. 229-262). Mahwah, NJ: Lawrence Erlbaum.

Grenz, S. J. (1993). Revisioning evangelical theology. Downers Grove, IL: InterVarsity.

Harvey, D. (1990). The condition of postmodernity. Cambridge, MA: Blackwell.

Held, B. S. (1995). Back to reality: A critique of postmodern theory in psychotherapy. New York: Norton.

Horton, T. V., & Wallander, J. L. (2001). Hope and social support as resilience factors against psychological distress of mothers who care for children with chronic physical conditions. Rehabilitation Psychology, 46, 382-399.

Howard, G. S. (1987, September). Toward a telic vision of human beings: A dialogue with Joseph F. Rychlak. Journal of Counseling and Development, 66, 12-19.

Kreeft, P. (1999). A refutation of moral relativism: Interviews with an absolutist. San Francisco: Ignatius.

Lee, C. (2001). The postmodern turn in family therapy. Christian Counseling Today, 9(3), 16-19.

Lee, C. (2002). Family systems epistemology: The evolution of a perspective on persons and relationships. In T. Speidell (Ed.), On being a person: A multidisciplinary approach to personality theories (pp. 243-295). Eugene, OR: Wipf and Stock.

Lindbeck, G. A. (1989). The church's mission to a postmodern culture. In F. B. Burnham (Ed.), Postmodern theology: Christian faith in a pluralist world (pp. 37-55). New York: Harper and Row.

Luepnitz, D. A. (1988). The family interpreted: Feminist theory in clinical practice. New York: Basic.

Lyotard, J-F. (1984). The postmodern condition: A report on knowledge (G. Bennington & B. Massumi, Trans.). Minneapolis, MN: University of Minnesota. (Original work published 1979)

MacIntytre, A. (1984). After virtue. 2nd ed. Notre Dame, IN: University of Notre Dame Press.

Middleton, J. R., & Walsh, B. J. (1995). Truth is stranger than it used to be: Biblical faith in a postmodern age. Downers Grove, IL: InterVarsity.

Murphy, N. (1990). Theology in the age of scientific reasoning. Ithaca, NY: Cornell.

Murphy, N., & Ellis, G. F. R. (1996). On the moral nature of the universe: Theology, cosmology, and ethics. Minneapolis, MN: Fortress.

Newberg, A., & D'Aquili, E. (2002). Why God won't go away: Brain science and the biology of belief. New York: Ballantine.

Nietzsche, F. (1996). On the genealogy of morals (D. Smith, Trans.). Oxford: Oxford University Press. (Original work published 1887)

Osmond, D. H. (1994). A physiologist looks at purpose and meaning in life. In J. M. Templeton (Ed.), Evidence of purpose (pp. 133-167). New York: Continuum.

Parry, A., & Doan, R. E. (1994). Story re-visions: Narrative therapy in the postmodern world. New York: Guilford.

Rychlak, J. F. (1994). Logical learning theory: A human teleology and its empirical support. Lincoln, NE: University of Nebraska.

Rychlak, J. F. (2000). A psychotherapist's lessons from the philosophy of science. American Psychologist, 55, 1126-1132.

Schneiders, S. M. (1989). Does the Bible have a postmodern message? In F. B. Burnham (Ed.), Postmodern theology: Christian faith in a pluralist world (pp. 56-73). New York: Harper and Row.

Smith, H. (1989). Beyond the postmodern mind. New York: Quest.

Snyder, C. R., Harris, C., Anderson, J. R., Holleran, S. A., Irving, L. M., Sigmon, S. T., et al. (1991). The will and the ways: Development and validation of an individual-differences measure of hope. Journal of Personality and Social Psychology, 60, 570-585.

Snyder, C. R., Sympson, S. C., Ybasco, F. C., Borders, T. F., Babyak, M. A., & Higgins, R. L. (1996). Development and validation of the State Hope Scale. Journal of Personality and Social Psychology, 70, 321-335.

Walsh, F. (1998). Strengthening family resilience. New York: Guilford.

White, J. E. (2003). Embracing the mysterious God. Downers Grove, IL: InterVarsity.

White, M., & Epston, D. (1990). Narrative means to therapeutic ends. New York: Norton.

Willard, D. (1988). The spirit of the disciplines. San Francisco: HarperSanFrancisco.

Zimmerman, J. L., & Dickerson, V. C. (1996). If problems talked: Narrative therapy in action. New York: Guilford.

CAMERON LEE

Graduate School of Psychology

Fuller Theological Seminary 

AUTHOR

LEE, CAMERON. Address: Graduate School of Psychology, Department of Marriage and Family, Fuller Theological Seminary, 180 N. Oakland Avenue, Pasadena, CA 91101. Title: Professor of Family Studies. Degrees: AB, University of California, Berkeley; MDiv, PhD, Fuller Theological Seminary. Specializations: family life education; clergy family research; narrative approaches to integration.

Correspondence concerning this article may be sent to Cameron Lee, PhD, Graduate School of Psychology, Fuller Theological Seminary, 180 N. Oakland Avenue, Pasadena, CA 91101. Email: [e-mail]