Making Trouble for Problems:

Therapeutic Assumptions and Research Behind the Narrative Practice of Externalizing Conversations




Feel free to contact me regarding this paper, A clinical research project submitted to the faculty of Argosy University/San Francisco Bay Area Campus in partial fulfillment of the requirements for the degree of Doctor of Psychology in Clinical Psychology. Point Richmond, CA September 2005

This clinical dissertation by Carl F. Hilker III has been approved by the committee members signed below who recommend that it be accepted by the faculty of Argosy University/San Francisco Bay Area Campus in partial fulfillment of the requirements for the degree of DOCTOR OF PSYCHOLOGY.


Clinical Dissertation Committee:

Mary Herget, Ph.D.


Jim Sparks, Ph.D.



Table of Contents


Chapter One: Study Formulation and Definition


Problem Statement

Purpose Statement

Significance and Clinical Rationale

Chapter Two: Methodology

Summary of the Study

Interpretive Process

Literature Sources

Definition of Key Terms

Chapter Three: Literature Review

Therapeutic Assumptions of Externalizing

Externalizing Problems

Objectifying and Personifying a Problem

Reducing Blame for a Problem’s Existence

Tracking the History of a Problem

Deconstructing Sociocultural and Political Effects/Assumptions

Creating Space to Consider Possible Alternatives

Allowing for the Creation of Alternative Stories

Building Personal Agency over Preferred Qualities

Not Assigning Causation to Problems

Making Deadly Issues More Discussable

Acting as an Oppositional or Anti-Language

Researching the Narrative Therapy Approach

Establishing a Narrative Perspective on Research

Impact of the “Interpretive Turn.”

Reconciling research with narrative values.

The effects of political and social views about therapy.

Influence of Narrative-Inspired Co-Research

Reciprocal connection between therapy practice & research methodology.

The use of externalizing practices in co-research.

Types of Research into Narrative Therapy

Evaluating narrative therapy in general versus specific practices.

Research on externalizing practices.

Chapter Four: Findings and Discussion

Evaluation of Externalizing Practices

Clinical Implications of this Review

Conclusions about Externalizing Practices

Where Would Narrative Therapy be Without Externalizing?

Analysis of Current Research


Implications for the Future

Future Research Methods

Proposals for Research





The narrative approach to therapy outlined by David Epston and Michael White is broadly premised on the assumption that people’s lives are socially constructed through the multiple stories they tell, and that these stories are colored by the cultural and historical contexts within which they occur. The process of externalizing is a narrative therapy practice that establishes a context where people experience themselves as separate from the problems in their lives. Such practices focus on the relationship between the person and the problem instead of upon a problem-person. The person is not the problem; rather the problem is the problem. More than a therapeutic technique, externalizing practices assist in exemplifying the social constructionist worldview of narrative therapy.

Accounts of the usefulness and therapeutic value of externalizing practices come from practitioner descriptions and anecdotal client accounts of the therapy process. Efforts to utilize methodological research for studying the therapeutic qualities of externalizing practices have been limited. Contributing factors to this dearth of data include the unique and sometimes complex social constructionist principles of narrative therapy, the political and social sensibilities of those among the narrative therapy vanguard, and researchers’ comparative lack of training in qualitative research methodology.

There are recent signs within the narrative therapy community and the qualitative research field which suggest that novel forms of method-based research involving collaboration between researcher and participant can yield promising results. This study proposes that research methods that employ the narrative practice of co-research offer excellent opportunities to evaluate the therapeutic qualities of externalizing conversations.


Chapter One: Study Formulation and Definition


Narrative therapy is a relative newcomer to the field of talk therapy – having first taken form in the late 1980s. In fact, many different themes have been emphasized in what is understood as the narrative metaphor, including ways of understanding people’s identities and vantages from which to understand problems and their effects on people’s lives (Morgan, 2000). Narrative therapy can also refer to particular ways of talking to people about their lives, problems, constraints, and their reflections on social, cultural, and political contexts.

The purview of this study is restricted to that version of narrative therapy chronicled by Michael White and David Epston. Their groundbreaking book, Narrative Means to Therapeutic Ends (1990), represents an initial comprehensive effort to illustrate their conceptualization of narrative therapy as well as the therapeutic effects of its operation. Since that time, a torrent of literature – by these and other practitioners; has continued to evolve our understanding of the meaning and uses for the narrative therapy which White and Epston are widely considered founders.

Grounded in postmodern and social constructionist modes of thinking about the world, the narrative approach or “stance” can be recognized by many of its characteristic qualities. One of its most pervasive and distinct qualities is the practice of externalizing . In stressing the point of its importance, consider Michael White’s response to a question posed to him regarding narrative therapy (Epston, 1993, p.161); “Michael, if you hadn’t invented that ‘talk’ we refer to as an externalizing discourse, where do you think your work would be today?” White’s reply was a conclusive, “Nowhere!”

Externalizing practices create a context in which problems are treated as separate from people. People are invited to discuss problems and constraints in a manner that separates them from these issues. This process is quite unique within a Western culture that values a scientific medical model that situates problems within individuals’ biology and character. Problems are conceptualized from the narrative vantage point “as existing outside people as a linguistic device to reorganize our thinking in ways that counter shame and blame, minimize defensiveness, and promote client agency or the capacity to act on behalf of themselves in relation to problems” (Madsen, 1999, p. 169).

However, the process of externalizing is more than simply a stylish therapeutic technique for talking with clients about problems. Externalizing conversations are a hallmark of the theoretical and therapeutic foundation of the narrative approach. It is conceived as a philosophical way of thinking about people, their problems, and informs the attitude by which therapists continually engage clients in conversation. Consider the following observation offered by O’Hanlon (1994):

I have to give you a warning; if externalization is approached purely as a technique, it will probably not produce profound effects. If you don’t believe, to the bottom of your soul, that people are not their problems and that their difficulties are social and personal constructions, then you won’t be seeing these transformations. When Epston or White are in action, you can tell they are absolutely convinced that people are not their problems. Their voices, their postures, their whole beings radiate possibility and hope (p.28).

Informed by philosophical ideas from writers such as Jacques Derrida and Michel Foucault, narrative therapy assumes a non-traditional stance where notions of reality are not rooted in positivist notions about the existence of “deep structures” to self-identity or “essential truths” about people. Such thinking is often referred to as structuralism.

Narrative therapy is, alternatively, informed by poststructuralist ideas. Poststructuralism refers to a movement away from the idea that there are deep or real structures in people; e.g. a real self, a true self, which can be discovered by experts. This outlook resists the search for essential truths about people and questions notions of therapist “objectivity” and “expertise” in therapeutic work.

Moreover, narrative therapy operates under the social constructionist assumption that “problems are anchored and supported by cultural discourses or taken-for-granted cultural prescriptions about how we should act” and make sense of our lives (Madsen, 1999 pg. 171). The way people understand themselves and their relationship to the world is developed and shaped by their interactions in the world.

Externalizing practices presuppose such ideas about culture and its influence on reality and self-identity, and actively seek to engage people in conversations that allow identified characteristics (particularly problems) to be seen as external to oneself. Externalizing discourse represents a manner of speaking that summons a social constructionist perspective and allows for the creation of alternative life narratives that open doors to an expanded array of choices in our lives.

As narrative therapy becomes a more established therapeutic approach, an increasing body of literature seeks to address the therapeutic qualities of narrative practices. This study specifically seeks to draw attention to the purported therapeutic qualities of externalizing practices and to critique the nature and depth of research that has been conducted to date using such practices.

Problem Statement

Narrative therapy is reaching into its second decade on the scene of psychotherapy and its community of adherents is expanding. Moreover, narrative practitioners and writers regularly profess the therapeutic qualities of narrative practices such as externalizing.

Yet, how can these therapeutic qualities best be understood? How persuasive are the arguments concerning its effectiveness? What type of work is conducted with clients that conveys that such practices are indeed beneficial to people? And lastly, how solid is the narrative-inspired research (often called co-research) that makes use of externalizing practices?

As the body and breadth of narrative-based literature accumulates, there appears to be a growing emphasis on validating the therapeutic effectiveness of narrative practices (such as externalizing), particularly as they relate to certain types of problems people encounter in their lives (such as eating disorders, domestic violence, depression, etc.).

A distinctive quality of the therapeutic stance of narrative therapy lies in the manner in which it borrows from the postmodern philosophical tradition. Similarly, narrative-based research appears to be unique from other types of research in the mental health professions. Conventional forms of research often search for empirical conclusions (as with quantitative designs) and typically rely on the researcher’s expert role to collect, codify, and interpret data (as is the case with many forms of qualitative methodology). Narrative-inspired research, conversely, focuses on the subjective nature of experience and seeks to de-emphasize the therapist’s role as an expert. In particular, the narrative enterprise of co-research is often employed as a research method that is not only used to expound on the therapeutic qualities of the externalizing process, but is conducted with clients in an externalizing manner of speaking ( Epston, 1999).

But, does the unique style of narrative-inspired research support the therapeutic claims made by its adherents? Is the research compelling enough to draw the interest and respect of those in the mental health community not accustomed to the narrative tradition? Is there more conventional research literature that supports the therapeutic claims by narrative therapy? Should it even be a goal of narrative-based researchers to ground their research in widely accepted design methodology?

Purpose Statement

The purpose of this study is to take an in-depth critical look into the narrative practice of externalizing; focusing intimately on its various uses and purported therapeutic qualities.

A considerable amount of literature exists on the various ways in which externalizing practices function. Much of the literature is sprinkled with various tantalizing therapeutic qualities attributed to externalizing practices, often within the context of how externalizing serves other narrative practices; such as deconstructive listening, relative influence questions, discovering unique outcomes, etc. An externalizing stance is additionally inherent to the narrative practice of co-research.

However, the uniqueness of this study is in the singular manner in which it focuses on the various ways in which externalizing practices are employed within narrative therapy and the extent to which they are evaluated through research. With this aim, the study answers the following research questions:

  1. What does the narrative therapy literature maintain as the salient therapeutic qualities and values attributed to the practice of externalizing?
  2. In what manner and depth has methodological research attempted to validate the therapeutic qualities of externalizing practices?

Significance and Clinical Rationale

Narrative therapy is an increasingly popular orientation within the field of psychotherapy; and while some practitioners in psychology seek to restrict the title of “theoretical orientation” to only a few approaches typically founded in the early 20th century, the narrative-inspired orientation arguably has a rich, albeit recent tradition grounded in social constructionist ideas about the human condition.

Narrative therapy proposes many innovative ideas and therapeutic practices. Key among them is the practice of engaging people in externalizing conversations. Sometimes viewed as simply a therapy technique, externalizing practices address wider implications about how clinicians view the world, understand clients’ problems, and by association, how clients view themselves and their ability to make changes in their lives.

While narrative-inspired texts discuss externalizing practices to varying degrees, this study is distinguished by the manner in which it focuses singularly on the use of and therapeutic value associated with externalizing practices. By doing so, this study provides a richer understanding of the theory and practice of externalizing approaches to clinicians experienced in narrative work, as well as to those new to the narrative approach. It also offers an opportunity to explore in-depth the types of research behind narrative therapy and externalizing practices.

An evolving body of narrative-inspired literature and research is being conducted that may assist in placing narrative therapy alongside other documented and researched therapeutic approaches. An objective of this study is to further that literary process, specifically as it relates to the practice of externalizing. This study’s uniqueness lies in the manner in which it focuses solely on the issues and therapeutic implications surrounding externalizing practices. By doing this, it represents novel work in the field of narrative therapy literature, and will assist in explaining a key narrative practice to the larger clinical psychology community.


Chapter Two: Methodology

Summary of the Study

This study assembles and illustrates the therapeutic qualities attributed to the narrative therapy practice of externalizing, reviews current research literature on narrative therapy and externalizing practices, and critically appraises the manner in which narrative-inspired research makes use of externalizing discourse . The overall effectiveness of this study derives from a comprehensive survey and critical analysis of the current literature.

Interpretive Process

This study comprises a comprehensive survey of the existing literature on the narrative therapy practice of externalizing and provides a critical evaluation of the research literature conducted on narrative-inspired methods. A final chapter will combine viewpoints from existing literature, as well as my own clinical interpretation, concerning the effectiveness of the narrative community in sharing its research practices with the larger mental health community in a compelling manner.

Literature Sources

In following the approach outlined above, this study has drawn from a wide range of scholarly-based literature. Specific sources include the following:

  • The primary literary works of Michael White and David Epston.
  • Literary text from other narrative and social constructionist-inspired writers and clinicians including, but not limited to Gene Combs, Jill Freedman, Victoria Dickerson, Jennifer Freeman, Richard Maisel, Gerald Monk, Alice Morgan, William Madsen, Karl Tomm, Jeffrey Zimmerman, etc.
  • Databases used to acquire published psychological articles for this study include, PsychInfo, Proquest Direct, InfoTrac, and Digital Dissertations.
  • Journal articles were also retrieved from various libraries including Argosy University and Alliant International University.

Key descriptor words for this study include, externalizing conversations/practices, narrative therapy, narrative therapy and research, and externalizing practices and research.

Definition of Key Terms

The following are definitions to key terms used throughout this study. These definitions are also cited as footnotes in the text to assist the reader.

Constructivism ; A term based on the idea that knowledge is constructed by a person’s prior knowledge as well as new personal experiences. Initially embraced by some in the therapy community, constructivism eventually was critiqued as being too relativistic and insufficiently attentive to the ways reality is also constructed through culture, class, race, etc.

Discourse – For the purposes of this study, the term will reflect that described by Hare-Mustin (1994, p.19-20), whereby discourse is “a system of statements, practices, and institutional structures that share common values…. The ways most people hold, talk about, and act on a common, shared viewpoint are part of and sustain the prevailing discourses.” Discourses have a powerful influence on how they allow for and shape people’s life stories, and they typically reflect established structures of social and power relationships. Examples of discourse include those about normative standards (body image, success, self worth, manhood, etc.), professionals’ role as experts, and discourses about pathology (Freedman & Combs, 1996).

Ethnographic research; Research relying on the notion that one can describe what people think by analyzing what they say (Fetterman, 1989). Ethnographic interviews are typically audiotaped, transcribed verbatim, and then analyzed to identify emergent themes and categories among the many words people use to describe a given experience. Within the narrative therapy application of ethnographic research, clients are actively engaged in the interpretive process of the meanings of their words and experiences of therapy.

Externalizing conversations ; this term can refer to the practice of externalizing the problem. Yet, this study also uses the term to encompass the pervading narrative therapy stance held with clients that regards people’s sense of self, reality, and problems as socially constructed. This stance is a distinctive element of the narrative therapeutic alliance between therapist and client.

Externalizing practices; Those efforts by the therapist to engage a client in conversations that seek to separate a problem, constraint, or internalized quality from the client’s own identity. Such practices allow for the notion that “the person is not the problem, rather the problem is the problem.”

Grounded theory – A methodology that attempts to develop theories that are grounded in the research participants’ own experiences. The role of the researcher is to reflect on the participants’ experiences in a manner that allows these experiences to be organized and categorized. The goal is to develop a coherent body of knowledge or theory that is derived from participant descriptions (Gaddis, 2004).

Qualitative research – Qualitative research does not rely on statistics to show confidence in the results. Confident conclusions are based more on the findings and interpretations that are produced by rigorous and systematized observation and documentation of the particular phenomenon in question (Maione & Chenail, 1999). Qualitative research is a general term used to describe a group of methodologies dedicated to the description and interpretation of social phenomena (Maione & Chenail, 1999). Such methodologies include, but are not limited to, ethnography, discourse analysis, narrative analysis, grounded theory, phenomenology, frame analysis, hermeneutics, and conversation analysis.

Quantitative research – Research that focuses on measuring and counting facts and the relationships among variables, and that seeks to describe observations through statistical analysis of data. It includes experimental and non-experimental research and descriptive research ( i.e. research that attempts to describe the characteristics of a sample or population).

Social constructionism ; Refers to the notion that a person’s sense of reality is malleable, and shaped largely by social and cultural forces mediated by language. The focus is on the processes (i.e. discourses) by which people come to describe, explain, and account for the world and their position in it (Gergen, 2001).

Therapeutic – The American Heritage Dictionary defines therapeutic as, “having healing or curative powers.” This study uses the word therapeutic in that context, and explores both the claims and research that speak to the “healing and curative powers” of externalizing conversations.

Unique outcome – Anything that does not fit with the dominant problem-saturated story. For example, if a client’s problem is social anxiety, a unique outcome could be an instance when the anxiety was overcome or simply didn’t present itself as a problem at that time. From a narrative perspective, it is important that the client perceive an incident as a unique outcome. If imposed by the therapist, the client may have misgivings about the incident truly being a unique outcome.

Chapter Three: Literature Review

Please note that the literature review presented below is a curtailed version from the original presented to my dissertation committee.

The goal of this portion of the chapter is to expand upon some of the operating assumptions regarding the therapeutic value of externalizing practices, and to critically review the manner in which narrative practices have been researched, with specific focus on externalizing conversations.

Therapeutic Assumptions of Externalizing

Certain assumptions or presuppositions are made within many intellectual enterprises in general and more specifically surrounding the artfulness and creativity of therapy and what may be considered therapeutic. A narrative therapist’s questions to a client often rest upon assumptions as to how both the question and answer will be useful to the client.

Therapists’ work with clients is driven by what is assumed to be therapeutic for a client at a given point in time. These therapeutic assumptions spring forth from a multitude of wellsprings; one’s past work with similar clients, information imparted by “experts” in the field, data borne from research, etc. The notion of what is therapeutic not only sustains the therapist during each session, but provides the lifeblood of the entire therapy enterprise in its continued refinement of “standard of care” and the evolution of the field.

Within the realm of narrative therapy, the implementation of externalizing conversations has taken a defining role. In fact, many of the major applications of narrative therapy; whether it be working with certain problems (such as eating disorders, depression, domestic violence, children acting out, etc.) or how to achieve specific results (such as discovering unique outcomes , deconstructing problem stories, or building alternative stories) involve the use of externalizing conversations. In many respects, the process of externalizing is the workhorse of narrative therapy.

But, why and how does externalizing work? What aspects of externalizing conversations do clients of narrative therapy find useful and therapeutic? And how well has it been thought through by those who assert its therapeutic usefulness?

In relatively informal ways, many narrative-inspired practitioners have utilized feedback from clients. For example, White notes that when soliciting feedback from those who consult him about their experience in therapy, “very often it is the engagement with externalizing conversations that is identified as a turning point for them in their efforts to satisfactorily address their problems and concerns” (White, 2000, pp.3-4). Specifically, he mentions that externalizing conversations made it possible for people to “separate their sense of identity” from the previous problem-saturated accounts of themselves (White, 2000).

The remainder of this section explores in-depth several of key therapeutic assumptions and elements of the narrative practice of externalizing. They are presented as follows:

  1. Externalizing problems
  2. Objectifying and personifying a problem
  3. Reducing blame for a problem’s existence
  4. Tracking the history of a problem
  5. Deconstructing sociocultural and political effects/assumptions
  6. Creating space to consider possible alternatives
  7. Allowing for the creation of alternative stories
  8. Building personal agency over preferred qualities
  9. Not assigning causation to problems
  10. Making deadly issues more discussable
  11. Acting as an oppositional language or anti-language

While not an exhaustive list, these elements are among the most frequently cited by narrative-inspired authors as the therapeutic qualities that render externalizing conversations useful. These qualities are not characterized as occurring in isolation from one another, but are seen as intertwined and interdependent. This section illustrates these assumptions in detail. A subsequent section critiques the thoroughness to which these elements have been found to be useful in therapy.

Externalizing Problems

When people turn to therapy for assistance, they have often reached a point where they believe there is something wrong with them; that they or something about them is problematic (Carey & Russell, 2002). By this point, the problem has become internalized. People assume the problem is a facet of their nature or inner-self. The problem has become fused with the person’s identity.

The narrative stance assumes that people can most easily examine the effects of problem-saturated stories of their lives when a discussion is conducted via externalized conversations. Externalizing establishes a context where people experience themselves as separate from the problem. Such a conversation shifts the client’s relationship to problems, and shifts the conversation to a focus on the relationship between the person and the problem instead of a singular focus on a problem-person.

Much of the literature suggests that the first step in the externalization process is to use the client’s language to describe the problem, modifying it so that the problem is objectified, and then questioning the objectified problem about its intentions and goals for the client. This process allows the problem to become a separate entity and thus external to the person or relationship – decoupling the connectedness between the person(s) and the problem (White & Epston, 1990).

This process often begins simply. For example, during an initial session a client may comment that he feels he has a problem with making major decisions in his life. The consulting therapist may ask externalizing questions such as: So, “indecision” has made it difficult for you to arrive at certain decisions in your life? Has long has “indecision” been influencing your life? What strategy does “indecision” employ to get you off-kilter and unable to make up your mind? The therapist in this example has used the term indecision in an externalized manner, and will determine if this or another term fits best for the client.

Clients often do not need a lengthy explanation of the externalizing process. Rather, the therapist simply engages the client in externalizing conversations in a comfortable and conversational manner. If the therapist maintains this linguistic stance during the consultation and asks externalizing- inspired questions, clients will begin to experience a difference between how the therapist is talking about the problem and how they have likely been thinking about the problem for a considerable time. As the therapist’s use of externalizing language takes hold with the client (which may take any number of sessions), the client may begin to think about his problem or constraint in a different fashion; one where he slowly becomes separate and distinct from the problem. Therapeutic opportunities now begin to emerge.

Michael White found the process of working with people to externalize their problems compelling in the way it helped in the struggle against these problems. In the landmark book co-authored with David Epston, he notes (White & Epston, 1990, p. 39):

I have concluded that, among other things, this practice [externalizing the problem]:

  1. Decreases unproductive conflict between persons, including those disputes over who is responsible for the problem;
  2. Undermines the sense of failure that has developed for many persons in response to the continuing existence of the problem despite their attempts to resolve it;
  3. Paves the way for persons to cooperate with each other, to unite in a struggle against the problem, and to escape its influence in their lives and relationships;
  4. Opens up new possibilities for persons to take action to retrieve their lives and relationships from the problem and its influence;
  5. Frees persons to take a lighter, more effective, and less stressed approach to “deadly serious” problems; and
  6. Presents options for dialogue, rather than monologue, about the problem.

Objectifying and Personifying a Problem

The idea of personifying the problem is closely linked to the quality of creating space. Yet, it does deserve its own discussion. Narrative therapy is quite distinct from other modalities of treatment, such as psychodynamic therapy, in the manner in which it strives to create a therapeutic setting for clients where problems are indeed experienced as something external to them. Creating a conversation where the problem has its own persona can be vital to experiencing it as external. Clothed with a discrete set of intentions, beliefs, agendas, and expectations that may be distinct from those of the client, the problem suddenly begins to take on a life of its own. Freedman and Combs (1996) also note that as the problem becomes externalized, it begins to be the focus of objectification rather than the person.

This process may indeed assist in creating space for experiencing new choices, but the narrative literature suggests that the therapeutic value of personification lies in how it assists a client in experiencing the problem as separate from her ; the realization that the problem may have beliefs and intentions that are not in the best interests of the client. Faced with this realization, she must decide how to respond.

Will the problem continue to make decisions for her or will she? Whose agenda is more important, hers or the problems? Has the problem lived up to its promises to her (if any where made)? What does the problem want for her in the future? What does she desire? How does the problem try to trick her into buying into its way of life rather than hers?

Such questions as these are considered useful in helping to separate the person from the problem(s). What questions like these have in common is their personifying and objectifying attitude about the problem. The problem is brought to life by the implicit notion that it has its own consciousness and purpose.

Reducing Blame for a Problem’s Existence

People who consult therapists often feel a sense of blame or shame for the problems they suffer. Indeed, many cultures hold value in people taking responsibility for their problems, and if a little blame and shame is felt along the way, that may be supported by many cultural conventions as well.

Narrative therapy takes a different tack. While certainly important for people to take responsibility for addressing problems in their lives, the narrative approach attempts to decouple the effects of blame and shame from the act of taking responsibility for one’s problems .

Madsen (1999, p.179) comments that blaming clients for their problems “never seemed particularly helpful.” He further elaborates that,

The process of externalization allows for the separation of blame and responsibility. By thinking of problems as external entities that invite, coach, or trick us into particular responses, we can remove blame from individuals. People are not to blame for their difficulties; problems are to blame for their difficulties. Problems can be seen as inviting particular ways of being. However, externalizing conversations also allow space for clients to consider whether they might want to decline such invitations. Questions can be asked that represent irresistible invitations to responsibility. We can promote responsibility by irresistibly inviting it rather than by demanding it. (p.179).

With blame and shame for a problem’s existence decoupled from the desire to take responsibility for one’s actions, the narrative literature maintains that therapeutic space is opened up whereby a person can feel more empowered to create and choose preferred ways of responding to the influence of a problem or constraint.

Tracking the History of a Problem

Tracking the history of a problem is also referred to as mapping the problems influence. Typically taking place within the initial session with a client, the process may also proceed throughout the course of the therapy as new experiences come to light. Tracking the problem’s history is often seen as a two-step process that lays the foundation for the discovery of unique outcomes and the co-construction of alternative life stories for the client.

Once a problem is named and separated (i.e. externalized) from the person, a narrative therapist may ask questions that inquire into the history of the identified problem. This history can include stories from the distant past, the last few months, or just a few hours prior. The goal is to establish a baseline for understanding the historical influence of the problem over the person’s life (Morgan, 2000).

It is important to maintain an externalizing stance when tracking the history of the problem. The nature of questions is such that the problem(s) remain separate from the person consulting the therapist. In addition, the conversation may use the identifying term for the problem (such as Self-Doubt or Imposed Silence). This way, the interaction between the person and the problem is more clearly visible. Such questions might include: When did you first notice the problem entering your life? At what point in the last year was the problem the strongest in your life? When the problem is influencing you, what are its requirements for how you should act or think?

When tracking the history of the problem in someone’s life, people often assume the problem maintains a static effect over time. Michael White developed the use of relative influence questions for assisting persons in externalizing the problem and tracking the history of the problem (White, 1986a). He often uses such questions during an initial interview to immediately engage the person in the activity of separating their lives and relationships from the problem(s).

An example of how relative influence questions operate is when a therapist asks a client to think about their life over the past year in terms of a percentage. The therapist may then ask: “Over the past year, what percentage of the time was negative thinking in charge of your life, and what percentage were you in charge?”

Such questioning may inquire about last month, last week, or maybe just before the client entered the therapist’s office; basically anytime during the past. The intent is to get the client thinking about the times and the degree to which the problem was in control of the person, as well as those times when the problem’s influence on the person’s life was relatively weaker.

When tracking the history of problems, White (1986a) posits that it is important to break relative influence questions into two modes of inquiry. The first type encourages people to map the influence of the problem in their lives and relationships. These questions aid people in identifying the problem’s sphere of influence across various interfaces; their behaviors, emotions, beliefs, attitudes, etc. When the problem’s influence is understood across multiple interfaces in one’s life, greater opportunities become available for discovering unique outcomes where the problem’s influence may be weak. These newly discovered areas can represent the framework for affirmative action and an increased sense of personal agency (White & Epston, 1990).

Once the influence of the problem has been identified, the stage is then set for the next mode of questioning – inviting invites clients to map their own influence in the life of the problem. These questions often bring forth information that challenges the prevailing problem-saturated story and assists people in discovering competences and skills that heretofore had been neglected under the duress of the problem.

An example of this might be a discussion in which a person recalls an incident(s) where she overcame her “public speaking fear” and was able to stand up in class and present. Questions might be asked such as; What quality was it that forced the “speaking fear” to take a back seat and put you in charge of what you wanted to do? Why was it important for you to not let the “speaking fear” take charge during this particular incident(s)?

Tracking the problem in a two-staged process affords the benefit of gathering a solid history about the problem as well as searching for clues to a person’s non-static relationship with and influence over the problem. The effect of these two modes of questioning is to encourage a person to move… out of a fixed and static world, a world of problems that are intrinsic to persons and relationships, and into a world of experience, a world of flux. In this world, persons find new possibilities for affirmative action, new opportunities to act flexibly (White & Epston, 1990, p. 42).

The new information that may come from the above-described discussion often results in the discovery of “unique outcomes”; neglected moments where a person was in control of the problem or contexts where the problem was not present. These unique outcomes can then be used to piece together alternative stories that are less problem-saturated and represent new opportunities in people’s lives.

Deconstructing Sociocultural and Political Effects/Assumptions

Earlier in this chapter, it was discussed how the narrative stance implicates cultural and historical influences in contributing to problems and constraints affecting people’s lives. By giving consideration to those cultural and political forces involved in shaping our identity, awareness can be gained about how life stories are shaped by broader cultural stories/traditions. As a consequence, new understandings of life that are influenced less by self-blame and internalized weaknesses can be constructed.

A significant component of externalizing conversations is that considerations beyond the individual can be taken into account. People’s relationships with problems tend to be shaped by history and culture. Externalizing conversations can explore the broader implications of how gender, race, culture, sexuality, class, and other relations of power have influenced the construction of a problem(s) (Carey & Russell, 2002).

In this context, externalizing allows for a deconstructive process that challenges those beliefs and ideas that maintain problems and are often viewed as taken-for-granted “truths” or cultural “common sense” knowledge. White (1995) states, “I believe that, although internalizing conversations obscure the politics of experience, externalizing conversations emphasize the politics of experience” (p. 24).

From a narrative stance, clients are invited to engage the therapist in conversations that unpack these stories and see them from different perspectives, so that the manner in which they have been constructed may be apparent (Freedman & Combs, 1996). Attention is always paid to the context in which stories exist (particularly problem-dominated stories) and the ideas and beliefs that help maintain their influence.

Exposing the politics of how identity is formed, externalizing conversations are viewed as providing clients a language in which to protest the influence of broader cultural discourses. Carey and Russell (2004) and Madsen (1999) further this idea through their discussion of how externalized language is utilized in the narrative practice of re-authoring conversations. These conversations involve the co-construction (i.e. between therapist and client) of preferred narratives that assist in addressing those constraints that brought the client to seek counseling. An externalizing stance is maintained as the preferred story is re-authored.

What do such questions look like? The conversations and particularly the deconstructive-inspired questions asked by narrative therapists have a distinct flavor. They are geared to assist the client in exposing the underlying cultural and political assumptions supporting dominant ways of thinking. The therapist listens for, and seeks to inquire about the assumptions that are suspected of serving the interests of the problem. Morgan (2000) offers some questions a narrative therapist might ponder and ask a client, including:

  • What are the underlying assumptions that enable the problem story to make sense?
  • What are some of the taken-for-granted ways of thinking and acting that are assisting the life of the problem?
  • What are some of your beliefs about a parent’s role in raising children?
  • How did these ideas develop?
  • How comfortable are you with these ideas that you inherited?
  • Which of these ideas do you find helpful or possibly restrictive in the way you would like to see your relationship with your daughter?
  • Thinking about those ideas you find helpful, what would I have seen you doing in the past that would have told me that these ideas were important to you back then?

Thinking about and asking these questions set the stage for a deconstructing conversation about the effects and assumptions of culture and politics in the life of the problem. In addition, the externalizing stance is claimed to afford the client increased flexibility for experiencing new possibilities for acting in relation to the problem.

Creating Space to Consider Possible Alternatives

Creating space via externalizing practices is one of the most universal therapeutic effects written about in the narrative-based literature (White & Epston, 1990; White, 1995; Freedman & Combs, 1996; Madsen, 1999; Maisel, Epston, & Borden, 2004). Moreover, the sense that an opening is created between a person and his problems is often one of the initial effects first noticed when engaged in externalizing conversations.

The idea behind the concept of creating space is that clients can gain room for the consideration of alternative stories and/or action (Morgan, 2000). Negative identities tend to have a fixed, unremitting quality to them; often leaving people little breathing space to view alternative stories and new courses of action. In such circumstances, externalizing conversations are used to help create space for the client to “experience an identity that is distinct or separate from the problem” (White, 1995, p. 23), as well as consider alternatives to the problem story.

Interestingly, when confronted by externalized questions, it often becomes evident that negative identity conclusions are only “thinly” developed (White, 1995, 2004). The social and cultural stories that perpetuate their troublesome existence often escape direct challenge. Externalizing conversations allow for the unpacking of these thin identity conclusions (White, 2004). In fact, White (2004) comments,

I believe that one of the primary achievements of externalizing conversations is this unpacking of the thin conclusions that people have about their own and about each other’s identity. In this activity, these conclusions are deprived of the truth status that has been assigned them; these conclusions cease to carry the authority that they did (p. 124).

A working assumption in much of the narrative-inspired literature is that people typically embrace this process and find it therapeutic by allowing them the space to think and act in more preferred ways. According to White, “In follow-up sessions, when we have discussed with persons their experience of the externalizing of the problem, many have remarked that it had the effect of ‘freeing’ them to act independently of the problem” (White & Epston, 1990. p. 66).

The literature also suggests that a common response to externalizing conversations has been a sense of relief; relief that a person is not the problem and that there are ways of getting more in touch with other, less problem-dominated narratives that speak about who he/she is (Carey & Russell, 2002). The creation of therapeutic room to think and act contributes to shedding light on other aspects of a person’s life that the problem(s) had been historically obscuring.

Allowing for the Creation of Alternative Stories

Conventionally, experience is thought of as simply what happens, and people store experiences as they happen and retrieve them through memory. The narrative process, on the other hand, is informed by the notion that experience is colored and shaped by the meaning people assign to each experience and that it is attended to (or not) as it seems relevant to the stories people are living. As such, when narrative questions are asked, the assumption is not that people can retrieve particular experiences with particular predetermined meanings. Rather, narrative questions allow the therapist to co-author experience with the client. Narrative questions do place a “spin” on a client’s experiences – beginnings and endings of experiences are suggested or highlighted, and portions of experiences are expanded upon in great detail while other segments are given shorter shrift (Freedman & Combs, 1996).

The intended goal is that listening with an externalizing attitude has a powerful deconstructive effect. From the therapist’s perspective, such a stance biases the therapist to interact differently with people than she would if she saw them as in possession of intrinsic characterological problems. An externalizing stance creates a different “receiving context” for the client’s stories, one in which the therapist can work with problems without viewing the client as pathological (Freedman & Combs, 1996).

When one is working with alternative stories of people’s lives, if a client mentions a particular character trait as if it is inherent to them (e.g. It’s my determination that helped me get the job.), this then becomes an opportunity to invite the person into an externalizing conversation that can lead to a thicker, more developed description of this trait. Unpacking the trait of determination can aid in learning about its history and how it is linked to helpful personal skills and knowledges (Carey & Russell, 2002). This unpacking can lead to further development of other preferred qualities, as well as enriching preferred stories in a person’s life.

Building Personal Agency over Preferred Qualities

The narrative literature asserts that externalizing conversations don’t simply focus on de-labeling people and their problem(s). Such conversations also take place with regards to preferred qualities as well (such as competence or determination). However, just as competence is a cultural creation (from a poststructuralist perspective), it is possible for the narrative therapist to ask questions about how this quality came to exist, its effects, what it says about the person, what other personal qualities it is linked to, and where it is most and least prominent in the person’s life. This process can make qualities such as competence more meaningful and relevant in a person’s life (Carey &Russell, 2002).

In addition, once a problem and its specific components have been externalized, clients can be invited to notice opportunities to take action against the externalized problem (Tomm, 1989). In an early paper by Michael White, he addressed the problem of schizophrenia with a client, asking the following narrative-inspired questions designed to internalize personal agency (White, 1987):

If it was possible to do so, would you like to limit the influence that schizophrenia has been on your life?… Can you see how schizophrenia has been coaching you into withdrawing and avoiding people?… How did you manage to defy schizophrenia’s instructions to avoid people and come to this meeting today?… What do you imagine this might tell you about your ability that you might not otherwise have noticed?… In what other ways have you stood up for yourself and not let schizophrenia push you around?… How ready are you to take a further step against the withdrawal habit that has such a grip on you?… Would you prefer to be a weak person with a strong habit or a strong person with a weak habit?… When you submit to schizophrenia’s efforts to push you into an unreasonable position, how does this invite your parents to do all the reasoning for you? (p. 19)

Such a line of questioning is a dramatic departure from what might be considered more traditional ways of working with a schizophrenic client. The client is being challenged to think about his own relationship with schizophrenia; as if the disorder was something distinct or external to him. The aforementioned questions are designed to be reflexive; meaning they allow the listener to ponder her/his role in the meaning of the question. From a narrative stance, a therapeutic assumption is that the very nature of these questions infuses the notion that the client indeed has choices, and that the client is an active agent in the course of her own life; even when facing challenges such as those involving schizophrenia (Tomm, 1989).

When questions strike a meaningful chord with a client, the questions’ implied notion of individual choice is internalized as part of the patient’s evolving identity. This can also constitute a new and powerful experience which can be used as part of a burgeoning alternative story. Such therapeutic conversations are designed to foster personal empowerment and lead to a heightened sense of personal agency.

In their article on exploring externalizing practices with families that have difficulty internalizing their own competencies, Kahle and Robbins (1998) illustrate how externalizing the success a family experiences can be useful in inviting the family to internalize their own competencies. They view the externalization of family successes as an additional use for externalized conversations; one less discussed in narrative literature. Whereas the process of externalizing the problem involves making the problem a separate entity from the client in order to depersonalize the problem, Kahle and Robbins (1998) suggest that,

Externalizing the success is an associated concept that is employed in order to counter another type of depersonalizing. Some families are not willing or able to take responsibility for victories over the problem. When asked questions designed to elicit the meaning that they have made about these victories, some families respond by providing explanations that negate their own involvement in these unique outcomes. In other words, the reasons the family provides for these unique outcomes are often externalized and, thus, depersonalized (p. 63).

Externalizing successes is seen as a supplement that can be useful to assist clients in the process of internalizing personal agency. It is not meant as a replacement for externalizing problems (Kahle and Robbins, 1998). Externalizing successes is illustrated in the following example:

A family consulting a narrative therapist about their problematic relationships with one another engages in externalizing conversations about the effects of anger and disapproval in their lives. As the effects of these problems are mapped out, the family discovers that there have been times when the problems have not been present or their influence has been minimized. In discussing why the problems are absent or reduced at certain times, the family may be reluctant or unable to articulate qualities within themselves that have assisted in overcoming the problems’ influence; instead attributing these unique events to chance or external events such as “It was a sunny day and we all felt happy”. At this point the therapist may believe the family members have depersonalized their victories over the problems. In response, the therapist might invite family members to entertain ideas of personal agency via a new externalizing conversation; one focusing on the qualities that led to the victory. Questions might be asked that get the family to ponder their behavior when the problems were absent; e.g. “Johnny, what did you notice about how you were feeling when the anger had less of a grip on you?” Johnny may respond, “I think I was being patient with my sister instead of reminding her of all the mistakes she usually makes.” If other family members agree on the role of patience in stemming the anger and disapproval, the therapist may chose to externalize the successful concept of patience in further discussions with the family. The therapist may ask, “What is happening when you are being patient? Where did this idea of patience come from?” and so on. The intention is to invite the family into externalizing conversations about the role and effects of the desired quality of patience so as to assist family members in the process of internalizing personal agency over the problems of anger and disapproval.

Not Assigning Causation to Problems

Externalizing conversations are typically thought of as benefiting the client. For instance, externalizing practices create a forum where people can become less identified with their problems or constraints, and alternative stories are allowed to take shape (Freedman & Combs, 1996; Madsen, 1999). Yet, externalizing practices also provide a unique perspective on problems and opportunity for the therapist as well. Amundson, Webber, and Stewart (2000) point out that,

Externalization “passes over” having to ask a problem from where it comes, avoiding the investiture of a problem with too much theory or falling into the danger of “believing leading to seeing.” To speak of the problem as a thing in itself permits us the most powerful ethical position we [therapists] can assume: to not inadvertently cooperate with any definition of the Other, no matter what the psychological, sociological, or political correctness offered by a particular symptom or theoretical system happens to be (p.27).

The practice of psychology often lends itself to theorizing about the origins of clients’ problems; with or without their consent or agreement. For instance, systems theory may find causation within the family system; whereby the problem or symptom needs to exist in order to maintain the family homeostasis. In psychoanalysis, problem formulation is an essential part of the work. The problem may be interpreted as a symptom of an underlying characterological pathology that has its origins in repressed childhood experiences.

However, externalizing conversations afford the narrative therapist a unique opportunity, “of not having to choose, not having to impose any single rendition, whether parent conflict, attention seeking, structural analysis, or psychological diagnosis” (Amundson, Webber, & Stewart, 2000, p.27). Externalizing practices allow for a non-deciding stance on the part of the therapist. They can also disarm internalizing discourses that seek blame for the problem’s existence (Morgan, 2000). Meanwhile the vigor of the therapy is maintained by turning over to the client (typically through the use of narrative-inspired questions) the important task of making sense of the problem, bringing it to life, detailing its history, and eventually tailoring appropriate measures for responding to the problem or constraint.

Externalizing allows for “theoretical pluralism” in regards to a problem’s possible origins and permits the therapist to be a “theoretical conscientious objector” (Amundson, Webber, & Stewart, 2000). In this way, the practice situates the therapist in a position that honors the poststructuralist perspective espoused by narrative therapy – whereby the myriad forces that shape the meanings we bestow our lives can be accounted for.

Making Deadly Issues More Discussable

Problems often have a grim and oppressive quality to them. By the time a person or family has reached the point of seeking consultation, they may be immobilized in the face of the problem. The problem(s) may feel like the dominant force in their lives. From a narrative perspective, the therapist wishes to respect the seriousness of the problem, yet the oppression and the immobilization it tries to foster are forces that are sometimes best to counter.

A narrative perspective often seeks to counter the oppressive nature of problems by employing externalizing practices in a way that reduces the deadly effect of the problem, thus reducing the grip it has on an individual or family and allowing for space to be created whereby changes can be made. This may mean using play, humor, irony, lightheartedness, etc. in the discussion of the problem or constraint.

Particularly in the narrative-based literature on working with families (e.g. Freedman, Epston, & Lobovits , 1997; Selekman, 1997; Nylund, 2000) practitioners discuss the importance of playfulness and creativity in facing serious problems. Making deadly issues more discussable often surfaces with families due the presence of children in the family and their unique developmental stages and family roles.

In fact, White (2000) notes that it was with children that he first began exploring the possibilities of externalizing conversations. Based on his sense of success with children, he expanded the practice to his work with adolescents and adults. Reflecting on children’s ability to engage in externalizing processes, White (2000) was struck by…their sheer delight in entering these more imaginative ways of speaking about what were so often considered to be deadly serious problems; and their joy in stepping into these conversations in ways that defied many a prediction about their continuing inability or powerlessness to address this predicament or that (p. 5).

Provided an opportunity, many children prefer to interact in a playful manner (Freeman, Epston, & Lobovits, 1997). Engaging children in a dialogue-saturated methodical and problem-solving manner often results in a child’s disengagement from the therapeutic process. Freeman, Epston, & Lobovits (1997) address the need for playfulness and creativity when tackling deadly issues with families in the following way:

The price of choosing seriousness for us as therapists may be the dampening of our own resources, such as the ability to think laterally, remain curious, be lighthearted enough to engage playfully with a child, and have faith that the situation is resolvable. Lacking these, we may have our wits dulled, lose our appeal to kids, or become overwhelmed. Do we dare to be playfully creative in the face of worrisome problems? What happens when we engage our imagination, humor, and resourcefulness in opposition to the deadly seriousness of problems? We believe this leads to the rise of inspired problem-solving and the downfall of serious problems (p. 3).

Too serious an approach on the part of the therapist may work to the advantage of the problem, particularly if the risk is of excluding or alienating the children of families who are seeking assistance. Freedman, Epston, & Lobovits (1997, p. 4) further suggest that, “When children and adults meet, play provides a common language to express the breadth and depth of thoughts, emotions, and experience; in this way, we share a lingua franca.”

While there are many approaches to making deadly issues more discussable, the use of externalizing conversations is widely discussed in the literature as a vehicle for making problems less serious, lighthearted, more discussable, and ultimately defeatable. As White (1989/1997, p. 6) has observed, externalizing conversations, “frees persons to take a lighter, more effective and less stressed approach to deadly serious problems.”

The literature also discusses creative non-verbal methods of externalizing that involve tasks such as art and family activities. For example, Selekman (1997) points out that,

Having the child go to combat with his or her family against the problem in the form of a family ritual or externalize the presenting problem in the form of a drawing on paper taps into the child’s spirit of fun, which helps lead his or her family in the direction of change (p. 119).

Again, it is at the intersection of playfulness and creativity that the entire family can join together and constructively address the constraints they face. Moreover, such externalizing practices act to lift the weight of shame and blame off of both children and adults in a family.

Acting as an Oppositional or Anti-Language

The narrative stance is one that is situated in a social constructionist framework; i.e. that our sense of what is “real” in the world and who we are (as individuals and as a society), is constructed by the language we use and the stories made possible through the use of this language. An impetus of narrative therapy is to call into question (i.e. deconstruct) not only dominant stories we might subscribe to, but also the dominant forms of discourse used to construct these stories as well (White & Epston, 1990; White, 1995; Freedman & Combs, 1996; Maisel, Epston, Borden, 2004).

If externalizing conversations are a workhorse of narrative therapy, one of their most powerful utilities is the manner in which they operate as an oppositional language to dominant discourses (White & Epston, 1990; White, 1995; Freedman & Combs, 1996). White (1995) remarks that a key point about externalizing conversations is how they introduce novel ways of thinking and speaking about a given problem, as well as alternative ways of acting in relation to the problem or constraint. He goes on to suggest that, “In promoting these externalizing conversations, we are engaged in an activity that is not entirely a pro-cultural activity (White, 1995, p. 42). This activity challenges the taken-for-granted cultural-bound ways of speaking about people’s lives and relationships. Epston (2005) eloquently adds to this theme by succinctly proposing;

As an anti-language its purpose is to make trouble for particular meanings, which have been disguised as universal, masqueraded as timeless, or taken on the guise of “nature” or naturalness. Another purpose is to challenge the guarantee of any structure of meaning; language, code, dialect; as being decisive and definitive (Epston private notes, 2005).

Externalizing conversations are a method for exposing that which is often hidden and presupposed in the taken-for-granted dominant language/discourse intertwined with our culture (White, 1995). The anti-language collides with the dominant language, exposing the hidden meanings, presenting a client with unmistakable juxtapositions by which to view the dominant language that has heretofore immersed their worldview. Externalizing conversations have been envisioned as a way to “exoticize the domestic” by inviting people to identify and unpack the cultural knowledges they live by (White, 1993, p.39).

The practices associated with externalizing problems are considered therapeutic counter-practices to the dominant culture. They engage people in a process with the intention of de-objectifying themselves and each other, and make room for the possibility for alternative life choices. As Pare and Lysack (2004) point out, “Externalizing talk is one example of the deliberate use of language as a “tool” or scaffold by which we [i.e. therapists] are acting into the conversation with the intention of inviting a person to entertain new possibilities.”

The anti and the dominant discourses generate “colliding worldviews”, each fraught with its own implications, both contesting for the speaker’s adherence (Epston, 2005). From the clients’ perspective, what is presumed to arise from such a collision is a unique therapeutic opportunity to understand problems, realize that options exist, mount counterstrategies against said problems, and create new linguistic tools for constructing alternative meanings and stories.

Through engaging in an anti-dominant language discourse, people are more prepared to separate from and objectify the problem(s) and devise an anti-problem stance. This anti-problem stance is often overt and combative in the way it speaks against and opposes the dominate language, particularly when dealing with deadly problems such as anorexia/bulimia, alcoholism, domestic violence, etc. When discussing the oppositional language, Maisel, Epston, & Borden (2004) stress that,

The main purpose of anti-a/b [anorexia/bulimia] language is the rolling back of the power of a/b, the re-voicing of the person, and the recreating of her world. Through this anti-a/b language, the person returns to the position of a critic of a/b and an advocate for herself and others who have been subjected to the abuses of a/b’s power (p.84).

But in confronting the condemning and marginalizing discourse of the problem’s dominant language, externalizing conversations can have a surprising tone as they inquire in ways that seek to counter the effects of the problem discourse. As a practical example of externalizing conversations as an anti-language tool, consider Maisel’s comments on the ways in which challenging the rhetoric of anorexia/bulimia requires an anti-language:

If a/b [anorexia/bulimia] speaks with arrogant authority, then Anti-anorexia speaks with humility, raising questions rather than issuing pronouncements. If a/b evaluates and degrades the person then an Anti-a/b language allows for the appreciation of the person and the critical scrutinizing of a/b. If a/b inscribes itself into the identity of the person, then Anti-a/b will expel anorexia from that identity and locate it in social and political contexts (Maisel, personal communication, May 24, 2005).

Lastly, while the anti-language must continually be constructed, it provides a “structure of resistance” that stands in opposition to the dominant structure (Epston, 2005). Condemnations and dead ends offered by the dominant discourse can be provided new, richer life-enhancing forms of expressions.

Researching the Narrative Therapy Approach

Art and science are not enemies. The painter operates within the bounds of physical laws; blue and yellow make green; when two objects occur in close juxtaposition, they appear related, and so on. The effective artist works within these laws to apply novel contexts and interpretations. In the same way, the effective [evidence based] therapist is more than a technician who applies sterile technologies to solve problems (Beutler, Bongar, & Shurkin, 1998, p. 75).

The above quote is an artful and in some ways hopeful interpretation of the relationship between practitioner and researching communities in psychology. But in practice, sometimes the rules of painting don’t adequately apply.

This section seeks to explore the various ways and degree to which narrative therapy has been researched, with particular emphasis on externalizing practices when applicable. The results have implications on the manner in which the use of externalizing is understood, as well as wider inferences about the future relations between narrative and research practices.

Overall, compared to other therapy approaches, narrative therapy has been receiving only modest attention in the research. Research focusing extensively on externalizing practices is even slimmer, with many of the relevant studies either giving externalizing only a basic discussion or evaluating its usefulness in conjunction with other forms of treatment such as group therapy, psychoeducation, etc. Evidence of the usefulness of externalizing is mixed. Some studies suggest clients find it much less important than other identified aspects of therapy, while other research implies that it has notably useful and distinct effects that warrant further investigation.

Establishing a Narrative Perspective on Research

The narrative tradition emphasizes a distinct set of values for both research and practice. A review of the literature reveals that a pivotal source of concern among narrative-inspired therapists has to do with how “traditional” research practices privilege professionals’ interpretations and understanding over those of clients (White, 1995; Epston, 2001; Gaddis, 2004).

Therapeutic ideas and practices have historically been influenced and constructed primarily by professionals in the field; whereas those subject to this expanding therapy knowledge historically have had little voice over what has been helpful or hurtful (Gaddis, 2004). The narrative worldview actively takes a stand against this precedent. Gaddis (2004, p.38) stresses that his perception of a narrative worldview assumes that “clients do not make sense of therapy in the same way that I do.” Consequently, he situates his narrative stance in a way that never assumes he as a therapist knows what is most helpful for clients. Clients must be intimately involved in their own therapeutic construction (Gaddis, 2004).

The intimate involvement of both client and therapist has widely been referred to as co-research within the narrative community (for example, White, 1995; Epston, 2001; Maisel, Epston, & Borden, 2004). This mode of working has an emerging tradition within the narrative community as both a research methodology and a therapeutically minded way of conducting therapy consultations.

Impact of the “Interpretive Turn.”

The narrative therapy tradition ushered in by practitioners such as White and Epston was heavily influenced by developments in the field of cultural anthropology. For instance, in the 1970s and 80s a dramatic development known as the “Interpretive Turn” took shape in the field of cultural anthropology, which had subsequent repercussions in the development of social work and therapy (Denborough, 2004).

According to Denborough (2004), this movement was associated with cultural anthropologists Clifford Geertz (1983), Renato Rosaldo (1992), Edward Bruner (1986), Victor Turner (1986), and Barbara Meyerhoff (1982), among others. Questioning what had widely been accepted practice in anthropology, these authors proposed that it was impossible for anyone to have an objective view in their research. They showed how anthropologists subjectively shape their research and in turn influence the communities they seek to study (Denborough, 2004). They called into question the presumed objectivity of the anthropologist, and stressed that their field must acknowledge how their own cultural beliefs, practices, ethnicity, etc. influence their research.

In response to the embedded interpretative nature of their work, these authors proposed an alternative form of research, a form of ethno-methodology that privileged the meanings and interpretations of the people being studied (Denborough, 2004). This ethnographic methodology called for inviting people of various cultures and groups to interpret and document their own lives and culture. Thus, the “subjects” were to become active research partners, and lend accountability to the research (Denborough, 2004). The narrative therapy concept of co-research has its origins in the Interpretive Turn movement. Co-research is discussed further in a subsequent section.

Reconciling research with narrative values.

Earlier in this study, it was discussed how narrative therapy’s poststructuralist and social constructionist ideas seek to uncover taken-for-granted assumptions and respect how knowledge is subjectively constructed. This is in direct contrast to the positivist, scientific method that seeks to discover knowledge that is independent of socio-cultural and historical context. It is from this positivist tradition that psychological research has its roots.

As mentioned earlier, the narrative worldview is one grounded in the idea that reality and truth are subjectively created constructs, not universal truths to be “discovered.” Observing the relative lack of research and empirical studies supporting narrative therapy, Etchison and Kleist (2000) note that this circumstance may be due to what they see as the constructivist orientation of narrative therapy, which is inconsistent with quantitative research methods . Research objectivity, which is adamantly denied by the constructivist perspective, remains a cornerstone of quantitative empiricism. Etchison and Kleist (2000, p.65) suggest that,

Constructivist approaches to researching therapy emphasize a qualitative understanding of one’s meaning given to experience (Nelson & Poulin, 1997) in context, without imposing the requirement of researcher objectivity. Participants and researchers in qualitative inquiry are regarded as coresearchers (Gale, 1993) who together explore the meaning of experience. Constructivist-based research places importance on the interaction between participants and researcher as a necessary component for quality data gathering and analysis (Merchant, 1997).

Gaddis (2004) discusses both his passion for narrative therapy and his interest in formulating a method of research that is both faithful to the narrative worldview and useful for advancing professional therapy knowledges. He presents two qualitative research methods that he believes are in some ways concordant with narrative principles: interpersonal process recall interviews (Elliot, 1986) and grounded theory analysis (Strauss & Corbin, 1990). He employed these methodologies in an earlier study of clients’ experiences of narrative therapy (Gaddis, 2002).

Interpersonal process recall is a unique type of interview method that uses audio or videotapes to stimulate recollections of past experiences (Gaddis, 2004). Participants watch tapes of past events (i.e. therapy sessions) and describe the experience they recalled during those events. Gaddis (2004) stresses the importance of “de-centering” the researcher’s role in collecting this information. An important feature is to allow the client to reflect freely on their experience, without the researcher guiding or overly-interpreting their responses.

Grounded theory is a methodology that attempts to develop theories that are grounded in the research participants’ own experiences. The role of the researcher is to reflect on the participants’ experiences in a manner that allows these experiences to be organized and categorized. The goal is to develop a coherent body of knowledge or theory that is derived from participant descriptions (Gaddis, 2004). Gaddis found this methodology useful for smaller studies, but ultimately argues that grounded theory risks emphasizing the researcher’s construction of meaning over that of the client.

Muntigl (2004) is also interested in examining narrative therapy, specifically how it facilitates client change. He demonstrates how a linguistic-semiotic analysis of narrative therapy is useful in exposing how the therapist uses narrative practices to assist a client in scaffolding alternative meanings for acting in the world. However, unlike the collaborative approach between researcher and participant that Gaddis (2004) seeks, the approach outlined by Muntigl (2004) focuses on third party observation and analysis of the discourse between therapist and client.

Couture and Sutherland (2004) propose that the methodology of discourse analysis is compatible with postmodern therapies. They see it as interactive between therapist and client, with a focus on the construction of meaning in conversations. Discourse analysis “investigates the specifics of interaction between therapists and clients with the goal of identifying moments or events of change in these interactions” (p.12).

Whereas Gaddis (2004) appears to be more of a practitioner seeking a research method that fits his view of narrative sensibilities, Couture and Sutherland (2004) appear to take the reverse approach; a research-oriented effort to address the unique sensibilities of postmodern therapy. These two sides may illustrate how some circles of practitioners and researchers are attempting to bridge the gap between the two camps in a way that benefits the lager mental health community.

Overall, while some researchers and practitioners appear to be seeking methodologies for researching narrative therapy, the volume of constructionist-focused qualitative research is modest compared with the much larger thrust of empirically-based research. Etchison and Kleist (2000) also note that only a minority of those seeking training in research methodologies and employed by major psychological journals are experienced in qualitative research methodologies. Consequently, there exists a gap of experience and occasional controversy over what types of research are deemed acceptable within given psychological communities. Michael White (1995) illustrated the perceived divergence between his role as a narrative-inspired practitioner and more formal research methodology in the following manner:

Those people who are practicing therapy, along with persons who seek therapy, are primary or basic researchers, and those people who collect data in a more formal way are the secondary or supportive researchers. I’ve always been interested in primary research, and find the continual demand from secondary researchers that primary researchers justify their existence to be quite tedious. If the secondary researchers in our field could go further in relinquishing the moral high ground… which would include the rendering transparent of the socially-constructed nature of their enterprise; then what secondary researchers do might become more relevant to what primary researchers do. I am sure they could have a very enriching collaboration (cited from Gaddis, 2004, p.47).

As discussed in more detail later in this study, there is a growing body of qualitative research designed to evaluate the effects of narrative therapy practices with various populations. These studies appear to be conducted by people who have a basic understanding and respect for the narrative therapy approach. However, they may not operate as passionately within what may be termed the narrative worldview as other narrative-inspired practitioners in the field. Their aspirations may be more research-focused and less political and social. Those conducting the research do appear distinct from those writer/practitioners that are largely responsible for the accumulating body of theoretical and ideological literature that continues to outline the principles of narrative therapy. The impact of narrative’s political and social sensibilities upon its accommodation with research is also explored in further detail in a later section.

It remains unclear the degree of acceptance the limited body of qualitative research has among those practitioners that have deep-rooted loyalty to key narrative principles such as the constructionist nature of research, locus of power away from the therapist/researcher, etc. In the quote above, White alludes to a promising collaboration between “primary” and “secondary” researchers. The degree to which each side has reached out to each other has historically been limited. Yet, the evidence cited earlier and in later sections suggests that the effort is gaining momentum.

It is important to note that the defined purview for this study is not intended to fully explore the context of the relationship between narrative therapy and research methodology. But, some mention of it is helpful as it applies to understanding the practice of externalizing and the opportunities for researching its effects. A concluding section in this study will evaluate the degree to which the written-about claims of usefulness for externalizing practices are supported by some manner of organized research.

The effects of political and social views about therapy.

At this point it is useful to add a point of qualification onto the conclusions drawn by Etchison and Kleist (2000). These authors focus primarily on the manner in which the constructivist values of narrative therapy create trouble for its integration with various research practices. There is another factor at play that may also be an obstruction; namely, narrative therapy’s valuing of the social and political effects of its own practice. The effect of this concern can be illustrated by contrasting narrative therapy with solution-focused therapy. While narrative therapy has had a volatile relationship with research interest and practices, solution-focused therapy has a history of actively embracing various types of qualitative outcome research and actively uses this research to promote the use of solution-focused and brief therapy practices (Miller, Hubble, Duncan, 1996). Similarly, Maione and Chenail (1999) discuss that, “although there have been few qualitative studies that investigate specific therapy models, the solution focused model has been the subject of several studies” (p. 68).

Solution-focused and narrative therapies share similar postmodern and poststructuralist assumptions about our sense of reality, the social constructionist nature of truth, and the primacy of the client’s involvement in identifying problems and their possible solutions (see de Shazer, 1988, 1993; Miller, Hubble, Duncan, 1996). However, solution-focused therapy appears to put less emphasis on the political and social effects of its practices, whereas loyal adherents to narrative therapy often pay close attention to the injustices and marginalizing qualities of social and political contexts (see Gaddis, 2004; Redstone, 2004). Narrative therapy rejects those culturally accepted research practices that are seen as supporting taken-for-granted cultural norms and that disregard marginalized accounts of ways of living. As such, narrative therapy reaches a philosophical impasse with research that solution-focused therapy may largely avoid.

Influence of Narrative-Inspired Co-Research

I have always thought of myself as doing research, but on problems and the relationships that people have with problems, rather than on the people themselves. The structuring of narrative questions and interviews allow me and others to co-research problems and the alternative knowledges that are developed to address them (p.178).

The above quote is how Epston (2001) described his interest in co-research. In fact, narrative-inspired co-research has been part of narrative practice since its popular inception (see White& Epston, 1990). Moreover, Freedman and Combs (1996) convey that, “In a very real sense, all narrative therapy is co-research. When we listen carefully to people’s stories we are doing research. When we ask meaning questions and preference questions we are asking people to join us in research” (p.287).

Epston (2004) has noted that he arrived at the term co-research around 1988-89 when he was working with family members afflicted with a rare skin disorder. He has been prolific in the area ever since; for instance, creating the Internet-based Archive of Resistance: Anti-anorexia/Anti-bulimia website (, and coauthoring a book exploring the use of co-research in assisting people in overcoming anorexia/bulimia (see Maisel, Epston, & Borden, 2004). Indeed, co-research has become a therapeutic methodology to address anti-anorexic/bulimic practices (Epston, 2004). The formal manner in which therapists consult with people about the effects of particular practices on particular problems (Freedman & Combs, 1996) may help to differentiate the look and feel of a co-research project. Another project involving co-research that is more in its beginning phase is the league of deconstructing addiction getting underway at the Dulwich Centre in Adeline, Australia (the website is

Co-research is a highly collaborative effort between client and therapist to come to unique terms with a problem(s) faced by the client. What results is a “co-production of knowledge” between the therapist and client (Epston, 2004). This effort typically serves three major purposes. First, co-research assists people in discovering ways of identifying and managing problems, and constructing alternative stories in a fashion unique to each person. Secondly, the project involves documenting (with the person’s permission) the unique journey undergone in naming, addressing, managing, and overcoming the problem. This often takes the form of letters, documents, and posts on the Internet similar to Blogs that constitute what are called “insider knowledges.” Consultations can also take place between people who have experienced a success over a particular problem and those that are still struggling with a similar problem. Third, a co-research project often asks the client to intimately describe what was helpful and unhelpful in the therapeutic process. This information not only benefits the therapist and the wider narrative community, but also serves to level the power hierarchy between therapist and client; which possesses a therapeutic component in itself for the person being consulted. In describing part of the process, Epston (2004) reveals the following:

When I first started meeting with these families we engaged in fairly thoroughgoing externalising conversations, in which the problem was a problem for everyone; and here I include myself. These conversations led to the co-production of certain knowledges about the problem, and ways of responding to it, that were of pragmatic value to the families (p.31).

Indeed, the work by Maisel, Epston, & Borden (2004) is an exemplar of the type of co-research currently being conducted within the narrative community. The process outlined by the authors intimately involves clients in constructing their own experience with the issue of anorexia/bulimia and assists them in furnishing personal accounts of protest against the influence of anorexia/bulimia. This documented work (combining client and therapist accounts) also provides other narrative-inspired therapists with professional knowledges that were derived by client’s own accounts of what was helpful and unhelpful in the course of therapy. Thus, useful and practical knowledge is added to the body of narrative therapy literature in regards to therapeutic work with anorexia/bulimia. The authors’ approach also embodies the kind of research that adheres to narrative principles about the locus of power and professional versus insider knowledges.

Thus, narrative-inspired practitioners have written about the usefulness of co-research in evaluating what was helpful and unhelpful in the therapeutic conversations themselves. In some ways, this has been a part of narrative practice for some time. During narrative therapy sessions, questions are often asked to ensure that the conversation is being experienced by a client as relevant and helpful. At the completion of therapy, questions are also asked that have clients evaluate particular practices and lines of enquiry implemented by the therapist (Epston & White, 1992; Morgan, 2000).

Interestingly, some within the narrative community may be feeling an inclination to incorporate or construct more clear elements of client-focused research methodology into the narrative approach. Recent literature has been published that seems to join the basic ideas of co-research more intimately with how it might better inform, validate, and expand the professional knowledges and practices of narrative therapists (see Denborough, 2004; Gaddis, 2004; Maisel, Epston, & Borden, 2004; Redstone, 2004; Clark et al., 2000). The intention is to further augment the claims made by narrative practitioners with clients’ documented reactions about the helpful and unhelpful aspects of their therapy experiences.

Reciprocal connection between therapy practice & research methodology.

As mentioned earlier, co-research has been a practice employed for some time, typically as a collaborative therapeutic effort for the benefit of clients struggling with particular issues. More recently, some narrative practitioners have been actively promoting its usefulness as a style of research that is commensurate with narrative principles of therapeutic change, useful in the construction of professional therapy knowledges, and also amenable to qualitative methods of analysis (see Gaddis, 2004; Clark et al., 2000; Madsen, 1999).

The narrative practice of co-research has in some ways become a political and philosophical statement amidst the pervasive traditional understanding of psychological research. For instance, Gaddis (2004) concedes that research is important in formulating therapists’ ideas, yet he expresses that, one of my greatest concerns has to do with how traditional research practices privilege professions’ interpretations and understandings over those of clients. I have attempted to re-consider therapy research so that its main purpose is to honor clients’ accounts of therapy. My hope is that this will enable us as therapists to be taught as much by clients as by other professionals (p. 37).

Co-research embraces the narrative sensibility of placing therapeutic power in the hands of clients and allowing them to have ownership of how they make sense of and manage their lives. But, the approach outlined by Gaddis cannot help but be political as it actively seeks to hammer out an approach to psychological research that is in many ways distinct from traditional research methodologies, which locate interpretive and prescriptive power in the hands of professionals. If traditionally generated research knowledge seeks an elevated status when it comes to making claims about what is “true” (Gaddis, 2004), than the narrative enterprise of co-research makes an effort to deny that notion or at least to reduce the implicit hierarchies of power and knowledge (Clark, Jankowski, Margee, & Springer, 2000).

Clark et al. (2000) note that, “Despite the increase in alternative forms of participatory research, there is still no clear methodology for including participants as co-researchers in the interview analysis and interpretation process” (p.57). In response, the authors propose structuring a form of co-research into a model of qualitative research methodology that utilizes social constructionist ideas as a foundation. Similar in makeup to later efforts by Gaddis (2004), Redstone, (2004), and Maisel, Epston, & Borden (2004), this method of analysis would go beyond more traditional forms of qualitative analysis by involving participants/clients more intimately in the coding and interpretation process of helpful and unhelpful qualities in the therapy experience. Informed by social constructionist theory, this research method brings together the subjective interpretation processes of both researcher and participant and transforms the research process into more of a relationship than a method. Note the following graphs that illustrate the difference between traditional forms of information analysis and validation, and that allowed under the method of co-research (Clark et. al. 2000).

In the traditional model, the principle investigator consults with participants about key themes. But, it is only the investigator that uses the text generated by the participants to determine the final makeup of the research codes and themes.

Within the co-research method, the roles of “expert” and “subject” become more blurred (Clark et. al. 2000). The subjective interpretation processes of researcher and participant are brought together to form a co-research group. The group works together to co-construct, codify, and validate the resulting themes.

Gaddis (2004) is also concerned for the durability of the narrative approach itself. He expresses concern at the large number of descriptive accounts of narrative therapy provided by professionals and the fact that fewer accounts are given by actual clients (Gaddis, 2004). He proposes that co-research, with its emphasis on clients’ accounts of what was beneficial in therapy, can be a research tool that strengthens the professional therapy knowledges within the narrative community in a way that is isomorphic with the community’s view of therapy. In other words, just as narrative therapy emphasizes clients’ construction of solutions to their own problems, so to should narrative-inspired research emphasize clients’ accounts of what is helpful and unhelpful during the process of therapy. Such practices will ensure that professional accounts of narrative theory do not usurp people’s subjective experience of the narrative therapy process.

Overall, co-research can be viewed as a form of research, in that it has a purpose of generating client-oriented accounts about what does and doesn’t work in therapy. This in turn, may be beneficial to professionals who are interested in finding creative solutions to various types of problems and constraints in people’s lives. Yet, its constructionist orientation and purposeful denial of the notion of researcher objectivity continue to place it in a marginalized grouping within the larger research community. In Epston’s discussion of co-research, his view on objectivity is clear. Epston (2004) remarks,

Unlike conventional research, the process of co-research does not claim to be objective, nor does it aspire to objectivity. The process itself is inextricably entwined with its purpose, which is to generate knowledge that can influence in preferred ways a person’s relationship with the particular issue for which they have sought counseling (p.31).

Indeed, the impetus behind co-research is often as a therapeutic tool with clients. Here as a final comment, it is worth noting a compelling reciprocal therapy/research characteristic of co-research practices. Some qualitative research suggests that involving clients in the research process as active co-participants produces a therapeutic effect of its own. Maione and Chenail (1999) cite several studies (including Gale, 1992; McNamee, 1988; Wright, 1990) that suggest the intimate and empowering process of researching clients’ perspectives about their therapy can have a therapeutic benefit on its own, even when intervention and change is not the intention of the researcher/interviewer.

This might come as no surprise to some who practice narrative therapy. Indeed, early in their use of the narrative metaphor with clients, Epston and White (1992) noticed the therapeutic effects co-research had on people: “When persons are established as consultants to themselves, to others, and to the therapist, they experience themselves as more of an authority on their own lives, their problems, and the solution to these problems” (p.17).

The use of externalizing practices in co-research.

A review of the literature reveals that the practice of co-research is intimately entwined with externalizing conversations. For instance, speaking about the co-research documents within the Archive of Resistance: Anti-anorexia/Anti-bulimia, Epston (1999) remarks that, “the archives can take many forms… but what is common to them all is their manner of speaking; anti-anorexia; an anti-language, a radical form of an externalizing conversation” (p.144).

Externalizing conversations provide much of the therapeutic power behind co-research practices, particularly when metaphors of resistance are being addressed that allow a client to reverse the vocabulary of self-blame and internalization, and mount a decided protest against a newly objectified problem. Examples of this type of work include that conducted by Maisel, Epston, & Borden, (2004) in the realm of anorexia/bulimia, Nylund (2000) in his work with children diagnosed with ADHD, and Freeman, Epston, and Lobovits (1997) in their creative accounts of working with difficult childhood problems.

Maisel et al. (2004) reveal that externalizing conversations allow for the creation of an anti-anorexia language. They note that:

In order for anti-a/b (anorexia/bulimia) research to bring to light aspects of people’s experience that contradict a/b’s claims and generate meanings that legitimate a person’s own experience, this co-research relies upon the introduction of an anti-language. It is only by supplying an anti-a/b language that the meanings generated by a/b can be rendered controversial and contestable (p.84).

It is important that this language is not imposed upon clients. That is where the uniqueness of co-research comes into play. Co-research allows for the active co-production of knowledge by clients and therapists (Epston, 1999). When working with anorexia/bulimia, Maisel et al. (2004) believe that, “the real experts on problems are those people who experience them first-hand” (p.81). Tapping into this experience through co-research allows for local or insider knowledges to be developed. These knowledges are regarded as more trustworthy and more valuable to the therapeutic process than systematized professional knowledges that typically speak for or about people but tend to place minimal value on people’s individualized experiences (Maisel, Epston, & Borden, 2004).

In working with families with children diagnosed with attention deficit hyperactivity disorder (ADHD), Nyland (2000) uses a highly collaborative approach with clients similar to co-research. In some ways his approach does differ from that of Maisel et al. (2004) in that he also actively incorporates aspects of solution-focused and collaborate language systems approaches to therapy. Yet, the manner in which client experiences inform his distinctive approach is reminiscent of narrative-inspired co-research.

Nyland (2000) notes that engaging in externalizing conversations is a vital aspect to his collaborative work with families. From the initial session, he seeks to use externalizing language to explore with families their understanding of the identified problem(s) and to get their personal accounts. Each family interprets the diagnosis of ADHD differently, and Nyland (2000) stresses that before exploring a family’s unique resources for managing their lives, he must work collaboratively in defining and documenting the problem. His book provides a comprehensive account of his approach to working with those children and families living with an ADHD diagnosis. Similar to the work of Maisel et al. (2004), Nyland’s (2000) collaborative approach is propelled less by psychological theory and more by clients’ own accounts of their experiences.

The work by Freeman, Epston, and Lobovits (1997) emphasizes playful and creative methods of working with children and their families. It also represents another collaborative therapeutic effort between therapists and clients that is influenced by co-research. As with the narrative tradition, clients’ views of problems and distinctive solutions are solicited and encouraged. Externalizing conversations help bring to life the creative metaphors that allow these children and families to identify, track the effects of, and mange their relationships with problems in their lives.

Moreover, Freeman et al. (1997) discuss how they work with clients to document their accounts of reclaiming their lives from problems. As can occur in a co-research framework, children sometimes act as consultants (either in person or in writing) to other children and families suffering from similar issues. What results are accounts of what proved useful for particular people facing a range of issues. What is common among these accounts is the externalizing fashion in which families relate to the issues for which they sought assistance.

As detailed above, externalizing conversations are integral to co-research practices. But, there is another aspect to this relationship. Co-research has also been used as a research methodology to explore and expand on the therapeutic uses of externalizing conversations.

For example, Stacey and Hills (2001) employ co-research to investigate the helpful and unhelpful ways externalizing has been used during therapy sessions. Externalizing often employs the use of metaphors; for instance the metaphors of resisting, overcoming, protesting, battling, coming to terms with, or learning to live with a particular problem, constraint, or issue. The authors believe that some narrative therapists restrict themselves in the diversity of metaphors used with clients when defining a client’s relationship to a particular problem. They note the popular usage of the protest metaphor; the idea that person should protest the existence of a problem, stand up to it and seek to eradicate its presence (Stacey & Hills, 2001).

Through their co-research, they come to conclude that this metaphor alone does not adequately represent the breadth or creativeness of narrative work (Stacey & Hills, 2001). Sometimes the protest metaphor is not the best fit with clients who seek alternative kinds of solutions to problems and issues. Co-research allowed therapist and client to work first-hand in developing an appropriate metaphor for their desired relationship with a problem. For instance, they discuss how the metaphor of balancing might be useful when a client seeks to confine a problem to the positive part it may play in their life. Balancing may prove useful in cases where the issue is something like altruism, over-responsibility, or worry (Stacey & Hills, 2001).

A goal of the authors is to forward and enliven narrative ways of working while avoiding transforming narrative theory into an ideology. Joining with clients to produce both client-specific knowledges, but also wider professional knowledges that may educate other narrative therapists helps to keep the focus of the work on the client and less on what narrative or externalizing theory teaches.

In a similar fashion, Freeman, Epston, and Lobovits (1997) warn that aggressive, “power over the problem” metaphors might be restrictive. Referring to aggressive and combative externalizing metaphors, the authors suggest, “They may also, through their heavy-handedness, discourage a lighter or more playful approach” (Freeman, Epston, & Lobovits, 1997, p.62). Arriving at this conclusion, the authors recommend that it is important to collaboratively consult with clients about their preferred metaphors and how particular descriptions may be rooted in culture and gender (Freeman, Epston, & Lobovits, 1997).

To conclude this section, it is worth mentioning a final relationship between externalizing practices and co-research. White and Epston (1990; Epston, 1999) suggest that the research process itself might create a form of externalization, particularly by having the problem and the experience of therapy be spoken about in an objectifying manner. An example might be when a family returns for a follow-up exploratory (co-research) session following the completion of therapy. The intention may be to gather first-hand client accounts of what was helpful and unhelpful during the therapy process. The act of and manner in which family members speak about the therapy experience, in itself, externalizes and objectifies that experience. They may speak about the experience as something external to them that had distinct effects; some helpful, others unhelpful. As part of this process, the problems and issues that brought the family to seek consultation in the first place are inclusively externalized since their presence was part of the therapy experience. Thus, the co-research process itself can reinforce the therapeutic value derived from the externalizing conversations that took place during and following therapy.

Types of Research into Narrative Therapy

Since the early 1990s there has been a steady but limited effort to employ outcome research to evaluate the effectiveness and therapeutic qualities of narrative therapy. These efforts have typically included various forms of qualitative studies (particularly based on ethnographic and grounded theory approaches), and a limited number of mixed method studies utilizing a combination of quantitative and qualitative strategies. Small scale case studies (five or fewer participants) are also a popular approach to evaluating narrative therapy. In several case studies, narrative methods are mixed with other therapy approaches such as motivational interviewing (Cowley, Farley, & Beamis, 2002), group therapy techniques, psychoeducation (Fristad, Gavazzi, & Soldano, 1999), and even psychodynamic theory (Diamond, 2000). No pure quantitative studies were found.

However, compared to the torrent of research that has been conducted over the last twenty years on cognitive based and other therapies (CBT, DBT, etc.), the amount focusing on narrative-inspired therapies is essentially a trickle. From a wider vantage, narrative therapy seems to be largely sitting on the shoreline operating within its own worldview as the stream of psychological research washes by. Certainly there exist (and this study has pointed out) fundamental narrative principles that often conflict with predominate styles of research methodology.

Five years since its publication, Etchison and Kleist’s (2000) article remains one of only a handful of scholarly articles to tackle the issue of research into the effectiveness of narrative therapy, and is regularly cited in dissertations and scholarly articles. As discussed earlier, the authors detail some of the values and ideas of narrative practice that are often at odds with research practices; for instance the fact that notions of empiricism are at odds with constructionist principles. Etchison and Kleist (2000) also suggest the paucity of research on narrative therapy is due to a lack of training in qualitative methodology among aspiring researchers.

An overall conclusion by Etchison and Kleist (2000) is that, “support for the use of narrative approaches with families is at best tentative given the small number of clinical studies” (p.65) However, the authors also indicate, based on their own review of available research, “that narrative approaches to therapy have useful application when working with a variety of family therapy issues” (p. 65). They further propose that, “narrative therapy is based on principles that are congruent with context-sensitive research methodologies (e.g. ethnographic, grounded theory) that deemphasize generalizability” (p. 65). In other words, they see a fit for qualitative research with narrative principles, and they recommend that narrative practitioners be sensitive and flexible to the unique dynamics of each client or family.

The enterprise of psychological research has indeed evolved over the last twenty years. Recent literature has explored ways of mending the gap between researcher and practitioner in regards to the use of postmodern therapies and research methodologies (see Clark et. al., 2000; Couture & Sutherland, 2004; Hertlein, Lambert-Shute, & Benson, 2004). Nevertheless, it is unclear to what extent such notions are accepted. The friction addressed by White (1995) between researcher and practitioner appears to remain in existence to varying degrees.

This study proposes that the conditions outlined by Etchison and Kleist (2000) have likely only modestly changed. There still appears to be a dearth of qualitative research assessing the use of narrative therapy. Moreover, much of the qualitative work continues to be smaller scale efforts or case studies that provide less robust findings for others in the psychological field to examine. Those researchers that do incorporate narrative practices into qualitative or mixed methods studies often discuss and utilize narrative therapy more in the context of a set of techniques.

This is in contrast, and in most respects contrary, to the beliefs of those practitioners that fashion narrative therapy within a larger worldview and reject the idea that narrative therapy can effectively be boiled down to a set of techniques (see White & Epston, 1990; Freedman & Combs, 1996; Gaddis, 2004; Redstone, 2004). Many of these practitioners are understood as the driving force behind the development of the narrative model of therapy. Similarly, their views have been among the most outspoken in how traditional and systematized research practices are in conflict with the social, political, and therapeutic principles of narrative therapy.

Thus to some extent, distinct communities are involved in the use of narrative therapy. The split is not simply between researcher and practitioner, but also between those practitioners who view narrative as a potentially useful set of techniques and those that embrace it for its larger philosophy and worldview. The current relationship between the various communities appears to be less an effort to reconcile differences, as Etchison and Kleist (2000) suggest is feasible. Instead, each community is using narrative therapy for their own pursuits; whether it is to generate mixed methods of outcome-based research or to broaden narrative’s application as a comprehensive approach and worldview.

Lastly, as mentioned earlier, there appears to be an invigorated effort within the narrative community to utilize co-research as a tool for building upon professional knowledges in a manner congruent with the narrative sensibilities (see Denborough, 2004; Gaddis, 2004; Maisel, Epston, & Borden, 2004; Redstone, 2004; Clark, Jankowski, Margee, & Springer, 2000). These efforts only partially fit with more commonly recognized qualitative research methodologies. Such work will certainly be of value to those already loyal to narrative principles, but it remains to be seen if such efforts grab hold of a larger audience beyond the narrative community.

Evaluating narrative therapy in general versus specific practices.

Many of the available qualitative studies also tend to focus on the narrative approach in general, as it applies to specific problems or populations. Narrative therapy’s application in specific clinical environments may be useful, as Etchison and Kleist (2000) suggest that narrative therapy has the flexibility to be tailored to specific client needs. Thus, such studies allow for unique outcomes and descriptions of how narrative practices can inform therapists and assist clients and families.

However, only broadly assessing the narrative approach, rather than expending some effort researching and documenting specific narrative practices, such as externalizing conversations, has limitations, and may have be doing a disservice to the narrative community. White & Epston (1990) have consistently held that narrative ideas are not simply techniques. Rather, they represent an overall approach that embraces a particular worldview based on poststructuralist and social constructionist principles about our sense of reality, as well as concern over the social and political implications of such a worldview.

Nevertheless, an argument can be made that the various practices of narrative therapy can be broken down, discussed, and their effects evaluated; such as the focus of this study on the use of externalizing practices. By evaluating the therapeutic effects of various narrative practices (such as externalizing), the psychological community can have a greater understanding of, and appreciation for, the distinct qualities that comprise the narrative approach to therapy.

Indeed, it appears that those who engage in narrative-inspired co-research have made significant efforts to examine the effects of specific narrative practices with clients, often so therapists can face “the moral and ethical implications of their practices” (White, 1993, p.57). However, these results are not presented as a comprehensive manner of research into the effectiveness of particular practices. The work by Maisel et. al. (2004) with anorexia/bulimia appears to be the closest approximation of documenting a narrative-inspired style of research. Thus, it remains to be seen if this methodology can be further developed and become recognized by a larger psychological community. The friction between what White (1995) referred to as “primary” and “secondary” researchers may remain a serious obstacle.

Research on externalizing practices.

No qualitative or quantitative research study was found that focused predominately on the practice of externalizing and its therapeutic effects. Nevertheless, an accumulating body of qualitative research has been conducted that include data on or discuss externalizing practices as part of larger studies and findings. Many of these studies examine narrative therapy in a more diffuse sense, where components of narrative such as externalizing, co-construction, and developing personal agency are mentioned as part of the approach, but not explicitly examined for their individual effects or contribution. Often, the studies are attempts at assessing (qualitatively) the effectiveness of using narrative approaches with specific kinds of problems or populations. The following is a review of several of these studies, with emphasis placed on findings related to externalizing practices.

In one of the few more comprehensive qualitative studies that highlight clients’ experience of externalizing practices, O’Connor, Meakes, Pickering, and Schuman (1997) examined families’ perceptions and attributed meanings of their experience in narrative therapy. The study sought to discover what families found helpful (i.e. therapeutic) and unhelpful in their experience. The selected families presented with a host of problems including conduct disorders, family violence, school problems, grief associated with parental death or divorce, aggression towards siblings, etc. These families were selected because they were being seen in family screenings by the narrative team at a university hospital outpatient clinic. Each family was in various stages of treatment, covering a span of 6 to 16 sessions over the course of one year.

The researchers used an ethnographic design guided by the question, “What is the family’s experience of narrative therapy?” The authors decided on this design for three reasons (O’Connor, Meakes, Pickering, & Schuman, 1997): (1) The research question required the possibility of a complexity of responses. Ethnographic research allows for such complexity. (2) The practice and process of narrative therapy has many similarities with an ethnographic interview. This type of qualitative research methodology is isomorphic with narrative practice. (3) Consistent with the role of clients in narrative therapy, ethnographic research allows for persons not to be regarded as subjects, but rather as participating co-researchers.

The authors asked four general questions; designed to elicit rich descriptions from the clients about their experience in narrative therapy. The questions were, “1) What has been helpful in the therapy? 2) What has not been helpful in the therapy? 3) What is your overall experience of narrative therapy? 4) What is an image or symbol to describe your experience of therapy?” (O’Connor, Meakes, Pickering, & Schuman, 1997, p.483). These primary questions included subsequent questions designed to facilitate enriched descriptions of the experience.

The data gathered from the audiotape interviews was coded using latent and manifest content analysis. The codes were developed using an inductive grounded-theory approach that examines written text and notes themes, commonalities, differences, and organizes them into common codes. The researchers then used a deductive method to measure the number of times each code appeared in the interview data.

Six major themes emerged from the data. They were: (1) externalizing conversations, (2) unique occurrence and alternate stories, (3) developing personal agency, (4) discussion of consulting and reflecting teams, (5) the narrative practice of audience;building, and (6) the helpful and unhelpful aspects of therapy.

Externalizing conversation was mentioned the least (15 times during the interviews). Developing personal agency and consulting/reflecting teams were mentioned the most (45 and 46 times, respectively).

Overall, the authors conclude that the practice of narrative therapy provides an excellent context for the ideas and practices that empower personal agency in family members. All the families reported some reduction in the presenting problem. The authors also suggest that the ethnographic method is congruent with the narrative worldview. The client’s experience in evaluating effectiveness of therapy was the core of the research. They note that the data supports the practices of building personal agency, reflecting and consulting teams, and externalizing conversations. Clients cited these three elements the most times. Interestingly, the authors call attention to the following:

The areas that clients notice the most in narrative therapy are personal agency and reflecting and consulting teams. The area they noticed the least is the externalizing conversation. What role does the externalizing conversation play in the therapy? Further research is needed (O’Connor, Meakes, Pickering, & Schuman, 1997, p.492).

Regarding clients’ comments on externalizing conversations, the authors did not attempt to draw out the therapeutic qualities noticed by the clients. However, they make the tantalizing comment that, “the use of externalizing conversation in conversing about the problem often requires a cognitive shift for clients. Once the shift has occurred, there is often a great reduction in the problem” (O’Connor, Meakes, Pickering, & Schuman, 1997, p.492). They suggest that externalizing aided in a cognitive therapeutic shift, yet they neglect to discuss the topic further.

The authors did include numerous quoted examples of clients’ responses to each of the six coded themes. An examination of the included client quotes regarding externalizing conversations indicated that the major therapeutic themes seemed to be that externalizing helped to, 1) reduce blame for the problem, and 2) make the topic of the problem more discussable by reducing some of the emotional tension. Of course, this analysis of selected quotes is quite informal and limited. But, it possibly points to a useful methodology (ethnography) and data source for a more detailed examination of the therapeutic qualities of externalizing conversations.

Additionally, the authors may have missed several useful opportunities to ask more detailed questions about client perceptions of externalizing conversations. Particularly in the case of the coded theme of developing personal agency, there are quotes that suggest the clients noticed the various ways the therapists and reflecting teams helped them assume responsibility in their lives. While the clients may not have spoken to it directly, or maybe the authors missed it, it has been this study’s observation that the narrative literature suggests that externalizing is a practice useful for, among other things, empowering clients with a heightened sense of personal agency. Assuming such externalizing practices were used during the course of this investigation, a major therapeutic aspect of externalizing practices may have been overlooked. Moreover, even if clients offered no comment in the interview on externalizing practices that aided in developing personal agency, this does not mean they were unaware of it. They simply might not have been asked the right question to elicit a response. Again, the authors designed the study to ask four broad questions about clients’ experience in narrative therapy, as well as “subsequent questions that could be used to facilitate a rich description of the experience” (O’Connor, Meakes, Pickering, & Schuman, 1997, p.483). A suitable question might have better elicited clients’ experience of externalizing as part of the process of developing personal agency.

O’Conner and associates published a later study evaluating therapists’ experience as part of a narrative therapy team (O’Connor, Davis, Meakes, Pickering, & Schuman, 2004). The researchers this time sought to answer the question, “What are therapists’ experiences of narrative therapy?” The study’s purpose was to discover narrative therapists’ perceptions and meanings of their practice of therapy.

As part of their literature review, the authors note that, “the literature on narrative therapy includes qualitative research, case studies, as well as personal and theoretical reflections. No quantitative studies were found” (O’Connor, Davis, Meakes, Pickering, & Schuman, 2004, p.24).

The study was comprised of eight narrative therapists who belonged to a narrative team in an outpatient clinic. These therapists were observed and interviewed. Again, O’Conner and his associates utilized an ethnographic methodology. Interview sessions were audio recorded and the transcripts were coded and clustered according to agreed-upon themes and concepts. The themes include a sense of success in reducing the clients’ problems as well as some limitations of narrative therapy.

This study did not discuss externalizing practices in a direct or thorough manner. However, externalizing conversations were mentioned under the larger themes of successfully reducing presenting problems and co-constructing alternative stories. The authors note that all eight participating therapists in the study found externalizing conversations helpful in their effort to co-construct alternative stories with their clients.

What was particularly interesting was that all of the therapists discussed how externalizing conversations promoted personal agency among family members by situating the problem outside of the client/family. This is somewhat of a contrast to the findings of O’Conner et al. in their 1997 study (discussed earlier) on clients’ perceptions of therapy. In that study, the authors report that personal agency was mentioned more by the clients as an important aspect of the therapy than externalizing conversations. However, in the study of therapists’ experiences, externalizing appears to be given greater prominence for a therapeutic role. Why is this? Do therapists put more emphasis on the experience and effects of externalizing conversations? Do clients sometimes not perceive the effects of externalizing practices, even when they wholeheartedly engage in them? O’Conner et al. don’t address these questions. In fact, the discussion of externalizing practices in their study and findings is brief. Their study’s focus is a broader look at therapists’ perceptions and meanings of their experience. The authors concede in their conclusion that, “More research needs to be done, research that moves beyond the beginning stage” (O’Connor, Davis, Meakes, Pickering, & Schuman, 2004, p.37).

In another study, Fristad, Gavazzi, and Soldano (1999) used a psychoeducational model of intervention with mood-disordered adolescents and their families. What was unique about this study was how the model of intervention incorporated methods of narrative-inspired externalizing practices. These practices were used to assist clients in differentiating the effects of a mood disorder diagnosis from their view of themselves, and to improve communication between clients and their families and between family members and therapists (Fristad, Gavazzi, & Soldano, 1999).

While this study poses some exciting ideas about the uses of externalizing conversations, the research conducted by the authors is weak in its assessment of the helpful qualities of externalizing. They refer to prior research that suggests that psychoeducational programs are effective in working with mood disorders. But unfortunately, this prior research did not encompass the additional use of externalizing practices. The authors’ contention that incorporating externalizing was beneficial is based primarily on a limited set of family case studies (Fristad, Gavazzi, & Soldano, 1999).

Archer (1997) also uses a case study method to promote the use of externalizing practices with individuals who have experienced natural disasters. The author uses a case example of a child who experienced the effects of a tornado. The author contends that the creative process of externalizing helped the child reclaim a sense of competence and cope with loss and vulnerability. Archer also links the externalizing approach to the more researched trauma work of Critical Incident Stress Debriefing (CISD). He notes that CISD “utilizes a reevaluation and reviewing process not unlike Michael White’s externalization format” (p.80). Archer also makes the connection that, “In the treatment of trauma, particularly the victims of natural disasters, the relaxation of rigidified trauma presuppositions provided by the externalizing process allows for space into the healing world of alternative self views” (p.80). The author concludes by noting that further research is warranted into the use of externalization as a debriefing and therapy tool for survivors of natural disasters.

Besa (1994) conducted a study to assess the effectiveness of narrative therapy in reducing parent/child conflicts. Six families comprised the study. Each was screened for parent/child conflict. A single-case methodology was used that employed a treatment package strategy that allowed for outcomes to be measured (for instance, by measuring reductions in targeted behavior/symptoms). The results were evaluated using three multiple baseline designs. All families participated in a form of narrative-inspired therapy that shared common approaches including externalization, relative influence questions, identifying unique outcomes, and assigning between-session tasks. According to Besa, the study’s findings suggest that narrative therapy was effective in reducing parent/child conflict, and may therefore be applicable to families experiencing similar parent/child problems.

Besa mentions externalizing conversations as a prominent and unique component of narrative therapy. However, he doesn’t attempt to account for the therapeutic value or non-value these externalizing practices had on the therapy. As with other studies, he chose to evaluate collective common elements of narrative therapy rather than isolate the therapeutic effects of specific narrative practices. In his concluding remarks, the author offers that, “Another area of future research is modifying this design to discover the relative effectiveness of each narrative technique. This could result in refining narrative therapy so that a given problem could receive the optimum balance of interventions” (Besa, 1994, p.324). He further proposes that specific types of single-case research designs can accomplish such a task (Besa, 1994).

In another study involving the use of externalizing conversations, Knight et al. (2003) reported the qualitative research findings from a preliminary study of adolescents with type-1 diabetes. Twenty adolescents with poorly controlled diabetes were allocated to a six week / one hour per week group therapy regime. Therapy consisted of a combination of motivational interviewing and externalizing conversations. The participants completed semi-structured questionnaires assessing their perceptions and management of diabetes before, immediately following, and 6 months after the intervention. The questionnaires were analyzed using content analysis. Similar to other studies, the authors did not isolate the effects of externalizing conversations. They instead spoke more broadly how the combined treatment assisted the participants in dealing more openly with the effect of diabetes and how the condition became viewed as less restrictive on their lives. They tentatively concluded that their group intervention strategy of motivational interviewing and externalizing conversations “led to an adaptive shift in illness perception” (Knight et. al. 2003, p.149). Even at the 6 month follow-up, the participants reported that diabetes was having a less restrictive influence on their lifestyle and that self-care behaviors were more common than prior to group therapy.

Goddard, Lehr, and Lapadat (2000) conducted a study that utilized focus groups comprised of parents of children who had been diagnosed with disabilities. Stories were collected in open-ended focus groups using criteria derived from what they viewed as constructivist principles of narrative therapy. The authors audiotaped the focus group sessions and used a thematic analysis grounded in a critical constructivist perspective to analysis the transcripts. The authors concluded that constructivist principles implicit within narrative therapy were used by these parents in coming to terms with their beliefs and experiences of their children who had been labeled as disabled. Moreover, they believe that the theoretical orientation of narrative therapy is conducive to the counseling and support needs of parents of children labeled as disabled.

Oddly, the authors never mention externalizing conversations by name, but they refer directly to it. For example, they report that, “the ideas that problems are not within people, and that parents’ voices make key contribution to discourse on disability, are particularly relevant to working with parents of children with disabilities” (Goddard et. al. 2000, p. 286). These parents were using a form of externalizing conversations that helped them define problems in a preferred and more manageable way.

In concluding this section, it is apparent that some qualitative research has been conducted that utilizes or at least mentions the effects of externalizing conversations. Analyzing the results of this body of research lends credence to the idea that externalizing practices do possess helpful qualities to clients.

However, no research evidence was found that focuses predominately on externalizing practices and explores its therapeutic effects in a comprehensive manner that involves both participant and therapist interpretation. As such, robust claims about therapeutic attributes are lacking at this time. In light of this lack of evidence and in recognition of the compelling prospects that have been exposed by existing research, efforts toward a more comprehensive analysis of the useful qualities of externalizing conversations may be warranted.


Chapter Four: Findings and Discussion

The purpose of this chapter is to integrate the preceding literature review with this writer’s own observations. The first section addresses the narrative community’s efforts to present the uses and therapeutic qualities of externalizing practices. A subsequent section critiques the body of research that has been conducted on narrative therapy and externalizing practices, and makes recommendations for future research efforts.

Evaluation of Externalizing Practices

While emerging as a focused approach less than 20 years ago, the narrative approach to therapy illustrated in this study has generated a considerable amount of scholarly and clinical interest in the psychological community. Although Michael White has been criticized for some of his interpretations of postmodern thinkers (see White, 1995), I believe the philosophical foundations he and David Epston (see White & Epston, 1990) assembled into a practical form of therapeutic intervention have had profound effects both on people’s lives and within the mental health community. The efforts by these and other authors have developed the narrative approach into a comprehensive clinical model/orientation, as well as a worldview sensitive to sociocultural and political implication.

Clinical Implications of this Review

This study has highlighted the use and value of externalizing conversations in narrative therapy – as outlined by those practitioners most committed to the practice. What is revealed is the extent to which externalizing conversations play a crucial part in many of the key practices of narrative therapy. Without this practice, many of the therapeutic aspects of narrative therapy would lose much of their potency. Many of the authors cited in this study go to exceeding lengths to describe the practice of externalizing and illustrate how it is incorporated into therapeutic discourse with clients.

Nevertheless, I believe it is also the case that the therapeutic qualities of externalizing conversations have yet to be adequately explicated. Externalizing is an intimate part of many narrative practices. Yet, no body of literature was found that systematically evaluates the relative importance of externalizing practices in light of outcome measures or on clients’ perceptions of the effectiveness of therapy. As this study advises, additional research into the usefulness and effects of externalizing conversations is indicated.

This study is novel for its illumination of the purported impact of externalizing practices, as well as how narrative therapy intimately relies on externalizing for its therapeutic interventions. The goal has been to document this practice in one body of literature, thereby furthering the narrative and broader mental health community’s knowledge of the uses and value associated with externalizing conversations.

Conclusions about Externalizing Practices

From a scholarly perspective, narrative therapy’s allegiance with social constructionist sensibilities about the importance of discourse is evident in the purposeful, creative and detailed manner in which narrative practices such as externalizing are described. Scrupulous attention is given in describing these practices, linking them to larger postmodern ideas, and detailing in a practical context their implications for clients and therapists alike. The result is a rich body of literary-based reporting of the therapeutic practices that have come to define narrative therapy.

Indeed, the eleven therapeutic qualities associated with externalizing practices that have been outlined in this study have all received considerable scholarly description and theoretical justification across various narrative-inspired authors. This rich and expanding literary enterprise presents like-minded practitioners with an assortment of resources from which to draw upon for their clinical practice. This project adds to that effort in some measure by combining many of these therapeutic descriptions into one study.

Outside of the theoretical work, specific clinical case examples are the typical means for depicting the therapeutic values of narrative practices such as externalizing. Again, these descriptions are richly detailed and harmonious with the theory behind narrative therapy, and are undoubtedly appealing to those practitioners who already espouse postmodern sensibilities. Moreover, the practice of working collaboratively with clients and consulting them about their experiences of the therapy process also has a powerful impact on the manner in which narrative practices are described and justified. Overall, externalizing practices are supported in the narrative literature on multiple fronts, including the following :

  • As a practical therapeutic measure; e.g. speaking in a manner that externalizes the problem and reduces the effects of internalizing discourses that promote feelings of internal defectiveness. Externalizing allows blame to be removed from individuals and families, and creates space for the consideration of alternative choices. Informal feedback from clients often supports these therapeutic claims.
  • Theoretical justification; e.g. externalizing as an anti-language that stands in opposition to dominant forms of social discourse. Externalizing recognizes and makes room for the socially constructed nature of people’s experience. It also allows for conversations that track the history of problems and deconstruct underlying sociocultural and political forces.

However, scholarly work on narrative therapy has been heavily focused in the realm of subjective discussion; i.e. framing narrative practices in theory, therapist accounts of the therapeutic process, incorporation of informal client feedback, as well as the sociocultural and political implications of the narrative worldview. What has been neglected, often purposefully, is an engagement with wider practices within the mental health community to provide supportive (research based) evidence of therapeutic efficacy.

The broader mental health community is actively engaged in various methods of highlighting the efficacy of specific treatment approaches as well as the utility of counseling/psychotherapy in general. Yet, narrative therapy continues to be on the periphery of such efforts.

As a comparison, consider cognitive based therapies for a moment. If therapeutic approaches can have a signature effect on the industry of therapy, that effect which belongs to cognitive based therapies is arguably the robust and systematic manner in which research data is generated and employed to advocate its helpful and functional qualities. Speaking broadly, books on cognitive therapy are also notable for the practical and concise language used to describe cognitive techniques. This is often followed up by an abundance of research data to support such techniques for specific therapeutic goals or with specific client populations.

By contrast, I would argue that one of the hallmarks of narrative therapy has been the painstaking and careful use of postmodern and social constructionist metaphors to situate the narrative approach. The collaborative nature of narrative work and the incorporation of client feedback about therapy are also distinctive of narrative therapy. Yet, I contend that these practices are immediately drawn back into supporting the theory behind the narrative worldview. The result is that scholarly narrative work is a complete effort, but one that resides loyally within its own worldview; a worldview that is at the moment distinct from prevailing cultural practices and sensibilities (i.e. modernistic vs. postmodern).

Whereas cognitive based therapies seem to be benefiting from their focus on specific techniques and utilization of fashionable (i.e. culturally accepted) research practices, narrative therapy remains a less fashionable outsider of sorts. This condition presents unique consequences for the practice of externalizing. Detailed literary accounts of its utility are abundant. Yet, validation of its purported effects and utility as a therapeutic practice are limited because its discussion has traditionally been intimately linked with the larger narrative therapy approach.

Nevertheless, as the axiom goes, the only thing that is certain is change. Indeed, cultural tastes and practices are continually changing. As this occurs, the field of psychology changes and adapts to these evolving sensibilities. It has yet to be determined if our culture will “grow” to accept more enthusiastically the postmodern and narrative therapy sensibilities that are so intimately intertwined. Alternatively, will those practitioners/researchers that advance narrative therapy make accommodations for prevailing mental health views in an effort to garner broader appeal? Those among the vanguard of the narrative approach seem unlikely to temper their political and social perspectives. Alternatively, a separate discussion of externalizing practices combined with concerted qualitative research to validate their utility could have unique appeal to a wider audience in the mental health community, especially in light of the current focus on evidence-based therapy.

Where Would Narrative Therapy be Without Externalizing?

In the introduction of this study, a quote from Michael White is used. In response to the question of where would his work be without externalizing discourse, he responded “nowhere!” In a more recent book, and maybe in light of his comment years prior, White (2004) has written that he often does not use externalizing questions in his work. While there has been a decade between these two comments, and certainly White’s therapeutic work has evolved over that time, what are practitioners to make of this seeming discrepancy?

Indeed, at least one author has described White’s comment of “nowhere” as a “bit of an exaggeration” (O’Hanlon, 1993, p.178). O’Hanlon (1993) suggests that in some respects, externalizing is more of a means than an end in itself; a method for reducing blame and resisting pathologizing language. He notes that White and Epston would likely be helpful without externalizing. O’Hanlon (p.178) adds, “I consider externalizing interventions (discourses) to be, in part, an invitation from the therapist to clients to develop a different relationship to their concerns or their troubles.”

In attempting to find a reasonable solution to this question, this study has attempted to parse out externalizing practices (e.g. specific efforts to externalize problems) from the broader concept of externalizing conversations. As defined earlier, these conversations are intended to represent the therapeutic stance that maintains a constant discourse that views clients as distinct from their socially constructed identities, sense of reality, and troubles.

In light of this distinction, it seems reasonable that White and others may conduct therapy without externalizing problems (i.e. engaging in the practice of externalizing). Yet, based on a review of the literature, the narrative stance that views people as distinct from their socially constructed world is not a feature that can be turned on or off. It’s not a technique or method. Rather, it is more precise to call it a worldview or attitude by which narrative-inspired therapists view the world. It is always in play in the therapeutic alliance between therapist and client. As such, externalizing has a special connection with narrative therapy; one that is deserving of this study to address its qualities and therapeutic effects.




It is my contention that the preceding review of the research on narrative therapy and externalizing practices leads to the following overall conclusions:

  • Research conducted on narrative therapy is qualitative or contains mixed methods in its methodology. No pure quantitative analyses were found among the circulating literature.
  • Narrative therapy continues to receive limited attention among the research-oriented community.
  • Much of the research that involves the use of narrative therapy is often combined with other forms of therapy; for instance motivational interviewing, or group therapy modalities. The result is that it is sometimes difficult to isolate and evaluate the effectiveness of the narrative therapy model.
  • Few studies have been conducted that isolate the usefulness of particular narrative practices such as externalizing. However, the limited research conducted thus far on externalizing practices offers promising implications that deserve attention by narrative practitioners as well as the broader mental health community.
  • Based on the lack of focused research on externalizing practices, it is difficult to parse out the specific therapeutic qualities of externalizing. The limited research and published case studies tend to focus on the effects of reducing blame, removing stigmatizing labels, and opening space for alternative interpretations of problems, events, constraints, etc. Additional focused research would likely shed more light on many of the more nuanced effects of externalizing practices.
  • An informal analysis of the current research suggests that clients may be receiving therapeutic value from externalizing conversations, but when asked, attribute this helpfulness to other therapeutic elements such as personal agency or the therapeutic alliance.
  • It is not just narrative therapy’s purported constructivist or social constructionist principles that have created impediments to the approach being researched. The political and social sensibilities of narrative therapy have at times been in conflict with the goals of researchers.
  • Recent movements within the narrative therapy community are attempting to bring together the narrative worldview with novel forms of qualitative research methodology that are in harmony with one another. The utilization of co-research principles is one such proposed methodology.
  • Co-research relies on externalizing conversations as part of its process. Additionally, co-research would be an excellent component of a qualitative method for collaborating with clients to discuss the helpful properties of externalizing practices. Such a study has yet to be performed in a comprehensive manner.


Implications for the Future

There is a strong argument to be made for future efforts into studying the helpful qualities of externalizing practices. Maione and Chenail (1999) point out that Common Factors research suggests that the specific techniques or applied therapy model is less important than other factors in the course of therapy . However, they concede that more research needs to be done on better understanding the interactions of Common Factors, and they also conclude that:

Although few clinical models have been studied to any great length, the research thus far supports the findings that therapist techniques and models play a role in the change process thus warranting further study. This further study will contribute to our knowledge on basic therapist communicative skills (p.78).

Further exploration into the therapeutic qualities of externalizing would do just that and more. As this paper discusses, externalizing is more than simply a technique. It is rather a purposeful frame of mind and deliberate method of discourse (communication) that invites people to think about their lives and problems in unique ways. Moreover, by studying specific practices such as externalizing more thoroughly, additional data may be generated to illustrate how clients may indeed benefit from specific techniques, but attribute the follow-on therapeutic effects to other broader factors such as the therapeutic relationship or the sense of personal agency. This research could then aid in facilitating understanding of the interactions among the Common Factors.

Specific externalizing practices may well bolster key common factors such as the therapeutic relationship in ways not yet adequately documented in the research; for instance, by making problems less deadly to discuss, reducing feelings of blame and defectiveness, and helping to build personal agency. Additionally, linking research efforts of specific narrative practices to Common Factors research would provide a platform for other mental health professionals to evaluate the therapeutic qualities of narrative therapy. As the name implies, Common Factors research encompasses many therapy modalities.

It certainly appears that specific models of therapy have only been lightly discussed in the research literature (see Maione & Chenail, 1999). In light of this, I propose there is a considerable need to evaluate the helpful qualities of externalizing and other practices associated with narrative therapy. Moreover, I recommend that evaluating externalizing practices would be the first logical step in any such effort for three main reasons. First, studies already exist that have, to various degrees, researched the use of narrative therapy. Second, several of these studies suggest that externalizing has useful therapeutic effects with particular issues and client populations. Third, it is my contention that externalizing is a uniquely important practice within the narrative therapy approach. As such, research highlighting its therapeutic qualities would be beneficial not only to refining the practice of externalizing, but it would have implications for the acceptance of narrative therapy in general. For instance, research on externalizing practices could shed light on the helpful qualities associated with the general externalizing way of thinking inherent to the narrative therapy approach.

It is important to spend some effort separating out the specific practices of narrative therapy in order to accurately assess their therapeutic qualities. As this study has suggested, those research efforts that have only studied narrative as a general model, while useful, have typically failed to adequately discuss the qualities of the model that are helpful to people. It is time for a more thorough investigation into the therapeutic qualities of narrative therapy. A good place to begin such research is with a focus on the effects of externalizing conversations.


Future Research Methods

As discussed earlier, elements of the narrative therapy worldview present unique obstacles to conducting research (see page 49). Yet, rather than becoming a deterrent, these obstacles can invite vigor as narrative-inspired researchers work to integrate narrative principles with evolving research methodology. Indeed, some are already attempting to do this (see Denborough, 2004; Gaddis, 2004; Redstone, 2004).

These efforts often involve explorations via co-research, such as that discussed by Epston (1999, 2004) and Maisel et. al. (2004). In addition, Madsen (1999) discusses how co-research projects can be fitted into a qualitative manner of outcome research that identifies the helpful themes or qualities in therapy. He notes that, “Emergent themes across interviews could be drawn out to develop grounded theory about contributions to positive outcome” (p.334). Other promising qualitative research methods include interpersonal process recall, comprehensive process analysis, ethnography, and discourse analysis. Indeed, the possibilities for making use of co-research in the construction of a research methodology have been addressed in the literature. It now simply needs to be acted upon in a thorough manner.

While narrative practitioners often consult with clients about their impressions of therapy and specific practices, more work needs to be done in documenting these accounts in a methodological research format. It is my contention that such projects would benefit the field of narrative therapy by ensuring that practitioner theory does not supersede client accounts of what was helpful in the therapy process. There would also be useful ramifications for externalizing practices. The current research literature reveals that externalizing indeed creates some distinctive effects upon clients. Understanding these effects to a greater degree is a worthy research goal not only for narrative practitioners, but for all clinicians who wish to understand how the use of discourse with clients can affect the therapeutic process.

Proposals for Research

In this section, I would like to offer two sketches for potential research projects that, (1) would honor narrative therapy sensibilities concerning the roles of the therapist and client, and (2) would be useful for examining the therapeutic qualities of narrative therapy practices.

The first example is a research method premised on Comprehensive Process Analysis (CPA; Elliot & Shapiro, 1992) that could also allow for the implementation of narrative-inspired co-research. CPA is a systemic, qualitative research procedure for analyzing significant events. It seeks to understand the following:

(a) the context out of which significant events arise (e.g., client coping style, therapeutic alliance),

(b) the important features of the event itself (e.g., therapist interpretation, client expression of interpersonal fears), and

(c) the impacts of the event (e.g., insight, decreased depression)

(Elliot & Shapiro, 1992, p.164-165).

CPA utilizes a Brief Structured Recall procedure (BSR; Elliot & Shapiro, 1992). This procedure is similar to interpersonal process recall (discussed earlier in this study) that has clients and therapists review audio or videotapes of entire sessions. The BSR procedure, however, only plays back those portions of the session identified by the client as important. The process of determining which events are important to the client can be accomplished either as a collaborative effort between the therapist and client, or determined by the client via the completion of a questionnaire following the session(s).

In the CPA method outlined by Elliot & Shapiro (1992), a researcher then interviews the client and therapist separately about the selected events; asking about the context of the event, the event itself, and its impact. The process results in separate testimonies about how each party felt a particular event was therapeutic and useful in the session.

A more narrative-inspired alternative would be a collaborative co-research session between the therapist and client, in which the selected events are discussed in a similar format. While this option may raise concerns such as possible therapist bias or clients saying things they think the therapist wishes to hear (social desirability bias), it does match more closely the spirit of narrative co-research. Therapist and client would be intimately involved in exploring those events in the session that were found to be therapeutic by the client. If this type of research is accumulated across many sessions and among various clients and therapists, common themes and therapeutic qualities can be identified and coded. This would be helpful in identifying specific therapeutic practices, their context, and the impact they have on clients.

As a second example, co-research projects can also be an excellent method for evaluating with clients the helpful and unhelpful qualities of externalizing practices. Narrative-inspired practitioners have expressed concern that more traditional forms of research emphasize professional accounts of therapy over those of clients (see Gaddis, 2004). However, employing the collaborative spirit of co-research allows the therapist to both define a research question – in this case, exploring the helpful and unhelpful qualities of externalizing conversations – and work in conjunction with clients to address the question in a way that taps the local knowledge and experience of the client. See the figure on page 56 for an illustration of co-research as a research project. This approach would allow for focused co-research projects on the effects of externalizing discourse.

This study has charted how the externalizing stance of narrative conversations is a pervasive feature of the narrative therapy model. As such, externalizing practices warrant specific investigation. By collecting co-researched accounts, identifying and codifying themes (again, possibly with client assistance), an accumulation of client-generated data on the therapeutic qualities of externalizing conversations would be made possible.

Co-research as a methodology is not without its limitations. Similar to that described above, the client or therapist may be led by ulterior motives to make claims or inquires that may not necessarily speak appropriately to the therapeutic process. Moreover, an approach that relies primarily on co-research may find a mixed reception among other in the mental health community. Only time and effort will tell if these or other methods prove useful in furthering clinicians understanding of what works and does not work for clients in therapy.




Amundson, J., Webber, Z., & Stewart, K. (2000). How narrative therapy might avoid the same damn thing over and over. Journal of Systemic Therapies. Vol.19, No.4. 20-31.

Andersen, T. (1987). The reflecting team: Dialogue and metadialogue in clinical work. Family Process. 26. 415-428.

Anderson, H., & Goolishian, H. A. (1988). Human systems as linguistic systems: Preliminary and evolving ideas about the implications for clinical therapy. Family Process, 27, pp.371-393.

Anderson, H., & Goolishian, H. A. (1992). The client is the expert: A not-knowing approach to therapy. In S. McNamee & K. Gergen (Eds.), Social construction and the therapeutic process (pp. 25-39). Newbury Park, CA: Sage Publications.

Archer, R. (1997). Tornadoes, boys, and superheroes: Externalizing conversations in the wake of a natural disaster. Journal of Systemic Therapies. Vol. 16, No. 1. 73-82.

Bateson, G. (1972). Steps to an ecology of mind. New York: Ballantine Books.

Bateson, G. (1979). Mind and nature: A necessary unity. New York: Dutton.

Besa, D. (1994). Evaluating narrative family therapy using single-system research designs. Research on Social Work Practice. Vol. 4, No. 3, 309-325.

Beutler, L. E., Bongar, B., & Shurkin, J. N. (1998). A consumer’s guide to psychotherapy. New York: Oxford University Press.

Bruner, E. & Turner, V. (Eds) (1986). The anthropology of experience. Chicago: University of Illinois Press.

Bruner, J. (1986). Actual minds, possible worlds. Cambridge: Harvard University Press.

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

Carey, M. & Russell, S. (2002). International Journal of Narrative and Community Work. No.2.

Carey, M. & Russell, S. (2004). Narrative therapy: Responding to your questions. Dulwich Centre Publications.

Chang, J. & Phillips, M. (1993). Michael White and Steve de Shazer: New directions in family therapy. In Gilligan S. & Price, R. (Eds.), Therapeutic conversations. New York: W.W. Norton & Company.

Clark, W., Jankowski, P., Margee, & Springer, N. (2000). Searching for a fit: Utilizing a co-research team as a qualitative method of analysis. Journal of Systemic Therapies. Vol. 19. No. 2. 56-64.

Couture, S. & Sutherland, O. (2004). Investigating change: Compatible research and practice. Journal of Systemic Therapies. Vol. 23, No. 2. 3-17.

Cowley, C., Farley, T. & Beamis, K. (2002). “Well, maybe I’ll try the pill for just a few months…” Brief motivational and narrative-based interventions to encourage contraceptive use among adolescents at risk for early childbearing. Families, Systems & Health. Vol. 20, No. 2. 183-204.

Crocket, K. (2004). From narrative practice in counseling to narrative practice in research: A professional identity story. The International Journal of Narrative Therapy and Community Work. No. 2. Dulwich Centre Publications. 63-67.

de Shazer, S. (1988). Clues: Investigating solutions in brief therapy. New York: Norton.

de Shazer, S. (1993). Creative misunderstanding: There is no escape from language. In Gilligan S. & Price, R. (Eds.), Therapeutic conversations. New York: W.W. Norton & Company.

Denborough, D. (Ed.). (2004) Narrative therapy and research. The International Journal of Narrative Therapy and Community Work. No. 2. Dulwich Centre Publications. 29-35.

Derrida, J. (1981). Positions. Chicago: University of Chicago Press.

Diamond, J. (2000). Narrative means to sober ends: Treating addiction and its aftermath. New York: Guilford Press.

Elliot, R. (1986). Interpersonal process research as a psychotherapy process research method. In Greenberg, L. & Pinsof, W. (Eds.). The Psychotherapeutic Process. New York: Guilford Press.

Elliot, R. & Shapiro, D. (1992). Client and therapist as analysts of significant events. In Toukmanian, S. & Rennie, D. (Eds.). Psychotherapy process research: Paradigmatic and narrative approaches. Newbury Park: SAGE Publications.

Epston, D. & White, M. (1992). Experience, contradiction, narrative & imagination: Selected papers of David Epston & Michael White 1989-1991. Adelaide, South Australia: Dulwich Centre Publications.

Epston, D. (1993). Internalizing discourses versus externalizing discourses. In S. Gilligan & R. Price (Eds.), Therapeutic conversations (pp. 161-177). New York: Norton.

Epston, D. (1999). Co-research: The making of an alternative knowledge. In Narrative and Community Work: A conference collection. 137-157. Adelaide, South Australia: Dulwich Centre Publications.

Epston, D. (2001). Anthropology, archives, co-research and narrative therapy. In Denborough, D. (Ed.). Family therapy: exploring the field’s past, present and possible futures. Adelaide, South Australia: Dulwich Centre Publications.

Epston, D. (2004). From empathy to ethnography: The origin of therapeutic co-research. In Denborough, D. (Ed.). (2004) Narrative therapy and research. The International Journal of Narrative Therapy and Community Work. No. 2. Dulwich Centre Publications. 29-35.

Epston, D. (2005). Private notes presented during February 2005 workshop in Berkeley California.

Etchison, M. & Kleist, D. (2000). Review of narrative therapy: Research and utility. The Family Journal: Counseling and Therapy for Couples and Families. Vol.8, No.1, 61-66.

Fetterman, D. (1989). Ethnography step by step. Newbury Park, CA: Sage Publications.

Fish, V. (1993). Poststructuralism in family therapy: Interrogating the narrative/conversational mode. Journal of Narrative and Marital Family Therapy. 19 (3), 211-232.

Foucault, M. (1979). Discipline and punish: The birth of the prison. Middlesex, England: Peregrine Books.

Foucault, M. (1980). Power/knowledge: Selected interviews and other writings. New York: Pantheon.

Freedman, J., & Combs, G. (1993). Invitations to new stories: Using questions to explore alternative possibilities. In Gilligan, S. & Price, R. (Eds.), Therapeutic conversations. New York: W.W. Norton & Company.

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

Freeman, J., Epston, D. & Lobovits, D. (1997). Playful approaches to serious problems: Narrative therapy with children and their families. New York: W.W. Norton & Company.

Fristad, M., Gavazzi, S. & Soldano, K. (1999). Naming the enemy: Learning to differentiate mood disorder “symptoms” from the “self” that experiences them. Journal of Family Psychotherapy. Vol. 10 (1). 81-88.

Gale, J. (1992). When research interviews are more therapeutic than therapy interviews. The Qualitative Report. [Internet-based publication], 1(4). Available at:

Gale, J. (1993). A field guide to qualitative inquiry and its clinical relevance. Contemporary Family Therapy. 15.

Gaddis, S. (2002). Narrative means to research ends: Learning about therapy from clients’ descriptions. A Clinical Dissertation. Syracuse University.

Gaddis, S. (2004). Re-positioning traditional research: Centering clients’ accounts in the construction of professional therapy knowledges. The International Journal of Narrative Therapy and Community Work. No. 2. Dulwich Centre Publications. 37-47.

Geertz, C. (1983). Local knowledge: Further essays in interpretive anthropology. New York: Basic Books.

Gergen, K. (2001). An invitation to social construction. London: Sage Publications.

Goddard, J., Lehr, R. & Lapadat, J. (2000). Parents of children with disabilities: Telling a different story. Canadian Journal of Counseling. Vol. 34 (4). 273-289.

Hare-Mustin, R. (1994). Discourses in the mirrored room: A postmodern analysis of therapy. Family Process. 33. 19-35.

Hertlein, K., Lambert-Shute, J., & Benson, K. (2004). Postmodern influence in family therapy research: Reflections of graduate students. The Qualitative Report. Vol. 9, No. 3. 538-561.

Huntington, J.F. (2001). A hermeneutical critique of narrative therapy. A Clinical Dissertation. California School of Professional Psychology.

Ingram, C. & Perlesz, A. (2004) The getting of wisdoms. The International Journal of Narrative Therapy and Community Work. No. 2. Dulwich Centre Publications. 49-56.

Kahle P. & Robbins, J. (1998) Re-authoring therapeutic success: Externalizing the success and unpacking marginalized narratives of competence. Journal of Systemic Therapies. Vol. 17, No. 3. 58-68.

Kaptain, D.C. (2004). Narrative group therapy with outpatient adolescents. A Clinical Dissertation. Iowa State University.

Kazdin, A.E. (Ed. in chief). (2000). Encyclopedia of Psychology, Volume 5. Oxford: Oxford University.

Knight, K., Bundy, C., Morris, R., Higgs, J., Jameson, R., Unsworth, P., & Jayson, D. (2003). The effects of group motivational interviewing and externalizing conversations for adolescents with type-1 diabetes. Psychology, Health & Medicine. Vol.8, No.2. 149-157.

Madsen, W. (1999). Collaborative therapy with multi-stressed families: From old problems to new futures. New York: Guilford Press.

Maione, P. & Chenail, R. (1999). Qualitative inquiry in psychotherapy: Research on the common factors. In M.A. Hubble, B.L. Duncan, & S.D. Miller (Eds.), The Heart and Soul of Change: The Role of Common Factors in Psychotherapy. Washington, DC: American Psychological Press.

Maisel, R., Epston, D. & Borden, A. (2004). Biting the hand that starves you: Inspiring resistance to anorexia/bulimia. New York: W.W. Norton.

McNamee, S. (1988). Accepting research as social intervention: Implications of a systemic epistemology. Communications Quarterly. 36. 50-68.

Merchant, N. (1997). Qualitative research for counselors. Counseling & Human Development. 30(1).

Meyerhoff, B. (1982). Life history among the elderly: Performance, visibility and remembering. In Ruby, J. (Ed.), A crack in the mirror: Reflexive perspectives in anthropology. Philadelphia: University of Pennsylvania Press.

Miller, S., Hubble, M., & Duncan, B. (Eds) (1996). Handbook of solution-focused brief therapy. Jossey-Bass Inc. San Francisco: CA.

Monk, G. (1996). Narrative approaches to therapy: The Fourth Wave. Guidance & Counseling, 11(2), 41.

Monk, G., Winslade, J., Crocket, K., & Epston, D. (1997). Narrative therapy in practice: The archaeology of hope. San Francisco: Jossey-Bass.

Morgan, A. (2000). What is narrative therapy? An easy to read introduction. Dulwich Centre Publications.

Muntigl, P. (2004). Ontogenesis in narrative therapy: A linguistic-semiotic examination of client change. Family Process. Vol. 43, No. 1. 109-131.

Nelson, M. & Poulin, K. (1997). Methods of constructivist enquiry. In Sexton, T. & Griffin, B. (Eds.). Constructivist thinking in counseling practice, research, and training. New York: Teachers College Press.

Nylund, D. (2000). Treating Huckleberry Finn: A new narrative approach to working with kids diagnosed ADD/ADHD. San Francisco: Jossey-Bass.

O’Connor, T., Davis, A., Meakes, E., Pickering, M., & Schuman, M. (2004). Narrative therapy using a reflecting team: An ethnographic study of therapists’ experiences. Contemporary Family Therapy. 26(1), 23-39.

O’Connor, T., Meakes, E., Pickering, M., & Schuman, M. (1997). On the right track: Client experience of narrative therapy. Contemporary Family Therapy. 19(4), 479-495.

Pare, D. & Lysack, M. (2004) The Willow and the oak: From monologue to dialogue in the scaffolding of therapeutic conversations. Journal of Systemic Therapies. Vol.23, Iss.1.

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

Redstone, A. (2004) Researching people’s experience of narrative therapy: Acknowledging the contribution of the “client” to what works in counseling conversations. The International Journal of Narrative Therapy and Community Work. No. 2. Dulwich Centre Publications.

Rosaldo, R. (1992). Culture and truth: The remaking of social analysis. Boston: Beacon Press.

Selekman, M. (1997). Solution-focused therapy with children: Harnessing family strengths for systemic change. New York: Guilford Press.

Stacey, K. & Hills, D. (2001). More than protest: Further explorations of alternative metaphors in narrative therapy. Australian and New Zealand Journal of Family Therapy. Vol. 22 (3). 120-128.

Strauss, A. & Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park, CA: Sage.

Thomas, L. (2002) Poststructuralism and therapy; what’s it all about? The International Journal of Narrative Therapy and Community Work. No. 2. Dulwich Centre Publications. 85-89.

Tomm, K. (1989). Externalizing the Problem and Internalizing Personal Agency. Journal of Strategic and Systemic Therapies. Vol. 8 (1) Spring.

Toukmanian, S. & Rennie, D. (Eds.).(1992). Psychotherapy process research: Paradigmatic and narrative approaches. Newbury Park: SAGE Publications.

Turner, V. (1969). The ritual process. Ithaca, NY: Cornell University Press.

White, M.(1986). Negative explanation, restraint and double description: A template for family therapy. Family Process, 25(2).

White, M. (1987). Family Therapy and Schizophrenia: Addressing the In-the-corner Lifestyle. Dulwich Centre Newsletter. 14-21.

White, M. (Summer, 1989/1997). The externalization of the problem and re-authoring of lives and relationships. Dulwich Centre Newsletter (pp. 3-20). Reprinted in M. White & D. Epston, Retracing the past: Selected papers and collected papers revisited. Adelaide, Australia: Dulwich Centre Publications.

White, M. (1991/1993). Deconstruction and therapy. Dulwich Centre Newsletter, No. 3 (pp. 21-40). Reprinted in Gilligan S. & Price, R. (Eds.), Therapeutic conversations. New York: W.W. Norton & Company.

White, M. (1995). Re-authoring lives: Interviews & essays. Adelaide, South Australia: Dulwich Centre publications.

White, M. (2000). Reflections on narrative practices: Essays and interviews. Adelaide, South Australia: Dulwich Centre publications.

White, M. (2004). Narrative practice and exotic lives: Resurrecting diversity in everyday life. Adelaide, South Australia: Dulwich Centre publications.

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

Wright, L. (1990). Research as a family therapy intervention technique. Contemporary Family Therapy. 12. 477-484.