Anorexia nervosa: A fresh perspective

“Risk factors” for anorexia

Discourses around what are termed “risk factors” for anorexia nervosa are also worth mentioning as they provide a richer understanding of the complexity of the multitude of discourses that surround and intersect the constitution of aetiological knowledges of anorexia (Tozzi et al., 2003). Halmi (2005) compiled suggested risk factors obtained from correlation and comparative studies. They were grouped as familial, individual biological, individual psychological, individual behaviours, cultural, and stressful life events.

Existing data on risk factors for eating disorders have been comprehensively evaluated (see, e.g., Bulik, Reba, Siega-Riz, & Reichborn-Kjennerud, 2005; Jacobi, Hayward, de Zaan, Kraemer, & Agras, 2004). Bulik et al. (2005) reviewed these evaluated results and commented on them in regard to anorexia. Their views are, inter alia, that:

For AN, the results are sobering. First, few longitudinal studies exist in which sufficient numbers of cases have been identified to enable the identification of risk factors. Second, for AN in particular, it is difficult to differentiate between early symptoms of the disorder and risk factors (e.g., dieting and high exercise levels). Third, few studies exist that enable an exploration of specificity of risk factors across the three eating disorder subtypes, with outcome variables often crossing both diagnostic and threshold boundaries. (p. S4)

It must be emphasised that in spite of the complexities in the aetiology and risk factors for anorexia, it is understood by dominant discourses to be an internalised, psycho-medical feminine phenomenon, ² that is, a person is anorexic, has anorexia, suffers from anorexia, and so on—in other words, a person is pathologised, and in some way personifies anorexic discourses.

Dominant treatment discourses: Foundational models

A number of psychological treatments have been used to target different aspects of anorexia, including distorted thoughts regarding food and body image; dysfunctional behaviours such as restrictive eating, binging, and purging; and interpersonal and intra-psychic issues (Bodell & Keel, 2010).

Attempts to influence the symptoms and prognosis of anorexia include (but are not limited to) ³ dietetic (Mehler & Crews, 2001), pharmacological (Treasure & Schmidt, 2002), psychodynamic (Dare, Eisler, Russell, Treasure, & Dodge, 2001), behavioural (Schmidt, 1989), behavioural and cognitive-behavioural (Gowers & Bryant-Waugh, 2004), group therapies (Gowers & Bryant-Waugh, 2004), feminist therapies (Fallon, Katzman, & Wooley, 1994; Orbach, 1978, 1986), integrated therapies (Steiger & Israel, 1999), and motivational enhancement programmes (Vitousek, Watson, & Wilson, 1998).

Most of these approaches are significantly informed by three basic psychotherapeutic models: psychoanalytic, cognitive-behavioural, and humanistic (Corey, 2005). These represent distinctive developments in psychotherapy, though each model has continued to evolve both within itself and in dialogue with the others; and they share the structuralist conceptualisation described above. The three models continue to dominate, providing foundational beliefs, assumptions, and practices to over 400 current psychotherapeutic approaches (Peavy, 1996).

Objectifying, pathologising and normalising practices in structuralist therapy

Over the past century, western culture has been pervaded by understandings of human action as expressing “internal state” identities, that is, actions depicted as expressions of essential structures of the self, located at the core of identity (White, 2007). Popular and professional acceptance of this “true self” is now so widespread that any human action or expression is liable to be interpreted as “a surface manifestation of these internal states” (White, 2004, p. 21). According to this dominant version, human life and personal subjectivity emanates either from the direct, “natural” expression of these self-essences (resulting in normative behaviour patterns), or else from distortions that, being “abnormal,” are termed “disorders” or “dysfunctions” (White, 2007, p. 101). According to Foucault (1971, 1976), it is the principal task of “medical perception” to construct the terminology of abnormality; and of distortions relative to “normality,” of “disorder,” and “dysfunction” (White, 2007, pp. 102–103).

Whether in professional or lay imaginations, the behaviour and attitudes of persons living with anorexia are contrasted with a socially derived “norm”; and anorexia is objectified as a pathology, not merely by recording food refusal, metabolic or hormonal change, and weight loss, but also by documenting the disordered personality, and possibly the “dysfunctional” family of “the anorexic.”

This approach is typified in the DSM-5 (APA, 2013). This authoritative source reinforces the dominant social construction of anorexia as internally objectified, that is, as being “in the individual” and as a disorder that people “have.” A similar representation of the medical model of anorexia nervosa appears in guidelines of the National Institute for Health and Clinical Excellence (NICE, 2004a, 2004b).

In deconstructing the objectifying and pathologising effects that are taken for granted in most psychotherapy models, Kaye (1999) submits that these models are informed by four assumptions: there is an underlying cause or basis of pathology; the location of this cause is within individuals; the problem is diagnosable (against socially constructed criteria of normalcy); and lastly, the problem is treatable via a specifically designed set of techniques. Treatment involves “a restructuring or reprogramming of behaviour in both individuals and families against some criterion of the normal, the deviant, the well-adjusted, the problematic and non-problematic” (Kaye, 1999, p. 21).

Dominant models of psychotherapy assume that therapists have the necessary linguistic and theoretical frame to discover and to solve the problem of anorexia (Kaye, 1999), even if people who live in relationship with the “disorder” do not experience this to be the case. A hierarchy has been established that privileges expert perspectives over those of clients, based on the belief that practices of psychological diagnosis and treatment are objective, “scientific,” and thus also authoritative. Within this discursive hierarchy, notions and measures of “universal truth” devalue, dishonour, disempower, and disenfranchise competing accounts, including hard-won knowledges and practices of resistance, survival, connection, love, sacrifice, strengthening of identity, and so on. Worse still, adherence to alternative accounts by persons living with anorexia (or their families) has sometimes been viewed as further evidence of “disorder.”

White argues that power, as recognised and sanctioned in western societies, is integral to the ascendency of structural accounts of mental illness, and associated technologies of diagnosis and treatment. As mentioned above, people subjected to this “objective,” modernist technology of power tend to be regarded as objects and are further positioned as passive, powerless beneficiaries of “expert” knowledge (Freedman & Combs, 1996). Following Foucault (1976, 1978, 1979, 1980a, 1980b, 1986, 1990), White interprets this “new technology of power” as a mechanism that, inter alia, “establishes an effective system of social control through what can be referred to as ‘normalising judgement’”:

The normalising judgement of people’s lives has been made possible through the development of a whole new technology of power that employs various schemes and continuums of normality/abnormality, tables of performance, scales for the rating of every human expression imaginable, and formulae for the ranking of persons in relation to each other. This technology of power also includes specific practices of evaluation that have the effect of inserting people’s lives into these schemes, continuums, tables, and scales. (2002, p. 43)

Furthermore, in much of today’s world the mental health disciplines are sites of (private) economic production, wherein practitioners add value and prestige by locating themselves within the paradigm of the medical model, without which it is impossible to obtain funding from health services or insurance providers. This is problematic, not only from a position that attempts to critique the paradigm of the medical model, but also from positions within the dominant paradigm.

Overview of efficacy and effectiveness of treatment approaches for anorexia

Since Freud and psychoanalysis, psychotherapies have claimed locations within the field of scientific epistemology, and have conducted research within the ambit of structuralist assumptions drawn from physical and natural sciences (Roy-Chowdhury, 2003). This opens psychotherapies to hypothesis testing, experimentation, verification through empirical observation and measurement, and a general responsibility to contribute to the “advance of science”; or at the very least to solve outstanding problems by scientific means. Within this paradigm, the randomised controlled trial is considered a superior means of testing interventions against other interventions, or against “placebo effect.” An efficacy study of this type features monitored, manualised treatments, randomised assignment of participants to treatment groups, and a control/comparison group (Howard, Monash, Brill, Martinovich, & Lutz, 1996). A naturalistic study that attempts to measure psychotherapy outcomes is an effectiveness study. A sizeable body of data has now been amassed regarding the efficacy and effectiveness of psychotherapies for eating disorders (e.g., Ben-Tovim et al., 2001; Dare et al., 2001; Eisler et al., 2000; Gowers et al., 2007; Lock, Agras, Bryson, & Kraemer, 2005; Lock & Litt, 2003; McIntosh et al., 2005; McIntosh, Jordan, Luty, Carter, & McKenzie, 2006; Pike, Walsh, Vitousek, Wilson, & Bauer, 2003; Richards et al., 2000; Steinhausen, 2002).

Technologies of treatment for anorexia nervosa do not emerge well from this body of research, which indicates limited and restricted “success” (Bergh, Brodin, Lindberg, & Sodersten, 2002; Bulik, Berkman, Brownley, Sedway, & Lohr, 2007; Button & Warren, 2001; Eckert, Halmi, Marchi, Grove, & Crosby, 1995; Kaplan & Garfinkel, 1999; Levenkron, 2000; Löwe et al., 2001; Richards et al., 2000). Rates of “recovery” are relatively low at between 11% and 40% (Ben-Tovim, 2003; Richards et al., 2000; Von Holle et al., 2008). Relapse is frequent (Eckert et al., 1995; Strober, 2010), as is the emergence of chronic symptoms of bulimia (Eddy et al., 2002). Other researchers found that former “anorexics” often continue to experience social and psychological problems (Button & Warren, 2001; Keel, Mitchell, Miller, Davis, & Crow, 2000); and the mortality rate for anorexia nervosa has persisted at around 10.5%, one of the highest in any area of psychiatry and psychology (Birmingham, Su, Hlynsky, Goldner, & Gao, 2005). Since most people first diagnosed with anorexia are young, the picture emerging from therapy research is exceptionally disturbing. Yet, it appears that such persons frequently do not appreciate professional interventions. Research indicates high drop-out rates (Eivors, Button, Warner, & Turner, 2003; Mahon, 2000), low levels of satisfaction among users of treatments for eating disorders; and increasing levels of chronicity for anorexia (Strober, 2004a, 2004b, 2010)—in itself a statistical predictor of worse outcomes.

Kaplan (2002) reviewed literature on psychological treatments for anorexia (especially randomised controlled clinical trials published over the past three decades) and identified less than 20 such trials, concluding that “there is little evidence on which to base treatment decisions regarding the psychological treatments for anorexia nervosa” (p. 235). A review by Steinhausen (2002) considered 119 outcome studies (of variable quality and published between 1953 and 1999), and found that “there was no convincing evidence that the outcome of anorexia nervosa improved over the second half of the last century” (p. 1284).

Bulik et al. (2007) conducted a systematic review of randomised controlled outcome trials from 1980 to September 2005 in all languages. Their selection criteria identified 32 studies, with 13 studies set aside as being of “poor” quality. This review found “the strength of evidence [for treatment effectiveness] to be variable, but generally unimpressive” (p. 317). Evidence for treatment efficacy was weak (p. 310); there was tentative evidence that cognitive-behavioural therapy may reduce relapse risk for adults after weight restoration is accomplished, but its efficacy in the underweight state and for adolescents remained unknown. These researchers found that family therapy had no supportive evidence for adults, but noted two studies suggesting that family therapy is superior to individual therapy for adolescents with shorter duration of illness. This review offers a rare glimpse of hope, implying that moving away from internalised, individualised pathology in order to focus on (relational collaborative) strengths, skills, and relationships could provide a source of hope and help for those living in relationship with anorexia (see below, Family therapy and Fresh approaches sections).

Lastly, the results of an Australian study (Ben-Tovim et al., 2001) of patients being treated for anorexia were in line with the general findings of Steinhausen (2002). Ben-Tovim (2003) agrees with Steinhausen that “the outcome of anorexia does not appear to have altered in the last 50 years,” a situation which “points to a field in need of fresh approaches” (p. 65).

Thus, while quality and results of research vary considerably, there is emphatic consensus that outcomes are unsatisfactory at best; and that with limited, provisional exceptions, no one therapy stands above the others. It appears, therefore, that “understandings” of anorexia based on structural theory have not been followed with effective technologies of treatment.

Family therapy

The concept of mental illness having specific family-related aetiologies was established by Minuchin (1974) in his structural family therapy. Minuchin’s approach featured a firm emphasis on pathological interactive familial processes as the cause, means of development, and sustenance of anorexia, and other disorders. This type of family therapy determined its goals and methods within a larger project of “normalising” “dysfunctional” families.

By the late 1970s an embryonic shift began in the direction of alternative family therapy models for anorexia (Dare, 1986; Dare & Eisler, 1997). These alternatives were initial challenges to the medical model as they were concerned with “facilitation of emotional communication and emotional literacy, and assisting family members in developing skills to better negotiate differences of opinion and attitude, recognizing that some rigidity of behavior and emotionality are at times associated with [anorexia nervosa]” (Le Grange, Lock, Loeb, & Nichols, 2010, p. 2). Furthermore, the Academy for Eating Disorders suggested that “families should be involved routinely in the treatment of most young people with an eating disorder. Exactly how such involvement should be structured, and how it will be most helpful will vary from family to family” (Le Grange et al., 2010, p. 4). These shifts in understandings led to a model for family therapy being developed as an outpatient treatment for anorexia at the Maudsley Hospital in London (Lock, Le Grange, Agras, & Dare, 2001). It is an innovative integration of a number of models of family therapy, where parents are seen as an important collaborative resource in the process of recovery, and prior notions of familial aetiology are rejected. Use of the model has been promising, with improvements in clinical outcomes that include significant drops in re-admission rates (Rhodes & Madden, 2005).

Expanding therapeutic arrangements

It may be timeous to explore less confining spaces than are currently constructed by dominant discourses around anorexia. The promising developments in psychotherapeutic treatments using forms of “family therapy” give impetus to such an exploration. The concept of “family therapy” offers a shift from viewing understandings of anorexia from an intra-psychic perspective to a focus on interactional or relational frameworks in which “anorexic behaviour” is performed. This approach offers different (post-modernist, post-structuralist) ways of talking about and locating anorexia, and thus a different treatment focus. In addition, the notion of “family therapy,” that is a therapeutic system of relational, interactional discourses, opens the process of psychotherapy for those in relationship with anorexia to something that can be openly observed, studied, and shared. In this process, the concept of “family” as an arrangement for collaborative use in therapeutic practices can be broadened to incorporate other persons (“members”) who also have an interest in and some relationship with the problem. This could create a more extensive but inclusive relational set-up that coalesces around the “problem,” which is usually collaboratively constructed as a person’s relationship with anorexia. This more inclusive approach may mean involving school friends, teachers, family doctors, sports coaches, and any other persons with a relational, communicative, relevant positioning with the “problem.” In other words the nature of the “problem” could determine the cluster of communicating persons in a broader collaborative, relational therapeutic arrangement (Anderson, 1997; Gergen, 2009b; Shotter, 1995; see also the concept of Open Dialogue: Aaltonen, Seikkula, & Lehtinen, 2011; Seikkula, 2008; Seikkula et al., 2006; Seikkula & Trimble, 2005). Such arrangements have already been formed in spaces and places beyond the traditional, and are evidence of the power of shared collaborative activities in times of need (Epston, 2000; see: The archive of resistance: anti-anorexia/anti-bulimia: it is listed as “A lifesaving archive of personal stories, essays, poetry, art, scholarship, and conversations about the body, anorexia, bulimia, perfectionism, and identity”). ⁴

From systems to stories: A collaborative narrative approach to therapy^5,6

Suggested collaborative approaches to alternative forms of psychotherapy for persons in relationship with anorexia could be drawn from the philosophy and ways of working in narrative therapy. It is a form of post-structuralist psychotherapy that is used in various collaborative, relational settings, including in both individual and community work. It is suggested as an appropriate alternative practice for persons in relationship with anorexia as it is based on an understanding that the possibilities of making meaning of human experiences arise from the social settings in which we live. It has been used successfully with adolescent boys in an outpatient setting (Kronbichler, 2004), where narrative ways of working were experienced as “helpful and effective.” It is almost a response to the disabling power and processes of modernist, structuralist ways of therapy, which have not been of any significant benefit to date for the treatment of anorexia. The shift from an individualised, psychologised notion of a person to a socio-cultural perspective requires the embracing and formulation of a new way of communicating in order to present and work with a different type of therapeutic process.

In contrast to the normative hierarchical relationship between therapist and client, narrative ways invite therapists to an exploratory and investigative position that implies no privileged access to the “truth” of a problem. Co-operative practices of narrative work require the therapist to relate to the client as a partner with specific expertise about their lived experiences, and ways of making meaning thereof in the construction of their subjectivities. A narrative therapeutic process is a deliberate yet respectful and reciprocal activity (Freedman & Combs, 1996) that de-constructs problematic meanings of life and ways of living constituted by dominant social discourses. It also embraces a re-constructive process, unearthing dormant competencies, skills, talents, abilities, and resources that inform the re-authoring of preferred ways of living and subjectivities. All members of a collaborative therapeutic arrangement constructively interact with others in this process, thus providing different understandings of identified problems, and of possibilities and support for a person in relationship with anorexia.

The opportunity for persons in relationship with anorexia to express their experiences in such therapeutic relationship that does not rely on subjugation of their knowledge of themselves, but, in fact, fosters an articulation of that knowledge, allows for the emergence of counter-plots to dominant totalising discourses of anorexia. A realisation that anorexia does not dominate the whole of who they are, or may be, can offer further encouragement to them to explore possibilities of acts of resistance to the influences of dominant discourses in their lives, and then to re-author (re-construct) their lives in preferred ways.

Opening spaces for new stories: Some aspects of and for narrative therapeutic practice with persons in relationship with anorexia

This section presents and discusses some of the more important assumptions and understandings that form the foundation of narrative ways of practising therapy; indicates some of the main differences between this post-structuralist approach to therapy and modernist, structuralist approaches; mentions the benefits of a post-structuralist, narrative approach to therapy; and offers a cautionary note in some circumstances. This discussion tends to follow a linear trajectory, with a notional beginning, middle, and end. This is done in an attempt to facilitate an understanding of working in narrative ways. However, I am fully aware of the fact that therapeutic processes, in the majority of cases, are far from linear, neat, and orderly, especially when therapist and client (and members of a larger therapeutic system) are co-authoring a preferred subjectivity with a client.

Initially, an exploration is undertaken with the client of the experiences that contribute to their making meaning of their dominant life story(ies). This is to examine the nature of dominant discourses that inform and constitute “problem-saturated” stories, and that, in turn, shape their subjectivities (identities). The exploration is focused on how clients construct their subjectivities on the basis of their specific experiences. In addition to revealing these constructed subjectivities, how persons in relationship with anorexia position themselves in their story lines in response to the influences of the disciplinary powers of dominant discourses is examined. The aim is not to “discover” the “truth” about who they are, but to explore how they construct modernist, structuralist “truths,” and the “reality” of how they see themselves and their relationships, through their experiences of being subject to the dominant discourses and voices of anorexia, and possibly other related discourses. ⁷

Their stories usually consist of what are termed “thin descriptions” (Geertz, 1973), which tend to state “facts” without too much interpreted meaning or significance. It is in this context that the process of “labelling” becomes prominent in self-construction of “identities,” and of constructions of them attributed to psycho-medical-health staff. Labels disempower as they tend to be reflective of negative ways of being, of disorders, of failures, of weaknesses and inadequacies. In addition, such descriptions tend to totalise the subjectivities of persons in relationship with anorexia—they leave little or no space for alternative descriptions of more complex and contradictory ways of being, or for the personal interpretation of their lives.

Such initial understandings are facilitated by adopting what is referred to as “deconstructive listening” (Freedman & Combs, 1996), which is required for accepting and understanding clients’ narratives without reifying or intensifying the powerlessness, pain, and pathological aspects of their stories (Monk et al., 1997). Deconstructive questioning is also used, but in a more purposeful and interventive nature. It invites clients to see their stories from different perspectives, to draw attention to how those stories are constructed (that is, exactly what they are—stories constructed by the power of the dominant discourses), to note their limits, and to discover that there are alternative narratives (Freedman & Combs, 1996). This process of “deconstruction” is a function of the first assumption that forms the basis of social constructionism (Gergen, 2009a)—“a critical stance is taken toward taken-for-granted knowledge” (Burr, 2003, p. 2). In other words, deconstruction reveals how these persons experience the effects of the disciplinary power of the dominant socially constructed discourses of anorexia. It also illustrates how they are subject to this power through the normalising “truths” of the dominant discourses that shape their ways of being and their relationships. Thus, deconstruction is used to interrogate or unpack the taken-for-granted assumptions that underlie the meanings of the experiences of their relationship with anorexia. It provides a process for understanding how the power of such dominant social discourses galvanises them into ways of accepting the social norms that underlie anorexic discourses.

A deconstruction process is used with other practices such as externalising conversations (McLeod, 2004; White, 2007; White & Epston, 1990), where the essential idea is to separate problems from the client, so that they do not accept the problems as core, embedded personal characteristics. Externalising conversations assist in separating clients from their relationships with anorexia that are constructed by dominant discourses, and which are causing them concern and distress. The use and process of externalising conversations is more important than trying to define the externalised problem(s). Externalising conversations can locate and separate many factors in a space “outside” clients; factors such as feelings, cultural and social practices, and metaphors. Through the process of externalisation, problems of relationships with anorexia can be named, spoken about, and personified as independent entities with a multiple array of descriptions that may change in nature during therapy. Anorexia tends to have an unbounded sphere of influence over the lives of those whom it affects. By being isolated and named in personally chosen words, and by having its characteristics described, anorexia becomes an externalised and bounded entity that has only a limited influence over lives, leaving clients with spaces for exercising alternative ways of being.

Being recruited into the externalising conversations by narrative therapists encourages clients to adopt a curious and questioning position to their problems—they become “investigative reporters” (White, 2011, p. 88) into their circumstances. In this role the therapist can co-operate with them in the externalising process, as it provides opportunities for them to explore the history of their relationship with anorexia. Then, with a more detached position in regard to their relationship with anorexia (and related dominant discourses), clients can provide details of the nature, effects, and consequences of dominant anorexic discourses, and the ways in which voices of discourses of anorexia influence their lives and relationships. Positioning clients as “investigative reporters” into their relationship with anorexia can also provide opportunities for them to decide on what steps to take to resist the influences of anorexia, to undermine the impact of anorexia, to distance their relationship even further from it, or even terminate their relationship with anorexia (White, 2011).

It must also be borne in mind that the disciplinary powers of anorexia can contribute to the formation of intense relationships with clients. The influential disciplinary powers exerted by anorexia tend to result in totalising, aggressive, and confrontational metaphors ⁸ from them. This poses concerns for the process of naming the problems and exploring the effects of the problems of anorexia on their lives and their relationships (White, 2011). One issue of concern is that discussions on the effects of anorexia, couched in metaphors reflective of intense forms of relationships with them, can foster even more intense interest in these forms of relationship during discussions in therapy (White, 2011). Another is that the nature of warring, battling, confrontational, and totalising metaphors tends to offer binary outcomes; that is, there is a winner and a loser. This offering of binary outcomes can present vulnerable risks if clients do not perceive themselves to be winners in the confrontations (White, 2011). A victim or loser identity can easily offer a concurrent construct to them of being a failure, incompetent, “less than,” abnormal, worthless, different, or of a sense of futility to life. Equally concerning is the possible opposite construct of being a hero in the struggle or battle—of attempting to vanquish the persistent and powerful attacks of anorexia on their own; of being the epitome of a hero, possessed of individual and independent attributes and resources to overcome the onslaughts of anorexia. When clients present their relationships with anorexia with such metaphors, it is possible to respect and honour their descriptions without encouragement or elaboration that may promote further embellishment on the metaphors (White, 2011).

There are likely to be expressions of resistance to the negative, totalisation of the power of anorexia when clients may allude to, or mention aspects of their lives within their anorexic-dominated experiences that they are able to appreciate and which are reflective of their more preferred ways of being and living. These are referred to as “unique outcomes” (White, 2007, p. 232; see also Goffman, 1961). Unique outcomes are behaviours, attitudes, emotions, and so on, that would not be expected in the context of the nature of problem-saturated stories. They constitute opportunities that, through questions and reflective discussions, can be developed into their new and preferred stories and subjectivities (White, 2011). Identifying unique outcomes and the possibility of construction of preferred stories go hand-in-hand during deconstruction activities.

The process of identifying unique outcomes, and then using externalising conversations can help create a “space” that makes it easier to encourage clients to talk about the way they would prefer to be, and what they would like to be happening in their lives. This is reinforced through the use of “relative influencing questions” that encourage clients to take positions in relation to their problem that are less damaging to their mental health and relationships (White, 2011). The idea is to assist them to identify their knowledge of their more preferred ways of being, to embrace that knowledge, and to draw on it to reconstruct their subjectivities in the process of re-storying (re-authoring) their lives. This process invites clients to re-author their experiences in ways that create richer and “thicker descriptions” (Geertz, 1973) of their identities, and enables them to re-connect (re-member) with people in preferred forms of relationships. This breaking away from thin conclusions and re-authoring new and preferred stories for their lives and their relationships is described by Freedman and Combs (1996):

Narrative therapists are interested in working with people to bring forth and thicken stories that do not support or sustain problems. As people begin to inhabit and live out the alternative stories, the results are beyond solving problems. Within the new stories, people live out new self-images, new possibilities for relationships and new futures. (p. 16)

In these circumstances a form of social constructionist therapy, namely narrative therapy, could assist in deconstructing the socially constructed dominant discourses of anorexia, and in reconstructing preferred subjectivities of persons in relationship with anorexia. This approach is particularly appropriate to working with persons struggling with the problems and challenges of anorexia (and related discourses), as narrative therapy is embedded in ideas and themes of social constructionism, the very paradigm in and from which notions of anorexia are constituted and maintained. The process of reconstruction that identifies and draws on clients’ knowledge about themselves, allows them to explore, with the therapist and other members of a collaborative therapeutic arrangement, their intentions or purposes that shape and guide their preferred ways of constituting their subjectivities, and their ways of living. It emphasises personal values, beliefs, hopes, and dreams that support their preferred behaviours, and highlights their principles of living and what it is that they stand for in life (Carey & Russell, 2003; White, 2001).