Self-illness ambiguity and anorexia nervosa

ABSTRACT

Self-illness ambiguity is a difficulty to distinguish the ‘self’ or ‘who one is’ from one's mental disorder or diagnosis. Although self-illness ambiguity in a psychiatric context is often deemed to be a negative phenomenon, it may occasionally have a positive role too. This paper investigates whether and in what sense self-illness ambiguity could have a positive role in the process of recovery and self-development in some psychiatric contexts by focusing on a specific case of mental disorder – anorexia nervosa.

KEYWORDS:

• Self-illness ambiguity
• anorexia nervosa
• self
• habits

1. Introduction

Living with a mental illness comes with many challenges that may be difficult to grasp for those who do not suffer from it. Among these challenges is one where people suffering from mental illness experience various forms of ambivalence and ambiguity concerning their illness, diagnosis and the self. ‘Self-illness ambiguity’ is a term coined to describe the difficulty to distinguish the ‘self’ or ‘who one is’ from a mental disorder or diagnosis (Sadler 2007). While various forms of ambivalence and self-ambiguity are part of our lives, self-illness ambiguity has been argued to be a common phenomenon in the psychiatric context and has become a topic of increased interest among philosophers and psychologists (Sadler 2007; Dings and de Bruin 2016; Dings and Glas 2020; Dings and de Bruin 2022). According to Dings and Glas (2020) self-illness ambiguity in a psychiatric context can take many forms and arises at two levels: of unreflective self-experience (e.g. where one experiences conflicting thoughts, emotions and desires) and reflective self-understanding (e.g. where one tries to make sense of the illness and its relation to one's self).

It is easy to see that self-illness ambiguity in a psychiatric context can be a problem and a source of distress for those who experience it. For example, it may be both difficult and painful to assess whether one's emotion or behaviour in a particular context is what they would feel or do independently of their illness, or an expression or result of their illness, or of medication they take/treatment they follow. Reducing and/or resolving self-illness ambiguity may thus be an important personal and therapeutic goal (Dings and Glas 2020). In their paper, Ding and Glas discuss various obstacles that may arise for resolving self-illness ambiguity and argue that self-management in the context of psychiatry requires not only symptom and emotional management, but also reducing self-illness ambiguity. Given that self-illness ambiguity will often result in various forms of psychological distress, it may thus require personal and professional (therapeutic) attention, as well as reducing and resolving. However, this is prima facie compatible with the possibility that some forms of self-illness ambiguity may have a positive role and need not require resolving, as is acknowledged in the current frameworks for self-illness ambiguity (e.g. Dings and Glas 2020; Glas 2019; Dings and de Bruin 2022; McConnell and Golova 2022).

The main goal of this paper is to contribute to this line of thinking about self-illness ambiguity and explore whether and in what sense self-illness ambiguity could have a positive role in the process of recovery and self-development in some selected cases. Questions about the nature and significance of self-illness ambiguity arise as soon as we recognize how varied and complicated the phenomena of ambivalence and self-ambiguity are in a psychiatric context. This paper explores what we can learn about self-illness ambiguity in a psychiatric context by focusing on a specific case of mental disorder – anorexia nervosa (AN). This choice is motivated both by the nature and prevalence of this disorder (Hudson et al. 2007; Arcelus et al. 2011). There is growing research on the nature of key symptoms involved in AN that can shed new light on questions concerning self-illness ambiguity. The issues of authenticity, self, and identity struggles are well documented in the context of anorexia nervosa (e.g. Hope et al. 2011; Glover 2014). To make the task manageable, in this paper I will only focus on two strands of research on AN: (a) recent (neuro)psychological research on the habitual and compulsive nature of some of the core symptoms in AN and (b) phenomenological and psychological research based on self-reported experience of people suffering from AN.¹ The second strand of research complements the first one by including patients’ reports and perspective into theorizing about anorexia nervosa.

Drawing on these two strands of evidence, I will make three tentative observations. First, the habitual and possibly compulsive nature of key behaviours involved in the symptomatology of anorexia nervosa suggests a complicated, stage-like picture of self-illness ambiguity at the level of pre- or unreflective experience (Section 3). Second, drawing on evidence from first-person reports (Hope et al. 2011) and phenomenological research on anorexia, I will propose that given the habitual nature of some of the core symptoms in anorexia nervosa and the fact that they often become compatible with one's self and values (in other words, can be ego-syntonic), self-illness ambiguity, although possibly painful and confusing, may also be a resource in the course of overcoming anorexia nervosa. Finally, I will briefly consider whether self-illness ambiguity needs to be resolved at the reflective level of self-understanding by conceptualizing anorexia as an inauthentic part of the self and how this might relate to the idea that persons suffering from anorexia nervosa are to develop a coherent self-narrative (Section 4).

2. Self-illness ambiguity and the need for resolution

We all occasionally have doubts about whether and to what extent our emotions, beliefs, desires and actions are true to ourselves or reflect who we are. We can sometimes be in two minds about whether, for example, an action is reflective of who we are or not. We may be conflicted about whether and which lifestyle to choose and differently respond to that conflict (Gunnarsson 2014). As a result, we may experience various feelings of ambivalence or self-ambiguity. For many of us, such feelings can be a source of significant unease (see, e.g. Frankfurt 1988; Coates 2017) and may thus call for some resolution.²

Unsurprisingly, feelings of self-ambiguity often arise in the psychiatric context where it might be particularly difficult to assess whether one's emotion or behaviour is what one would feel or do independently of their illness, or an expression or result of their illness, medication they take, or treatment they follow. Self-illness ambiguity can be defined as a difficulty to distinguish the ‘self’ or ‘who one is’³ from a mental disorder or diagnosis in various contexts (Sadler 2007). Receiving a diagnosis is in itself an important factor that may affect one's self-conception and lead to significant tension. But the sources and forms of self-illness ambiguity in a psychiatric context are many (Dings and de Bruin 2016; Dings and Glas 2020). As in other cases of self-ambiguity, self-illness ambiguity may be a source of significant distress.⁴ Patients may struggle with feelings of ambivalence (Karp 1994), confusion, or self-doubt (Inder et al. 2008).

In a recent paper, Dings and Glas (2020) provide a useful two-level framework to characterize different forms of self-illness ambiguity that can arise in a psychiatric context. First, self-illness ambiguity can arise at the level of unreflective (or pre-reflective) experience. At this phenomenological level, ‘the source of intentional action might be experienced as neither internal nor external, but ambiguous’ (Dings and Glas 2020, 344). Persons suffering from mental illness might not recognize the symptoms of their illness, but nevertheless experience feelings of confusion and self-doubt, of being lost, and of not feeling like themselves, resulting from which might be unclarity concerning whether they are in control of their actions, thoughts and emotions (336–337). Second, self-illness ambiguity may arise at the level of reflective self-understanding:

[S]elf-ambiguity may also pertain to the conceptual level, where agents reflect on who they are, and form a self-conception or self-narrative, but are confronted with conceptual difficulties such as ‘who am I’, ‘what is a mental disorder’ and how do mental disorders relate to who I am. (Dings and Glas 2020, 344)

In this case, a person suffering from an illness may struggle to make sense of how the illness itself relates to their self, and whether and in what sense it might be a part of their identity. For example, to what extent is a seasonal depression suffered every winter related to who one is and their identity? Importantly, on this account, there is a strong relation between self-illness ambiguity at the pre- or unreflective level and self-illness ambiguity at the reflective level, such that pre-reflective experiences of confusion about one's actions, feelings and thoughts lead to questions, asked at the conceptual level, about the relation between self and illness (Dings and Glas 2020, 336).⁵

Dings and Glas (2020) note that the self-illness relation in a psychiatric context is best construed as a diachronic process where the relation between self or who one is and one's illness is negotiated not only at different levels but also throughout time in a series of stages. Social factors may be a crucial part of that process. How a certain diagnosis is perceived by society and by other people with that diagnosis may influence how a person will relate to it. Close relations and social networks may play a vital role in the process of self-understanding by either helping or hindering one's attempts to understand the relation between one's illness and the self (Shea 2010; Dings and Glas 2020; de Haan 2020).

Self-illness ambiguity is a complicated phenomenon. According to Dings and Glas (2020), this is so, at least partly, due to how mental illness is conceptualized in current psychiatric practice. One factor is the so-called reification of mental disorders as sets of symptoms, experiences and behavioural patterns defined by diagnostic manuals. When so construed, mental illness can easily be seen as an entity external to who one is.⁶ Another problem concerns feelings of decreased trust and reliability of one's experiences in a psychiatric context. This may, according to Dings & Glas, lead to increased reliance on expert authority and diagnostic manuals, but it need not always lead to better self-understanding (see Tekin 2019). A different source of complexity comes from our limited understanding of what the ‘self’ is and how best to account for it in both descriptive and normative terms. Finally, ambiguity as such may take different forms in one's experience and come in different shades and strengths.

Given that self-illness will often involve difficult, confusing and possibly painful feelings that lead to distress, Dings and Glas propose that self-management in the context of psychiatry requires not only symptom and emotional management, but also reducing or perhaps even resolving self-illness ambiguity. This can be done, according to them, by achieving congruence between one's reflective self-narrative and one's unreflective bodily and affective experiences. The interplay between these two levels is in their opinion vital for this task. The narrative perspective could thus offer a way out of experienced self-illness ambiguity, because it can provide means for increasing such congruence. So understood, reducing or resolving self-illness ambiguity in the course of self-management and self-understanding may thus pertain to important personal and therapeutic goals. There is no doubt that various forms of self-illness ambiguity may require such attention because they may result in various forms of distress. Dings and Glas’ proposal to include reduction of self-illness ambiguity as part of self-management in mental illness is a convincing suggestion that requires further reflection on the tools that might be relevant for implementing it in different therapeutic contexts.⁷

The need for reduction and/or resolution of self-illness ambiguity in some cases is prima facie compatible with the possibility that some forms of self-illness ambiguity may have a positive role and need not require resolving. This is acknowledged in the current frameworks for self-illness ambiguity: It has been observed that there are many ways a person might relate to their self-illness ambiguity, not all of which will be a call for action (Dings and Glas 2020). According to Glas (2019), in some cases, the experience of self-ambiguity may open up new possibilities and be enriching for a person. According to Dings and de Bruin (2022), in some cases, one's identification with a mental illness may make recovery feel as a threat to one's identity (e.g. see Kokanovic, Bendelow, and Phillip 2013 for discussion of such cases in the context of depression). Integrating one's illness into one's narrative about oneself is an option but ambiguity may, nevertheless, linger. In a recent paper, McConnell and Golova (2022) argue that a third form of self-illness ambiguity arises when a narrative is formed and involves uncertainty about whether one's self-narrative represents who one really is, often resulting from a mismatch between self-narrative and evaluative stance. This type of ambiguity is, in their opinion, a useful phenomenon in therapy for addiction and should be first induced and embraced in the course of therapy, before it can be resolved.

The aim of this paper is to contribute to the exploration of whether and to what extent self-illness ambiguity may play a positive role in the process of recovery and self-development. Arguably, this can help us better understand self-illness ambiguity in psychiatric contexts, as well as its possible practical and ethical implications. In what follows, I will try to make progress on this task by focusing on self-illness ambiguity in the case of anorexia nervosa. Given the purely theoretical nature of this inquiry, and limited space, the goal is by no means to present an exhaustive account of self-illness ambiguity in anorexia nervosa or draw definitive conclusions about its role. Rather, it is to draw attention to some of the issues that are pertinent to this mental disorder and may shed light on self-illness ambiguity more broadly.

3. Anorexia nervosa: habits and the self

Anorexia nervosa (AN) is a mental disorder typically characterized by persistent food restriction, specifically low body weight, as well as a lack of recognition of the seriousness of the illness. It affects approximately 1% of women around the world (Hudson et al. 2007) and has one of the highest mortality rates among psychiatric disorders (Arcelus et al. 2011; Keshaviah et al. 2014). Currently used diagnostic criteria for anorexia nervosa from DSM-5 (APA 2013) have broadened the previously used ones by removing a specific low-weight guideline (previously less than 85th percentile), changing weight phobia to be explicitly or implicitly present (instead of requiring explicit endorsement), and eliminating the amenorrhea criterion (absence of at least three consecutive non-synthetically induced menstrual cycles). Moreover, an ‘atypical AN’ diagnosis was created for persons who do not have low weight, but otherwise meet DSM-5 criteria for AN. Two subtypes of anorexia nervosa are commonly identified in the literature: the restrictive subtype where people place severe restrictions on the amount and type of food they consume, and may restrict consumption of certain types of foods (e.g. carbohydrates, fats), as well as engage in calorie counting, skipping meals, and/or obsessive rules or rigid thinking about their food habits. The restrictive subtype may but need not be accompanied by excessive exercise. The other subtype of anorexia nervosa also involves binge eating and purging behaviour aimed at facilitating weight loss (APA 2013).

A number of risk factors have been identified in the empirical literature, such as: being female, being an adolescent, having an obsessional style, but their exact role in the occurrence of anorexia is limited and still far from understood (Jacobi et al. 2004). The research on the neurobiological, psychological and possibly cultural mechanisms that may underlie the etiology and development of AN is growing and it would be impossible to do it justice in a paper like this. I will focus here on two recent strands of research in psychology and philosophy that, in my view, may be helpful to discuss self-illness ambiguity in anorexia nervosa. I will use the two-level framework proposed by Dings and Glas (2020) to structure the presentation of this material in the remainder of the paper.

Recent (neuro)psychological research on the habitual and compulsive nature of some of the core symptoms in anorexia nervosa can provide interesting material when considering self-illness ambiguity at the level of un- or pre-reflective experience in that disorder. Part of the general description of AN symptomatology is that persons suffering from anorexia nervosa exhibit a continuing drive for thinness that results in continuing restrictions on weight goals (Barbarich-Marsteller, Foltin, and Timothy Walsh 2011). In recent years, these characteristic behavioural patterns of extreme dietary restrictions (and occasionally over-exercise) have been argued to provide evidence for the compulsive nature of anorexia nervosa (Godier and Park 2014, 2015). Robbins et al. (2012) define compulsivity in the context of psychiatric disorders as a trait in which actions are repeated in a persistent manner despite adverse consequences. A model example is the drug-seeking behaviour in substance abuse that persists despite adverse effects. Researchers working on the compulsive nature of behavioural patterns observed in AN draw some parallels between those behavioural patterns and addiction (e.g. Kaye et al. 2013; Godier and Park 2014, 2015). An interesting parallel concerns the usual dynamics of the two. The initial period in both cases can be described as a phase of reward seeking, where weight loss in AN is experienced as rewarding or pleasurable (Scheurink et al. 2010; Park, Dunn, and Barnard 2011, 2012). Next, it has been reported that despite adverse consequences for general health and mental well-being, it becomes increasingly difficult to refrain from weight loss behaviours (Park, Dunn, and Barnard 2011, 2012). Both substance addiction and anorexia nervosa have been reported to correlate with decreased cognitive flexibility (e.g. Hildebrandt et al. 2015) resulting in a rigid cognitive style that has been argued to contribute to compulsivity (Fineberg et al. 2010). A qualitative study from Godier and Park (2015) provides some evidence that seems to corroborate the compulsive nature of behaviours involved in AN. In particular, the subjects in the study reported that they experience the compulsive nature of weight loss behaviour as crucial for the continuing persistent form of their disorder and a significant barrier for recovery.

In line with observations about the compulsive nature of some of the key symptoms and behaviours in anorexia nervosa, there has been increased research on the neurobiological and neurocognitive models that could explain the persistent maladaptive eating behaviour in anorexia nervosa as a learned habit. At the neurobiological level, changes in the cortical-striatal circuits that are associated in the literature with compulsivity (Robbins 2007; Fineberg et al. 2010) and addiction (Everitt and Robbins 2013), have been reported in anorexia nervosa (Rothemund et al. 2011; Godier and Park 2014). According to Uniacke et al. (2018) the development and maintenance of such behaviours are consistent with current knowledge about habitual learning from cognitive neuroscience. The dieting behaviour is initially experienced as rewarding, although the reasons for why that might be the case may be specific for different persons. Experienced rewards involved in shaping this behaviour may be quite different, including changes in their physical appearance, adherence to what they think is a ‘healthy’ diet, a relief in anxiety, an increased sense of self-control. An initially goal-directed behaviour (e.g. to change appearance, release anxiety) followed by reward becomes gradually less dependent on the outcome and more resistant to change (Sysko et al. 2005; Mayer et al. 2012). According to Uniacke et al. (2018), food restriction observed in more advanced stages of the AN is less linked to original reward and more dependent on variety of learned external and internal cues, which may suggest that fronto-dorsal-striatal circuits may underlie the compulsive nature of persistent habitual food restriction behaviours in AN. This suggestion is compatible with several recent proposals that see such habit loops as dependent on experienced anxiety and link anorexia nervosa with the anxiolytic (i.e. anxiety reducing) nature of starvation (O’Hara, Campbell, and Schmidt 2015; Lloyd et al. 2019).

Finally, there is a growing evidence and increased interest in the nature and role of other plausibly habitual and compulsive behaviours that sometimes accompany the key symptoms of anorexia nervosa, such as increased urge for movement, physical and mental restlessness (Holtkamp, Hebebrand, and Herpertz-Dahlmann 2004; Casper et al. 2020) and restrictive over-exercise (Keyes et al. 2015; Riva et al. 2021). Increased evidence pointing to compulsive and habitual nature of weight loss behaviours involved in anorexia nervosa might perhaps explain at least to some degree why self-starvation involved in anorexia nervosa is such a persistent phenomenon (Walsh 2013).

The observations and evidence of the habitual and compulsive nature of some of the central behaviours in the AN symptomatology provide interesting material for discussion of self-illness ambiguity, in particular concerning the level of un- and pre-reflective experience. Do persons suffering from anorexia nervosa experience self-illness ambiguity at the level of un- and pre-reflective experience? This is a complicated matter and my discussion here is only speculative, given the data points presented in this section, and focuses only on charting some possible interpretations. One reason to expect that self-illness ambiguity at the pre-reflective level does not occur early on in anorexia nervosa is this: At least initially the behaviour of decreased food intake or dieting in anorexia nervosa is (typically) consciously chosen by a person. The direct consequence of the behaviour of decreasing food intake may take many forms (improved self-image, decrease in anxiety, feelings of being in control) and is typically experienced, at least initially, as a rewarding outcome of that behaviour. At least initially, the core behaviours that underlie the symptomatology of anorexia nervosa are consistent with decisions and values of the people who suffer from it (in other terms, they are ego-syntonic). McConnell and Golova (2022) make a parallel observation in the context of addiction.

The matter is, however, more complicated than that. One reason to expect that persons suffering from anorexia nervosa experience self-illness ambiguity of the pre-reflective type at some point in the course of their illness is this: Although the low food intake behaviour in anorexia nervosa is initially experienced as one's own choice that leads to some attractive and rewarding outcomes, the behaviour gradually becomes habitual or even compulsive and starts to influence the life of a person in a much broader sense. Given a variety of possible adverse consequences (e.g. loss of energy, health issues, adverse psychological effects or stigma), the behaviour may occasionally be in tension with other experiences, behaviours and goals of a person – self-illness ambiguity at a pre-reflective level may arise. For example, a decision not to consume a high-calorie product may be a result of the dieting behaviour or an expression of one's overall preference.

A further complication suggested by the evidence of the habitual and compulsive nature of anorexia nervosa is this: As many habits, the low foot intake behaviour may become ingrained and internalized by the person. The maintenance of this kind of habit loops may affect the presence and role of self-illness ambiguity in AN. We are not our habits, but many habits can shape us without us being ambivalent about their role in our lives. One hypothesis is then that the habitual nature of key symptoms in anorexia nervosa may constitute a barrier for experiences of self-illness ambiguity at the pre-reflective level, which would have otherwise come to surface earlier rather than later. Another hypothesis to consider here is that when self-illness ambiguity arises at the level of pre-reflective experience, at least initially, it is not properly attended to, because of the adopted explicit goals of maintaining low body weight and controlling the food intake; or is attended to, albeit by means of maintaining the habits and control (more on this in Section 4).

Finally, the process of treatment may affect whether and to what extent self-illness ambiguity will arise at the level of pre-reflective experience. Persons suffering from anorexia nervosa and undergoing treatment for it will be challenged in ways that may induce pre-reflective experiences of self-illness ambiguity. Suppose a person decides that they do not want to eat a particular type or amount of food. They may then struggle to demarcate, phenomenologically, if it is really them that does not want to eat or whether it is a result of them suffering from anorexia nervosa. It seems to be part of everyday life that one sometimes does not want to eat something, however, for someone struggling with AN when trying to deal with the symptoms of AN in order to recover this may turn out to be a phenomenologically ambiguous matter.⁸

It is important to note here that habitual behaviours that underlie key symptoms of anorexia nervosa typically occur in social environments. It is likely that self-illness ambiguity will come to the surface in such social contexts, where family, friends or colleagues who are not immersed in those habits will point them out and possibly evaluate them.⁹ Indeed, this is why people suffering from anorexia nervosa will often shy away from social situations that require eating. Experiences of self-illness ambiguity are socially situated and can be induced by external awareness and evaluation of one's symptoms and illness.

The above observations and tentative interpretations are at least partly compatible with the currently used criteria for anorexia nervosa in DSM-5, where the persistent lack of recognition of the seriousness of one's low body weight is part of the third criterion (criterion C) for receiving the diagnosis. The dynamics of how anorexia nervosa develops via habit loops, as suggested by the evidence presented in this section, makes it structurally more likely that the core symptoms of AN will be initially experienced as ego-syntonic and as part of one's self. This may, in turn, affect whether and to what extent self-illness ambiguity will arise at the level of un- or pre-reflective experience in anorexia nervosa. As many other entrenched habits, the habits of starvation in anorexia nervosa may feel, at least at some point and to some extent, like home. This is compatible, as explained above, with the possibility that self-illness ambiguity at the pre-reflective level will either eventually arise or be properly attended to, e.g. because of experienced negative outcomes of starvation or as a result of treatment.

To conclude, the evidence pointing to the habitual and compulsive nature of behaviours aimed at restrictive food intake and weight loss, suggests a complicated, stage-like picture of how experiences of self-illness ambiguity may arise at the pre-reflective level in anorexia nervosa, making the onset of the illness feeling initially perhaps more at home than it might be in cases of other mental disorders. For example, depression has been argued to involve an experience of loss of possibilities (Ratcliffe 2014), a state that is arguably foreign to who one otherwise is and their plans. This stage-like picture is compatible with the current frameworks of self-illness ambiguity that emphasize the diachronic nature of the phenomenon (Dings and Glas 2020).

4. Self-illness ambiguity as a resource and anorexia nervosa

Initially rewarding habits of restricted food intake can and often are endorsed by persons suffering from AN. The characteristic persistent lack of recognition of the seriousness of low body weight observed among persons suffering from anorexia seem to corroborate this point. Nevertheless, anorexia nervosa involves some internal conflicts and tensions that manifest itself in its behavioural and psychological symptoms, which will eventually rise to the surface of one's experience and may take the form of self-illness ambiguity at both the pre-reflective and the reflective level. In this section, I present evidence from recent philosophical and phenomenological research on anorexia nervosa that can help us understand such conflicts concerning self and anorexia nervosa and their manifestation at the level of both pre-reflective experience and the reflective level of self-understanding. Drawing on phenomenological research on anorexia nervosa and a qualitative study from Hope et al. (2011) that provides important material concerning issues of self and authenticity, I will suggest that self-illness ambiguity at both levels may in some cases be an important resource for recovering from anorexia nervosa, without necessarily deeming anorexia as an inauthentic part of one's identity. I will focus here on patients’ reports and the phenomenological approach, which provide an additional source of evidence for theorizing about anorexia nervosa and, I believe, should be also considered when investigating self-illness ambiguity in AN. This approach also responds to recent pleas for greater inclusion of patients’ perspective and phenomenology into psychiatric investigation and practice (e.g. Andreasen 2006; Tekin 2020; Larssen et al. 2022).

Let us first consider the nature of tensions experienced in anorexia nervosa. Recent phenomenological research on anorexia nervosa questions the overly cognitive description of AN, as present in the current diagnostic criteria such as DSM-5 (APA 2013). While diagnostic criteria for AN have always placed emphasis on the dysfunctional beliefs and faulty cognitions about one's own body (e.g. criterion C: Disturbed by one's body weight or shape, self-worth influenced by body weight or shape or persistent lack of recognition of the seriousness of low bodyweight), it has been argued that they overlook important changes and disturbances in bodily awareness (Fuchs 2021) and bodily experience (Osler 2021) that are arguably characteristic for anorexia nervosa. According to Fuchs (2021, 1), anorexia nervosa is better understood as ‘a fundamental disturbance of embodied self-experience’ rather than disturbed body image. Drawing on psychological research on anorexia, phenomenological work concerning embodiment and reports of patients’ experience, Fuchs argues that persistent self-starvation in anorexia nervosa can be explained as a manifestation of a particular type of conflict: a fundamental alienation of the self from the body. In his view, this conflict has a temporal dynamic that involves: first adopting an external look at the body, then the resulting reification of one's body, and, finally, gradual alienation of the body from the self. The body may be experienced as an obstacle, as impure, and in need of rigid control. Drawing on patients’ reports and the phenomenological tradition, Osler (2021) describes bodily experience in anorexia nervosa as the following sequence of stages. The early stages involve experiencing one's body as unruly, threatening and in need of control. The body is experienced and treated as an object. At least initially, persistent dieting behaviour provides successful and rewarding means of control resulting in experiences of empowerment and strength. Olser argues that, as self-starvation in anorexia nervosa progresses, the body and the focus on controlling it take over the control of the individual's life, possibly hindering other choices and courses of action.

Despite some interesting key differences, in both of these accounts (Fuchs 2021; Osler 2021; see also Leder 2013, 2021), the authors apply an embodiment approach to understand anorexia nervosa in order to explain the dynamics of the disorder and how it can be experienced and conceptualized by patients at different stages. Both accounts rely on first-person accounts of persons suffering from anorexia nervosa. Both of them draw on an important phenomenological distinction (Husserl 1989): between the body that is experienced and lived by the person, the body we have, i.e. body-as-subject, and the physical body in the world, i.e. body-as-object. In both accounts, a disturbance in embodiment observed in anorexia nervosa is explained as a result of losing one's typical connection between oneself and one's body, thus shifting from experiencing body-as-subject to body-as-object – hence the reification of the body. In both studies, an emphasis is put on how this experience of one's reified body can be internalized and accepted as one's own and, as a result, how anorexia can take control over other aspects of one's life.

These observations are prima facie compatible with the evidence concerning the dynamics of anorexia nervosa and the habitual and possibly compulsive nature of the low food intake behaviour that I summarized in the previous section. The dynamics and initial endorsement of anorexia nervosa suggest that these habits can have a significant level of control over one's life. In light of these observations and the phenomenological research on experiences and tensions in anorexia nervosa, I will now suggest that both experiencing and acknowledging the tension between the control of the body in AN and the self can result in an important shift in the recovery process from AN. Arguably, this is where self-illness ambiguity (at both levels) may come into the picture. There are several reasons for why self-illness ambiguity may play an important and transformative role in the course of anorexia nervosa. First, the presence of self-illness ambiguity at the pre-reflective level may be a sign that a person can experience the tension between the control imposed by the persistent low food intake behaviour and the experienced negative outcomes of that control over their life. This can take various forms and concern experiences of hunger, specific food craving and desires, tiredness. Arguably, when such tensions surface to the level of one's experience, they may be instrumental in the process of gaining insight into the negative consequences and costs of the habitual self-starvation in anorexia. The pre-reflective experience of self-illness ambiguity may thus constitute a more nuanced alternative to how the habits of anorexia nervosa are initially experienced. Following the framework suggested by Dings and Glas (2020), experiencing self-illness ambiguity at the level of one's thoughs, emotions and intentional action in anorexia nervosa would be also a prerequisite for self-illness to occur at the reflective level. This will be discussed now.

Arguably, self-illness ambiguity at the reflective level may also be a welcome phenomenon that can provide some personal and therapeutic resources in the process of recovery from anorexia nervosa. In order to motivate this idea, I will now present and discuss some evidence of problems concerning self and authenticity in AN based on a qualitative study from Hope and colleagues (2011). Hope and colleagues (2011) provide extensive material in support of the claim that at least at some later stages of anorexia nervosa, persons suffering from AN struggle with questions concerning authenticity, self and the role of anorexia nervosa in their life. The timeframe is important here, given that evidence presented in the study concerns patients’ experiences and perspectives later in the course of suffering from AN. They conducted semi-structured interviews with 29 female participants suffering from anorexia nervosa who either had been or were in treatment at the time of interviews. The goal of the interview was to ask patients about their feelings and thoughts about several aspects of anorexia nervosa, including understanding of AN, attitudes towards compulsory treatment within NHS UK, and the role of family, attitudes about weight and body shape, as well as views on the impact of anorexia nervosa on decision making (Hope et al. 2011, 20). Although the interviewers did not mention the notion of authenticity or raised explicit questions about identity, most participants themselves raised these problems and thus provided extensive material concerning them. Hope et al. (2011, 22) describe how burdensome living with anorexia nervosa is, as revealed by reflections provided by participants and point to a particular kind of conflict that participants admitted to the treatment programme become aware of: between extreme anxiety around eating and maintaining weight, i.e. maintaining the behaviour of low food intake, and that behaviour eventually leading to death (22).

The authors list four main issues or domains concerning self, authenticity and anorexia nervosa that were reported and discussed by participants in the study. (1) Many participants used the notion of ‘authentic self’ and described the relationship between self and anorexia nervosa as problematic. This relationship was differently described and problematized, as for example: anorexia being separate from the real self, with separation being differently experienced, or as anorexia being one of the two parts of the self (Hope et al. 2011, 22). In the second case, the participants often described the two parts of the self as being in conflict. Some felt that anorexia nervosa was in conflict with the authentic part, whereas others did not feel the same and expressed the views which suggested that they have internalized AN as part of their authentic self (23). (2) Several participants reported the impact of anorexia nervosa on their personal abilities and skills, in both negative and positive terms. The authors observed that there seemed to be a systematic difference between reported behaviours, judgements and values that correlated with being at a low or higher weight (24). (3) Some participants reflected on experiencing ambivalence and conflict between desires pertaining to different parts of the self, one of which is anorexia nervosa (24). A group of participants reported what may perhaps be interpreted as self-illness ambiguity at the unreflective level when describing their food preferences (25). (4) Finally, in contrast to reports and issues presented in points (1)–(3) above, a number of participants in the study described anorexia nervosa as an integral part of themselves, or of who they are. This was despite the fact that, as a result of receiving treatment, they were not anymore significantly underweight. Although anorexia was not always presented as welcome or desired at that stage, it was nevertheless described as part of their identity (25).

The authors summarize the main findings gathered from these interviews as follows. They see the participants’ preoccupation with the idea of authenticity and self as an indicator that the idea or notion of authenticity can have a valuable role in the process of self-understanding and self-development at the later stage of anorexia nervosa and during treatment (Hope et al. 2011, 26). They identify three main ways (albeit in a different order) in which the issues of authenticity can be conceptualized at the level of self-understanding by persons suffering from anorexia nervosa (26):

1. anorexia can be conceptualized as an authentic part of the self or part of one's identity and explicitly endorsed which may result in lack of motivation to engage with treatment in order not to suppress or remove that authentic part of the self – a position that is common in the early stages of anorexia nervosa (also endorsed in the Pro-Ana movement¹⁰)

2. anorexia may be conceptualized as undesirable and problematic (causing suffering), but nevertheless as constituting an authentic part of one's identity. This may result in the attitude that one should not completely get rid of AN, but learn to live with it and minimize harms it brings.

3. anorexia may be conceptualized as an inauthentic (though powerful) part of one's identity and this may provide grounds and motivation to overcome it.

According to Hope et al. (2011, 27) the inner conflict between anorexia nervosa and the self, and the idea of authenticity in particular can be valuable for self-development by: giving a person space and motivation to develop the self one wants, by helping to develop a level of self-esteem that is crucial for change, by helping to better understand one's motivation in the context of key symptoms of anorexia nervosa and by providing means to explore what makes one happy and satisfied (27). Thus Hope et al. (2011) see the idea of authenticity as an important tool for personal self-development, self-creation and discovery for persons suffering from anorexia nervosa.¹¹ They also make a rather controversial suggestion that the evidence of experienced inner conflicts and struggles with authenticity may in some cases be treated as an indicator that the capacity to refuse treatment in anorexia may be compromised.¹²

Hope et al. (2011) provide extensive and convincing evidence that issues concerning self and authenticity arise in the course of anorexia nervosa, at least at the later stages and, in their opinion, play an important role in the process of recovery. The reported struggles with authenticity described by those who are in the treatment process illustrate that vividly. I will now suggest how self-illness ambiguity may fit that evidence. Arguably, experiences of self-illness ambiguity at the pre-reflective level may be involved in transitioning from the self-reflective attitude described in (i) where anorexia is both experienced and endorsed as an authentic part of the self or part of the identity that controls the behaviour and decreases motivation for recovery (a position that is common in early stages of anorexia) to the self-reflective attitude described in (ii), where anorexia is thought of as undesirable and problematic but nevertheless as an authentic part of one's identity. The upsurge of experienced (or attended to) tensions between the controlling dieting behaviour and its outcomes and the transition from (i) to (ii) may allow for the self-illness ambiguity to arise in a conceptualized form at the level of reflective understanding. The conflict between anorexia and the self can become reflectively conceptualized in the course of that transition. It is of course possible that the transition from (i) to (ii) and the struggles reported in the study constitute an additional source of anxiety and stress. Nevertheless, self-illness ambiguity at both levels may provide a valuable resource for self-development and recovery for those suffering from anorexia nervosa, even though, as I will now suggest, considering anorexia as inauthentic need not be vital for that process to be initiated.

Hope et al. (2011) emphasize the role of the idea of authenticity as grounds for developing the reflective attitude described in (iii), where anorexia is thought about as an inauthentic part and where this may provide grounds and motivation to overcome it. I believe that the important resources can become available to those who suffer from anorexia nervosa already when adopting the reflective attitude described in (ii). There are two reasons for why I want to emphasize the significance of self-illness ambiguity that arises at that stage for the process of recovery and self-understanding. First, starting to experience and/or to attend to conflicts and tensions that result from the habitual behaviour of low-food intake may suggest an increased flexibility in how anorexia is experienced, from something that feels like home to something that causes suffering too. Second, this kind of self-illness ambiguity at the pre-reflective level is necessary for a conceptualized, reflective form of sell-illness ambiguity to arise. The two forms may work here in tandem. First experiencing and then gradually conceptualizing anorexia nervosa as something that may be problematic and undesirable is a starting point for experiencing and conceptualizing anorexia as something one might want to address and change and for becoming motivated to do so. Adopting the reflective attitude described in (ii) is, thus, a vital sign not only of an increasing insight into one's condition, but also of experiencing and acknowledging the downsides of the control imposed in AN. This may be seen as a particularly important step for recovery in anorexia nervosa, where, as mentioned earlier, the lack of recognition of seriousness of the illness is a key symptom and an important obstacle for entering the process of treatment and recovery.

However, it is not clear, I think, whether conceptualizing anorexia nervosa as an inauthentic part of one's identity or self is necessary to provide grounds and motivation to overcome it, or whether it is even desirable to think about anorexia in that manner for that very purpose. A more general possibility seems relevant here, namely that some parts of one's identity or authentic self may be such that one would want to change them, get rid of them and/or might not be proud of them. I have argued in Section 3 that recent research suggests that key symptoms of anorexia nervosa are habitual and compulsive. The ingrained nature of the habits of dieting and self-starvation often makes them an important part of lives and experienced identities of those who suffer from anorexia nervosa, at least at some point in their life. An initially rewarding behaviour transformed into a dangerous and compulsive habit may be resistant to become fully externalized. The embodied nature of disturbances experienced in anorexia nervosa, as suggested by recent phenomenological research, may provide yet another explanation of why disengaging from behaviours that underlie key symptoms of anorexia nervosa is particularly difficult. Supposing that these interpretations are correct, then an objectified body takes control over one's life, in which case reconceptualizing that body as inauthentic may not be a promising way to proceed.

Despite Hope et al.'s (2011) emphasis on conceptualizing anorexia as an inauthentic part of the self, it is still possible that anorexia nervosa may constitute to some extent an important part of one's identity, at least at some point in time. First, as research on the psychology and neurobiology on anorexia nervosa suggests, the low food intake and dieting behaviours become part of one's ingrained habitual bodily and neurobiological functioning, at least at some point and to some degree. Second, as research in the phenomenology of anorexia nervosa suggests, they may be a manifestation of particular struggles and conflicts and in that sense have some functional significance for the individual. Third, anorexia nervosa can become an important part of one's history and life experience. The process of recovery and self-development in anorexia nervosa often requires gaining perspective, developing new habits and even endorsing new values, but it need not and perhaps should not require entirely severing the anorexia part of one's history and identity or reconceptualizing it as inauthentic. To sum up, self-illness ambiguity may provide an important resource not only for noticing the tensions that pertain to one's illness, as suggested above, but also for gradually seeing oneself, one's goals, values and actions in abstraction from the ingrained and often rewarding habits of anorexia nervosa. Thus, in the case of anorexia nervosa, inducing, not reducing self-illness ambiguity, may be an important resource for the process of recovery, at least initially.

How exactly self-illness ambiguity could help with that process and how it could be utilized in the therapeutic work are difficult questions and addressing them in a satisfactory manner goes beyond the scope of this paper. One idea could be that attending to and working with experiences of self-illness ambiguity may be helpful for the assessment of currently endorsed values and goals, as well as the discovery of other, alternative preferences, values and behaviours. Here I draw on a chapter discussing issues of authenticity and identity in eating disorders by Jonathan Glover (2014, 354–359), where he suggests that assessing behaviours and values endorsed as part of anorexia nervosa may be a helpful strategy at certain stages in the process of recovery and self-creation (or self-development). According to Glover, one way to do so might be by comparing them with more generally endorsed ideas about what a good human life is like, although this strategy may be less available at the early stages of anorexia nervosa. Another strategy may be the so-called ‘thank you’ test, where people who might be at some point incapable to see the seriousness of their condition and the value of treatment, at later stages realize it and become grateful (to doctors, therapists, family and possibly themselves) for changes they made. Another way of drawing on the newly induced self-illness ambiguity can be by addressing problems related to external pressures and other-dependence that are often reported in anorexia nervosa (Glover 2014, 362–365; see also Fuchs 2021). Glover suggests that an important therapeutic goal in the process of recovery is to gradually discover which values one would like to make one's own.

Finally, McConnell and Golova (2022) argue that self-illness ambiguity may also arise when one experiences uncertainty about whether one's self-narrative represents who one really is and that this kind of ambiguity is a useful phenomenon in therapy for addiction and should be first induced and embraced, before it can be resolved. An interesting parallel may perhaps be drawn here with the role of self-illness ambiguity in the course of treatment in anorexia nervosa. Some of the struggles described in the Hope et al. (2011) study seem to concern conceptualizing AN and the issue of whether and how anorexia should be integrated into one's narrative identity. But here as well, it is an open question whether a stable position on the matter should be achieved or whether some residual self-illness ambiguity may be conducive to forming a narrative. This leads me to a final point and qualification: given how varied and heterogenous experiences of illness and self may be, a one go-to suggestion for whether self-illness ambiguity ought in the end to be resolved may not be possible. We may need to be pluralists about the exact role of self-illness ambiguity.¹³

5. Concluding remarks

The above observations concerning anorexia nervosa can shed interesting new light on self-illness ambiguity and its significance for the process of recovery¹⁴ and self-development. Some mental disorders, like anorexia nervosa, seem to be initially ego-syntonic by involving key symptoms that take the form of ingrained and compulsive habits. Self-illness ambiguity may in such cases be a welcome resource by providing means for experiencing and/or attending to tensions related to such behaviours, by gradually gaining insight into the seriousness of one's condition and, arguably, by offering some means for questioning old and discovering new preferences and values. To the extent that anorexia nervosa may be initiated by the choice of specific behaviours and may become part of one's history, life experience and identity, the idea that we should be resolving self-illness ambiguity should, in my opinion, be given less emphasis for the purposes of recovery, as well as therapeutic and personal goals. This is compatible with the importance of narrative approach to self-development that may provide useful tools for understanding, integrating and living with the anorexia part of one's history and identity. It remains an open question whether self-illness ambiguity may be a resource in the case of other mental disorders.¹⁵

Acknowledgements

I would like to thank Roy Dings, Valentina Petrolini and Tad Zawidzki and the audience of the Philosophy and Psychiatry Talk Series organized at the University of Granada for helpful comments and discussions. I thank the anonymous reviewers for this journal for very helpful comments and suggestions on the manuscript. I would also like to thank the editors of this issue: Leon De Bruin and Roy Dings.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Anna Drożdżowicz

Anna Drożdżowicz (Inland Norway University of Applied Sciences) works in the philosophy of mind and language, as well as in the philosophy of psychiatry.

Notes

1 A full discussion of the (neuro)psychological and cultural mechanisms in AN goes beyond the scope of this paper. I leave the issue of whether AN is best understood as a disorder of the self (Amianto et al. 2016) for another occasion.

2 Ambivalence need not result in self-(illness) ambiguity. For a useful way of distinguishing between ambivalence and self-illness ambiguity in the psychiatry context, see Dings and de Haan (2022).

3 The notion of ‘the self’ or ‘who one is’ are notoriously difficult to spell out in a precise manner. Providing a detailed discussion of these notions goes beyond the scope of this paper.

4 Self-illness ambiguity may be understood as continuous with other forms of self-ambiguity that people experience outside the context of psychiatry. See Dings and de Bruin (2022) for a specific defense of this position and overview of empirical evidence that self-ambiguity is more common in everyday life than is often assumed.

5 I thank an anonymous reviewer for emphasizing this point and encouraging me to include it in this work.

6 A related issue is that current diagnostic manuals, by focusing on well-defined sets of criteria, do not provide viable resources for self-understanding for those who receive diagnosis (Tekin 2019).

7 For extensive discussion of limitations that come with the narrative approach to self-illness ambiguity, see Dings and de Bruin (2022).

8 I thank an anonymous reviewer for suggesting this point and a helpful example to illustrate it.

9 I thank an anonymous reviewer for suggesting that I address this point explicitly in the paper.

10 Pro-ana Internet websites contain messages promoting an extreme form of thin-ideal (Delforterie et al. 2014).

11 Erler and Hope (2014) argue for a more general claim, namely that the concept of authenticity is particularly valuable for self-development in the context of mental disorder in general and for people suffering from mental disorders and experiencing inner conflict between disorder and the self.

12 Hope et al. (2013) argue that some of the ways in which autonomy is compromised reduce competence to decide on treatment.

13 I thank Valentina Petrolini for suggesting to me that such idiosyncrasies may lead to a pluralist approach to the role of self-illness ambiguity.

14 I use the term “recovery” in a somewhat generic way, as it is often done in medical and psychological literature. In a recent paper, Friedman (2021) argues that the concept of recovery, as used in medical literature, blurs important distinctions between biomedical, phenomenological, and social perspectives. Self-illness ambiguity may be relevant for all three.

15 One case that might worth investigating in the future is generalized anxiety disorder (GAD, APA 2013).