An embodied belonging

Amenorrhea and anorexic subjectivities

Karin Eli


Until the publication of the DSM-V in 2013, amenorrhea was one of the four criteria that comprised anorexia nervosa. Diagnostically, amenorrhea played a definitional role, dividing the ‘strictly’ anorexic from their ‘subthreshold’, menstruating peers; however, the implications that menstrual cessation, and menstruation itself, held for the lived realities and identities of women with anorexia remain under-explored. In this article, I examine the positioning of menstruation and amenorrhea in the narratives of Israeli women diagnosed with eating disorders during the eras of the DSM-IV and DSM-IV-TR. I find that the participants’ narrative uses of amenorrhea mirrored, and at times explicitly engaged with, the official diagnosis of anorexia nervosa. Notably, although the participants invoked amenorrhea as a defining sign of illness, they did not cast menstruation as a sign of health rather, they spoke of their menstrual periods as contradicting their anorexic-identified selves. Amenorrhea, then, emerged as central in the embodied making of anorexic subjectivities.


Agripas, Jerusalem, February 2013Agripas, Jerusalem, February 2013

Introduction: Encountering amenorrhea

Several months before beginning the first round of my fieldwork research in Israel, I began recruiting participants through an informal network headed by Miri,[note 1]All names used in this article are pseudonyms. To maintain confidentiality, the participants’ identifying details have been concealed or altered.
 a participant with whom I was personally acquainted. I had originally designed the study to focus on anorexic women, and this description held true during Miri’s initial recruitment efforts. In March 2005, soon after she began recruiting potential participants, Miri sent me the following email message: 

I sent messages to all the people that I thought could be of use to your work, yet only 2 of them gave me their approval so far. One is [name]. I don’t know if you could use her for your work, cause she is defined “ED Nos.” which means eating disorder non specific. She’s anorexic but she never lost her period so that doesn’t answer the DSM criteria. [English in original]

While Miri defined her friend as anorexic, she implied that, for the purpose of research, this insider definition of anorexia would not suffice. As I had never discussed menstrual criteria with Miri, her emphasis on her friend’s menstrual status revealed a construction of amenorrhea as a defining characteristic of the disorder, one that might separate wanted and unwanted research subjects. Still, given Miri’s extensive experience with psychiatric ‘patienthood’ (see Brumberg 2000, 43), upon its associated participation in clinical research, I was not entirely surprised. This construction of amenorrhea was reinforced when I first called Miri’s friend, and she immediately told me, ‘I never lost my period’. Menstruation, then, seemed to disqualify women from claiming anorexia as their illness, with menstrual periods making an attenuated anorexia, a (presumably) lesser disorder whose incomplete and undefined status must be repeatedly disclosed, lest the researcher, interested in anorexic ‘truth’, be deceived. I realized then that far from being a universal experience among women with anorexia, amenorrhea was loaded with serious implications for identity validation, clinical recognition, and status hierarchy.

Following my exchange with Miri, I began to wonder: what happens when women are faced with the liminal status of being a menstruating anorectic? How is the biology of belonging negotiated? Amenorrhea was one of the four diagnostic criteria of which anorexia nervosa, as an officially coded Diagnostic and Statistical Manual (DSM) disorder, was comprised. As defined in the DSM-IV-TR (American Psychiatric Association 2000, 589), a diagnosis of anorexia nervosa required the following:

A. Refusal to maintain body weight at or above a minimally normal weight for age and height (e.g., weight loss leading to maintenance of body weight less than 85% of that expected; or failure to make expected weight gain during period of growth, leading to body weight less than 85% of that expected).

B. Intense fear of gaining weight or becoming fat, even though underweight.

C. Disturbance in the way in which one’s body weight or shape is experienced, undue influence of body weight or shape on self-evaluation, or denial of the seriousness of the current low body weight.

D. In postmenarcheal females, amenorrhea, i.e., the absence of at least three consecutive menstrual cycles. (A woman is considered to have amenorrhea if her periods occur only following hormone, e.g., estrogen, administration.)

While anorexia nervosa was included in the first edition of the DSM (American Psychiatric Association 1952), amenorrhea became part of the official diagnosis of anorexia nervosa thirty-five years later, when it was added to the DSM-III-R (American Psychiatric Association 1987), signaling its definitional usefulness. As Alan Young (1995) argues, beginning with the DSM-III (American Psychiatric Association 1980), American psychiatry redefined mental health disorders as analogous to ‘physical’ disease, such that psychiatric diagnoses became reliant on observable and measurable symptomatic expressions (1995, 96). This approach facilitated the standardization of diagnoses and allowed for the case aggregation and statistical analyses necessary for epidemiological studies (Young 1995, 102). Following this definitional logic, it is possible to observe how amenorrhea was mobilized as physiological ‘evidence’ of disorder. 

At the time of its official diagnostic debut, amenorrhea had been well established as a central feature of anorexia nervosa for more than a century, beginning with the very definition of the disorder (see Gull 1874; Laségue 1873a, 1873b). The psychiatric portraiture of anorexic amenorrhea left little doubt that this symptom was not merely the ‘natural’ correlate of severe weight loss. Though lacking consensus on the causation of anorexic amenorrhea, clinical reports consistently noted that many women with anorexia developed amenorrhea before reaching underweight (Bemis 1978). The absence of menstruation came to be viewed as having a significance that transcended weight loss, potentially implicating a unique etiology, and justifying the inclusion of amenorrhea as a separate diagnostic criterion. Thus, though it can be assessed only through self reports, amenorrhea was characterized as ultimately useful: a measurable physical symptom of discrete temporality, easily demarcating anorexic women from their thin (but healthy) counterparts, while drawing clinicians’ attention to the underlying and unseen morbidities of anorexia, such as osteopenia (Attia and Roberto 2009, 581). 

With the impending publication of the DSM-V (American Psychiatric Association 2013), anorexic amenorrhea re-emerged into the scientific limelight as the object of (another) substantial diagnostic revision: the American Psychiatric Association decided to remove amenorrhea from the definition of anorexia nervosa (2010). This removal reflected the increasing scientific scrutiny that was leveled at amenorrhea in recent years, with numerous studies examining, and indeed challenging, the very usefulness of amenorrhea as a diagnostic criterion. At the crux of these studies was the ambiguity that characterizes anorexic amenorrhea. Clinically, the amenorrhea criterion has proven a source of diagnostic stringency, excluding a large percentage of ‘subclinical’ anorexic patients from the diagnosis of anorexia nervosa, and relegating them to the catch-all diagnosis of Eating Disorder Not Otherwise Specified (EDNOS). However, studies comparing people diagnosed as anorexic with their ‘subclinical’ (EDNOS) peers consistently found that these groups do not differ on psychological, behavioral, and demographic measures (Cachelin and Maher 1998; Watson and Andersen 2003; Abraham et al. 2005; Gendal et al. 2006; Roberto et al. 2008).

The etiological connection of amenorrhea and anorexia nervosa, moreover, remains unknown. As Cachelin and Maher (1998) argue, amenorrhea was once considered a manifestation of ‘a primary impairment of hypothalamic malfunctioning’, and, since anorexia was thought to result from such primary malfunctioning, amenorrhea became a definitional requirement. Neuropsychological hypotheses have indeed suggested that anorexic amenorrhea indicated leptin-related adaptation to semi-starvation, luteinizing hormone abnormalities, metabolically mediated affective states, or elevated cortisol levels in states of psychological distress, among other phenomena (Golden and Shenker 1994; Copeland et al. 1995; Mantzoros et al. 1997; Marcus et al. 2001; Gendall et al. 2006). However, no hypothesis has proven satisfactory. Finally, even if amenorrhea continues to be conceptualized as an important criterion for anorexia, a number of researchers have called attention to its inherent limitation: its diagnostic usefulness is constrained to women of childbearing age, who do not use hormone-based contraceptives. This limitation excludes ‘non-stereotypical’ anorexic patients, including men and postmenopausal women, as well as sexually active (and other) women who take contraceptives, therefore restricting the usefulness of amenorrhea both in diagnosis and in clinical studies (Attia and Roberto 2009).

Yet these discussions of amenorrhea’s usefulness as a criterion have excluded the voices of anorexic women themselves. Not merely a diagnostic entity, amenorrhea is a meaningfully embodied experience. As anthropological texts have shown, menstruation is imbued with cross-cultural and culturally specific meanings (see Buckley and Gottlieb 1988), and the very experience of menstrual processes, such as premenstrual symptoms and menopause, is socially dependent and locally embodied (Gottlieb 1988; Lock 1993). Thus far, however, anthropological analyses of anorexia have paid limited attention to amenorrhea. In those analyses that engage with menstrual loss, amenorrhea is interpreted in reproductive and sensory terms. Rebecca Lester’s (1995) perspective on contemporary anorexia draws on her analysis of medieval women’s starvation practices, which she terms ‘food asceticism’. She suggests that these women were concerned with the breaching of their physical ‘boundaries’ through ‘biosexual processes (menstruation, sexual intercourse, pregnancy, childbirth, and lactation)’ (1995, 190). These processes, argues Lester, ‘both challenge the boundaries of the body and often become the foci of complex ideological and value systems that encompass, produce, and replicate concerns about a woman’s autonomy versus her dependence in a social framework’ (1995, 190). She thus interprets the rejection of food as an attempt to control ‘the boundaries of the self’; in this way, the non-menstruating fasting woman achieves a ‘solidifying’ of her physical boundaries (1995, 190). This ‘solidifying’ is accompanied by a distancing of the individual woman from her family and society – a highly meaningful process, given that the medieval women who practiced self-starvation also practiced (often solitary) religious asceticism in Catholic monastic orders.

Megan Warin (2003, 2005, 2010) also approaches amenorrhea as an embodied experience. Having conducted fieldwork research with forty-six anorexic people, forty-four of whom were women, Warin argues that menstrual fluids inspire a sense of abjection and are therefore rejected by anorexic women as part of a greater attempt to ‘erase, cleanse and even make disappear their own “dirty” and “disgusting” bodies’ (2003, 89; see also 2010, 146–50). This rejection, she argues, has social significance that transcends individual embodiment: ‘in negating the biological grounding of their bodies (of menstruation, pregnancy and childbirth) the women in this research were rejecting potential relationships that would connect them in terms of kin relations and social obligation’ (Warin 2005, 42). For both Lester and Warin, then, amenorrhea is highly meaningful in the dual dimensions of individual embodiment and social relations, as it provides anorexic women with a means to redefine their bodies as well as negotiate their expected social roles as wives, mothers, and reproducers of kinship.[note 2]Another line of analysis is offered by feminist and poststructuralist writings on anorexia, which interpret amenorrhea as negating oppressive constructions of fertile femininity (Bordo 2003, 155–56; Malson and Ussher 1996, 505).

The meanings of anorexic amenorrhea are not limited to the symbolic and sensory implications of purity, (in)fertility, and kinship. Using an inductive analysis, I examined anorexic amenorrhea as positioned within the narratives of the women who participated in my research, delineating the ways in which they invoked amenorrhea as part of their histories of eating disorder. Their narratives suggest that anorexic amenorrhea constitutes an embodied experience of defined illness, and is a centerpiece in the negotiation of anorexic subjectivity and belonging, wherein the diagnostic narrative of anorexia nervosa is invoked to legitimize suffering vis-à-vis self and others.  


This article is based on anthropological fieldwork research that took place from November 2005 to October 2006, and constitutes part of a larger study concerning the subjective experience of eating disorders in Israel.[note 3]The study was approved by the University of Oxford’s Social Sciences and Humanities Inter-Divisional Research Ethics Committee (under the auspices of the University of Oxford’s Central University Ethics Committee, June 2005), and by Israel’s Kupat Holim Clalit (health care fund), Helsinki Ethics Committee (August 2005).
 Thirty-five women and one man, all of whom were Jewish and Israeli, took part in this study. A large part of my research was carried out in Tel Aviv and the Sharon area, where I was based; however, I travelled to various regions of Israel to meet the participants, such that one third of the participants were from the Northern, Southern, and Jerusalem regions. Most of the meetings took place in a variety of non-clinical locations, chosen by the participants, including their homes, cafés, parks, university campuses, and shopping centers. Several meetings took place in a psychiatric facility in which one participant was hospitalized. The second phase of the study took place from July to September 2011, when I conducted follow-up meetings with twenty-three of the original thirty-six participants. I limit the present analysis to the narratives collected in 2005–2006, as the more recent interviews largely focused on processes of chronicity and recovery.

The participants ranged in age from seventeen to thirty-eight at the time of first interview, with a median and mean age of twenty-five years. They were recruited through three main routes: an informal network, largely developed by my key participant (n=11); a clinical network headed by therapists at a government-subsidized mental health clinic (n=15); and an Israeli online discussion board for people with eating disorders (n=8). Of the remaining two participants, one was an acquaintance, and the other was recruited through an eating disorders support organization. There was, however, some overlap between the groups, as five of the private network participants were also members of the discussion board or had received treatment at the clinic, two participants recruited at the clinic were members of the discussion board, and four participants recruited via the discussion board were acquainted with participants recruited through the informal network.

Recruitment procedures included several steps, which differed according to the recruitment route. The informal network participants were first contacted by the study’s key participant, who had met them all previously through clinical and non-clinical (mostly online) eating disorder contexts. The clinical network participants were initially approached by a clinical team member (social worker, therapist, or graduate student). In both the informal and the clinical networks, after potential participants indicated their interest in the study to their ‘mediator’, and had agreed for their contact details (email address or, in most cases, telephone number) to be shared with me, I contacted them to explain the study further, answer any questions they had, and ask if they might wish to participate. The participants recruited through the eating disorders discussion board were self-selected. I posted three public messages on the discussion board; each message included information on the study and invited readers interested in the study to send me a private message through the discussion board’s messaging system. I communicated with these participants through message exchanges and telephone conversations, provided them with information about the study, answered their questions, and asked if they might wish to participate. This recruitment process also served as a screening process, through which I could obtain some details on the potential participants’ eating disorders and past or current treatment, as well as make sure that they were old enough to participate. Several discussion board members also contacted me to ask general questions about my work, without the express purpose of participating, and I therefore did not recruit them to participate in the study. When I met with the participants, regardless of recruitment route, I began the first interview by providing an information sheet and a consent form, either in Hebrew or in English, both of which included detailed explanations of the study and what participation might entail; I also invited the participants to ask me any further questions they might have had before the interview. All considered, about one-third of those recruited – through a ‘mediator’ or through self-selection (on the discussion board) – did not participate in the study, having either openly decided not to participate or ‘dropped out’ (with or without specifying their reasons) during the recruitment process; in one case, I excluded a self-recruited sixteen-year-old because she was too young to participate.

All participants had anorexia nervosa, bulimia nervosa, or EDNOS (subthreshold anorexia or bulimia), and most were diagnosed during the acute initial stages of their disorder.[note 4]Of the four participants who were diagnosed years after the inception of their eating disorder, two said they had either bulimia or an undefined eating disorder, and were of average weight or overweight throughout their disorder; one said she had both anorexia and bulimia, and maintained borderline underweight throughout; and one said she had anorexia but was diagnosed in retrospect (when she had recovered physically, but not psychologically).
 At first interview, the participants’ duration of eating disorder ranged from six months to twenty-three years, with the majority living with the disorder for five years or longer. It is impossible to determine with certainty the exact diagnoses of some participants; while it is clear that all had severe experiences of eating disorder, some participants openly opposed their diagnosis or used different diagnostic labels to define their experience. For example, two participants who said they were diagnosed with bulimia explained they rejected this term in favor of ‘eating disorder’. Moreover, several participants spoke of multiple eating disorders, some of which (particularly conditions they defined as ‘compulsive eating’) went untreated, and, most likely, undiagnosed. And many participants who identified as anorexic or bulimic also described food-related practices associated with the ‘other’ disorder (binge eating or purging in the case of anorexia, fasting in the case of bulimia). Through fieldwork, I have found that diagnostic divisions between anorexia and bulimia do not capture the continuum of practice that characterizes eating disorder, and I have therefore chosen not to solve the diagnostic ambiguity presented by some participants.[note 5]The official distinction between anorexia and bulimia has been problematized in the clinical literature. In a critical analysis of EDNOS, Fairburn and Bohn (2005) argue that a ‘transdiagnostic’ category, uniting all eating disorders, would be more meaningful than the extant diagnostic divisions. Similarly, Milos et al. (2005) argue that, since patients with anorexia and bulimia demonstrate a ‘diagnostic flux’, when followed longitudinally, ‘a transdiagnostic approach’ would best capture the core similarity between the eating disorders.
 Still, while not imposing diagnostic definitions on participants, my fieldwork research engaged deeply with diagnosis, the details of which emerged as an important feature of the participants’ identity discourses.

The participants ranged along the spectrum of disorder and recovery, and most participants could not be assigned easily into the categories of ‘disordered’ or ‘recovered’. The participants’ engagement with these definitions was imbued with attenuation and negotiation, and did not always correspond to their disordered or recovered status as it might have been clinically defined. Amidst this definitional complexity, it is, however, possible to say that all participants received treatment either during the time of the study or prior to it, and that, with the exception of one participant who was in the initial stages of acute disorder, all participants were in varying stages of long-term disorder or processes of recovery. As for those participants who continued to meet clinical criteria for eating disorders – their narratives reflected their years of eating-disordered experience, as well as the undulations of recovery and relapse, such that there were broad narrative similarities between participants, however positioned along the spectrum of disorder and recovery.

This article is based on the semi-structured interviews that formed the core of my fieldwork research. The interviews were digitally recorded. I transcribed all recordings and analyzed the transcriptions, coding the texts to establish thematic frequencies and patterns. Following this process, I translated the salient interview selections from Hebrew to English, collating them to form complete sets of thematic excerpts.

Menstruation or lack thereof was not a focal point of my study, nor was it subject to direct questioning. The participants were not directed towards discussion of menstruation, and I questioned a few of them about it only after they mentioned menstruation or amenorrhea in their narratives.[note 6]There is no Hebrew equivalent for ‘amenorrhea’; the participants mentioned amenorrhea not as a word, but as a concept, describing it as ‘I didn’t have a period’ or ‘I was without a period’.
[note 7]Those participants who mentioned menstruation or amenorrhea only did so in narrative contexts related to eating disorders. I therefore kept my follow-up questioning focused on the contexts they mentioned, and did not ask about their pre-disorder experiences of menarche or menstruation.
 This was a decision I made early in my research, wishing to acknowledge the full spectrum of eating disorders (including anorexia, bulimia, and EDNOS), as well as avoid the biomedical reification of amenorrhea as an indicator of anorexia. Nonetheless, of the thirty-five women who participated in the study, eighteen mentioned menstruation or amenorrhea at least once in their narratives. Most of these women said they were diagnosed as anorexic (defined or subthreshold) or had dual diagnoses of anorexia and bulimia; however, this group included two women with bulimia who also had a history of underweight (and, in the case of one, a possible, albeit undiagnosed, bout of anorexia). My analysis is therefore based on the unprompted mentions of menstrual periods in the narratives of these eighteen participants. Such analysis, focused on spontaneous expressions, allows for an understanding of the positioning of menstruation within participants’ illness narratives, exploring the context in which menstruation was mentioned, what role it had in the timeline and interpretation of illness, and what these mentions communicate about the participants’ conceptualizations of menstrual periods. 

Addressing the fertility question

When I began fieldwork in Israel, I expected to encounter accounts of anorexic amenorrhea anchored in the theme of fertility. Israel’s fertility rate (3.03 children per woman in 2010 [OECD 2013]) exceeds those of all other OECD countries, and, as the sociological and anthropological literature on Jewish Israeli women emphasizes, maternity and reproduction are formative elements in Israeli women’s national subjectivities (Berkovitch 1997; Benjamin and Ha’Elyon 2002). Women’s infertility, then, is positioned as both a personal and a national concern, one which merits state-supported medical intervention, with the government subsidizing in vitro fertilization (IVF) for each woman who qualifies for treatment, funding IVF cycles for up to two live babies per patient (Birenbaum-Carmeli 2004; Waldman 2006; Sperling 2010). Concerns about demographic growth may underpin Israel’s pronatal policies, but researchers note that Israeli women’s engagements with fertility issues also encompass a number of socio-religious, communal, familial, and experiential dimensions, where childbearing is framed as a woman’s pathway to personal fulfillment and social belonging, and where familial kinship is framed as the foundation of nation and society (Remennick 2000; Birenbaum-Carmeli 2004).  

Given the centrality of fertility in Israeli society, and the importance of motherhood to Israeli women’s subjectivities, it seemed likely that the participants’ discussions of amenorrhea would relate to these wider social meanings. Indeed, Sigal Gooldin’s (2002) ethnographic analysis, which she based on fieldwork at a children’s eating disorders ward in Israel, explores amenorrhea as a sign of looming infertility. In Gooldin’s (2002) account, amenorrhea is the key to a uniquely Israeli anorexia: an anorexia that constitutes a fusion of ideals, expressing both globalized thinness and Israeli fertile femininity, thus forming the amalgam that Goodlin terms ‘a glocal disorder’. Along these analytic lines, Gooldin argues that, for young Israeli women with anorexia, potential reproductive ability is the core concern in treatment-seeking and recovery, and that amenorrhea is central to the rationales and practices that define eating disorder treatment. She writes that the ward’s treatment team encouraged parents to hospitalize their daughters using discourses of future infertility; accordingly, when the goal of menstruation had been reached, nurses celebrated first (resumed) menstrual periods as public events at the ward. The patients, too, writes Gooldin, viewed the resumption of menstruation as a goal of treatment, as they wanted to become mothers and would not imagine themselves otherwise. This, Gooldin argues, is where a line can effectively be drawn between Israel’s young anorexic women and the so-called classic patients described by Hilde Bruch (1974): whereas the latter welcomed amenorrhea as a desired cancellation of femininity, the former yearned for menstruation and the restoration of their reproductive potential.

I discovered, however, that, for many of the women who participated in my research, amenorrhea held meanings unrelated to fertility.[note 8]While it may seem counterintuitive that the adolescents who participated in Gooldin’s (2002) study expressed greater concern with fertility than my study’s adult participants, Gooldin notes that her study’s adolescent participants endorsed dominant discourses that link motherhood and military service as essential to normative Israeli feminine identity. Whereas Gooldin’s participants engaged with these discourses in the future tense, the great majority of my study’s participants had already experienced military service (or exemption from service, in most cases on mental health grounds). It is quite possible that, being adults, and having engaged with military institutions, my study’s participants had greater capacity to express ideas that diverged from the dominant social discourses espoused by parents and clinicians.
 As might have been expected, narrative connections between fertility and menstruation (or lack thereof) appeared in this study as well. Yet these connections appeared less frequently than anticipated. Of the eighteen women who mentioned menstruation or amenorrhea, five did so in the context of fertility. For two of these women, fertility appeared in its expected context: Galit, who had two children after recovering from anorexia and bulimia, linked her past amenorrhea with the difficulties in conceiving she experienced years after her periods resumed; Danielle, who had recently become engaged, spoke of her desire to have children as a driving force for recovery, drawing a direct connection between anorexia, amenorrhea, and infertility. The other women who spoke of amenorrhea and fertility, however, offered perspectives that unsettled normative notions concerning the desirability of reproduction. Kinneret, who struggled with hercontinuing menstruation, told me she had ‘no aspiration to have children’; similarly, Stav, who said she was ‘bothered’ by the sight of her menstrual fluid, explained that amenorrhea, which equaled infertility, would enable her to become unattractive to men and thus avoid the emotional attachment she dreaded; and Nili, who began to menstruate after several years of amenorrhea, said her amenorrhea was ‘convenient’, since it freed her from having to take contraceptive pills. Notably, these engagements underscored the significance of fertility in the construction of Israeli feminine identity. While some chose to embrace it and others to negate it, fertility, for these women, constituted a narrative element against which to define themselves as normative or subversive (see Riessman 1990). 

Overwhelmingly, however, the participants in the present study embedded amenorrhea in neither fertility nor femininity (which was mentioned by four women), but in discourses of illness, health, and recovery. For thirteen of the eighteen women who mentioned menstruation in their narratives, amenorrhea was directly linked to disorder, invoked as a clearly recognizable indicator of illness, and used as a near synonym of anorexia. Amenorrhea thus became a crucial component in the articulation of an anorexic identity. With amenorrhea indicating infertility and biomedical (DSM) diagnosis, it is possible that amenorrhea, invested with the implicit equating of illness and infertility, was used as culturally appropriate shorthand to demarcate the healthy (and fertile) from the anorexic (and infertile). However, as the participants’ narrative uses of amenorrhea mirrored, and at times explicitly engaged with, the official diagnosis of anorexia nervosa, amenorrhea was more evidently embedded in the context of biomedically-defined disorder. The rest of this article discusses this interweaving of amenorrhea, illness, and subjectivity.

Amenorrhea as a sign of lost health

The first time Aya and I met was on the day she was diagnosed with anorexia nervosa. Aya was eighteen years old, on the brink of joining the Israeli Defense Forces, and somewhat overwhelmed by the rapid deterioration of her condition. After months of continuous self-starvation and compulsive exercise, she had become severely underweight, and was concerned about her impending military service. ‘Will they be able to see it on me?’ she asked, worried that she had come to embody her disorder so evidently that her superiors might immediately discharge her, since, in Israel, people with eating disorders can be (and frequently are) disqualified from military service on mental health grounds. Despite the severity of her condition, Aya was uncertain about her disorder; it was only when her psychologist officially diagnosed her that she accepted the term ‘anorexic’. Her doubts, she explained, resulted from the fact that she continued to eat, albeit in a very restricted way, and from the absence of the physical symptoms which she considered essential:

I just took it and told myself that anorexia is from a certain weight, that starvation means not eating at all. Because for me [starvation] was a fast. I didn’t know that it can be – that people eat, they just eat very little and – don’t know, and they don’t have to be at a weight where they need hospitalization, where their hair must fall out, or where their period must stop. Like, it was for me, the physical definitions – so I looked for the physical signs. And I didn’t find them, so I didn’t see myself as belonging to this category. So today she just finally included me in this issue.

In Aya’s narrative, hospitalization, hair loss, and amenorrhea were equated as manifestations of utter, defined, and severe illness. All three signs were positioned at the extreme of disorder, and all involved an external and undeniable loss of health. In not findingthese signs within her, and with her body presumably still demonstrating health (and continuing to menstruate) Aya felt unable to name her condition. Aya’s symptom-based definition of anorexia, it should be noted, was not simply a case of idiosyncratic definition. Aya told me that, before seeking psychological care, she had read about eating disorders. The image to which she adhered, an image of defined disorder based upon physical symptoms of lost health, was one propagated by popular discourse of clinical standards – a portrayal of the stereotypical anorectic as utterly fragile, incapacitated, and amenorrheic.[note 9]See Lager and McGee (2003) for a critique of media portrayals that essentialize anorexic women as radically thin and conspicuously ill.
 A menstruating woman, therefore, has a body that speaks of the very health that the ‘anorectic’, by definition, cannot have.

Whereas Aya was uncertain about having anorexia (prior to the authoritative act of naming via diagnosis) due, in part, to her menstrual status, other participants spoke of denying that they had anorexia, despite realizing that they became amenorrheic. For Emily, who had been ill with anorexia for more than a decade, menstruation was of great narrative importance. When I asked her when she became ill with anorexia, her response centered on amenorrhea:

Emily: It’s also impossible to say when I became sick because it’s, I’m sure you know it, it’s something that’s, like, not –

KE: It’s a process –

Emily: That’s it – but officially, like, when I was already without a period and without such things, that was at age eighteen. At age eighteen my period stopped, and then, like, I got osteoporosis throughout my body and such beautiful things.

Emily found it very difficult to define the beginning of her eating disorder; at one stage, she told me she remembered having disordered eating habits as an infant. Amenorrhea, therefore, functioned as a marker of recognizable, defined disorder: the culmination of a ‘process’ of becoming. This initial indicator of lost health, as conveyed by Emily, then caused osteoporosis – an implicitly inevitable complication (and one that led Emily to considerable agony due to stress fractures). Yet, while she defined her anorexia as beginning with the loss of menstruation at eighteen, it was not until several years later that she began to identify as anorexic. When speaking of her initial denial of this definition, Emily again cast amenorrhea as central: ‘I said all the time that there’s no way that I’ll be anorexic because I love eating too much. Even when I was already without a period several months I still claimed there’s no way because I love eating too much’. By focusing on amenorrhea as the central feature of disorder, and by emphasizing her own chosen obliviousness to it, Emily conveyed the severity of her denial. Amenorrhea, as Emily cast it, was the crucial proof of anorexia: the ultimate disorder-defining feature, an undeniable sign of illness that she, somehow, chose to ignore. In citing her lost menstrual periods, Emily did not have to illustrate the gravity of her condition; it was understood that I, as her audience, would immediately understand that which the lack of menstruation signaled, as amenorrhea was a synecdoche for the disintegration of the body, the most evocative, and revelatory, symptom of all.

Just as the narrative positioning of amenorrhea conveyed the magnitude of denial, so did some participants mention their lost menstruation to capture the severity of their conditions. Nili, who developed anorexia immediately before she was drafted to the military, managed to conceal her eating disorder from her commanders, and eventually became a commanding officer despite her ongoing disorder.[note 10]I have written elsewhere about the experiences of participants whose eating disorders developed, worsened, or subsided during their military service (Eli 2014).
 Speaking of her decision to join the professional, volunteer military ranks, she told me:

I was a good commander. … They promoted me. … I was professional, they appreciated me at work, so why not? And I was sick, yes, I was still sick. I still didn’t have a period. The entire military service I didn’t have a period, the entire military service. The last time was at basic training. From that moment, nothing until only about a year after my discharge.

Nili collapsed the entirety of her illness into the single sign of amenorrhea, the juxtaposition between her military profession and her physical disintegration powerfully conveyed through this central indicator of lost health. Amenorrhea was positioned in this excerpt as the ultimate, undeniable, and universally understood proof of illness; there was no need for further elaboration – Nili’s self-portrayal as a paradoxically anorexic officer was complete.

It is important to remember that while participants constructed amenorrhea as an individual indicator of illness, they positioned it as an absence that transcended the personal and merited social recognition of one’s illness. This construction of amenorrhea was evident in Vered’s narrative. Vered had been ill with anorexia for about a decade when we first met; at the beginning of her first interview, she told me, ‘I talked so much about my disease that I prefer you’ll ask [questions]’. I then asked if she had previously participated in research, and she said she had been interviewed, quite awkwardly, by a panel of American medical students. Our discussion quickly turned to issues surrounding medical training. Physicians were trained inadequately, Vered argued, relating the following anecdote in support of her argument:

I was at a gynecologist’s once, and – don’t know, somehow we got to the fact that I didn’t have a period several years because of the disease, and, umm, and then he – and it’s not someone I know, it’s someone I went to once, and after that one time it was absolutely enough for me and I didn’t see him again and that’s for the best. And I told him that I’m sick with anorexia, several years without a period – I didn’t have a period for several years, and just the questions that he asked me, I tell you – I was, I won’t even give him the credit to say these are the questions of an ignorant [person], [these are] the questions of an idiot, do you understand? It’s just unbelievable, there’s no – it simply gives me chills to recall it. He’s asking me stupid questions, like, it’s good that he didn’t ask me what I didn’t eat, how much I did eat, and – wow, and he started telling me that they conducted studies that it’s actually good for the body to fast… things I read myself, that it’s good to cleanse the gastric system, but he talked about it actually as a lifestyle – that they did research … that it’s actually necessary to live on a very restricted amount of food, I even remember that he detailed the amounts, understand, and to whom did he choose to give those details?

Amenorrhea was the pivotal point of Vered’s story, instigating the physician’s advocacy of dietary restriction. But amenorrhea was more than a mere trigger for offensive commentary. In employing the device of repetition: ‘I’m sick with anorexia, several years without a period – I didn’t have a period for several years’, Vered positioned her menstrual status as central to this story. She cited her amenorrhea as an undeniable sign of illness, with its protracted duration exacerbating – and cementing – her anorexic status. Amenorrhea was thus cast as revealing both her severe condition and the physician’s transgression, such that the answer to her concluding, rhetorical question, ‘to whom did he choose to give those details?’, could only be: to a woman so severely self-starved that she did not menstruate for years. Notably, Vered could have chosen other evocative markers of illness in this context, particularly her frequent past hospitalizations and nasogastric feeding. Yet amenorrhea, as a quintessential indicator of anorexia, sufficed.

Menstruation and deceptive wellness

One of the primary goals of therapeutic interventions for anorexia nervosa is the restoration of menstrual periods, and regular menstruation is employed as an important, though insufficient, indicator of recovery in the clinical literature (Strober et al. 1997; Pike 1998). With the participants’ casting of amenorrhea as indicative of illness or lost health, one might expect that the resumption – or continuity – of menstruation might be constructed as an indicator of health or recovery. However, as the narratives of several women revealed, that was not the case. ‘Recovering’ participants whose menstrual periods resumed spoke of a dissonance between this presumed (external) recovery and the ongoing (inner) disorder, implying that menstrual status revealed nothing of their disordered realities. Nili, who was previously quoted about her amenorrhea during military service, outwardly led a healthy and productive life. When I asked about her self-definition, she told me, ‘I live at peace with my fixations’; for her, however, that was not recovery:

Today? I’m still disordered. I’m not – I’m at normal weight, but, in the head, I have a great mess. That means, umm, my eating is strange today as well. Umm, wow, it embarrasses me to talk about it, what’s today, because it’s easy to talk about what’s past. Today I see a dietician once every two weeks, umm, gain-lose-gain-lose, now I’m in – like, I returned, I always lose weight in the summer, gain in the winter, now I’m again at my normal weight, like, it returned just in the past two weeks. Now I’m at excellent weight, there’s a period, not normally, but there is, sometimes yes, sometimes no, or something like that, but it exists somewhere. In terms of food, it’s complicated for me today, too.

Mentioning her ‘normal’ or ‘excellent’ weight three times, Nili appealed to the most apparent, and presumably salient, facet of anorexia, only to deflect it. She portrayed the ‘normalcy’ of her weight not as a sign of recovery but rather as part of her ongoing disorder, a process of patterned recovering/relapsing – just as her period may have resumed, but ‘not normally’, indicating the continuing existence of disorder. Contrasting her symptom-free appearance with the disordered behavior and cognition that she labeled ‘strange’ and ‘messy’, Nili conveyed that the essence of her disorder, that which rendered her ‘still disordered’, was the hidden, internal process of obsession and compulsion, a process she shielded from public view and which was a source of embarrassment to her, but which, recognized or not, defined her condition. The reappearance of Nili’s menstrual periods, though undoing her anorexic definition, was therefore a deceptive indicator of recovery, its irregularity speaking of tumult, of an undulating health, not fully regained. 

Oryan, an eighteen-year-old recovering from anorexia (to which she referred, in her narrative, as ‘eating disorder’), similarly portrayed her process of recovering through a demarcation of ‘inner’ and ‘outer’ aspects of disorder: ‘Physically, like, physically, I recovered, like, I gained weight, I got my period. But there are still the thoughts… because it’s here twenty-four hours, it sits here twenty-four hours. It doesn’t move’. Oryan experienced the resumption of her menstrual periods not only as unreflective of her ongoing eating disorder, but also as an unsettling event. At the conclusion of her initial narrative, in which she outlined the course of her disorder, Oryan said: 

And until now I cope with it, [it’s an issue] of how to cope with the turnaround, and suddenly after a year and a bit that I didn’t get my period, getting my first period which is, wow, which says that you’re [at your target] weight now, which says that you’ve grown fatter.

Oryan was weighed at her dietician’s office every week; clearly, she was aware of the weight-gaining process, of the target weight she was supposed to achieve, and of the fact that, as she ate according to a weight-gain regimen, she would, indeed, achieve this weight. Yet her first menstrual period – not the number on the scale – was the marker of change. It spoke to a presumed ‘truth’ of the body, a regained health she could not deny. And in its symbolic magnitude, Oryan’s first period was like a second menarche: just as menarche might symbolically demarcate childhood and womanhood, so did this period demarcate Oryan’s past ‘sick’, ‘thin’ self from the self she now conceptualized as fat. But this ‘recovering’, postmenarcheal self was by no means healthy. On another occasion, when I asked about her self-definition, Oryan again referred to the contrast between her menstrual status and her experience of ongoing disorder:

Physically, everything’s good. Like, I’m above [target] weight, I’m getting my period. But the thoughts and the wish to lose [weight] only increased, and accepting myself the way I am only grew more distant, like. But I hope it will be good, and that it’ll pass. Because, you know, I have the – every week, I get weighed, let’s say, and then, let’s say a while ago I lost [weight]. There’s no better feeling than that. How do I define myself? As someone who wants to lose [weight] but her body doesn’t allow her. Once it allowed me, today no more.

Creating a dichotomy between her body and her thoughts/emotions, Oryan offered a dual conceptualization of her condition, referring to the physical aspects of weight and menstruation as the reflection of a clinical or public concept of recovery, only to cast the physical as the mere appearance of recovery, its falseness brought into relief through her continuous, even exacerbated, disordered feeling and thinking. She illustrated this dichotomy through the example of her ongoing enjoyment of weight-loss, implicitly contrasting the therapeutic context in which her weekly weighing occurred (the eating disorders clinic) with her continuous use of this clinical measure as an indicator of a disordered target. In her concluding description of herself, Oryan deepened the dichotomous relationship of disordered self and seemingly recovered body, portraying her body as acting in opposition to her will, imprisoning her disordered wish within an unrelenting, deceptively recovered frame.

Menstruation and the making of an undefined anorexia

My initial encounter with the role of amenorrhea – Miri’s email message, recounted at the introduction to this article – did not imply that participants simply accepted this criterion without challenge. Indeed, the category of EDNOS (which the participants interchangeably called ‘undefined eating disorder’ or ‘non-specific eating disorder’) was subject to contestation. As the default diagnosis for all eating disorders that did not meet full clinical criteria for either anorexia nervosa or bulimia nervosa, EDNOS was a nondescript, amorphous category, used to describe anything from subthreshold anorexia or bulimia to eating disorders that exhibit symptoms of both, alongside binge eating disorder (American Psychiatric Association 2000, 594–95).[note 11]In the DSM-V (American Psychiatric Association 2013), EDNOS has been replaced with Other Specified Feeding or Eating Disorder (OSFED). Binge eating disorder is now a separate, specified diagnosis.
 Most participants who were diagnosed with EDNOS therefore chose to describe their disorders through the ‘defined’ categories of anorexia and bulimia, usually presenting no engagement with their own undefined diagnosis. However, in the cases of three women who were hospitalized for subthreshold restricting anorexia, the EDNOS category was explicitly stated, explicated, and resisted through the reinterpretation of menstruation. One such incident occurred with Kinneret, who had been ill for six years at the time of her first interview, and who continued to meet all clinical criteria for anorexia, save for amenorrhea. When I used the word ‘anorexia’ to refer to a phase in Kinneret’s eating-disordered experience, she immediately corrected me, saying  ‘it’s not anorexia, like, according to the diagnosis’. As she explained earlier:

I didn’t lose [my] period. I was fifty-five [kilos], I went down to thirty-eight, but my period [remained]. The truth is that I hated it. I have friends who, one kilo below target weight, half a kilo below target weight, and enough, they don’t have it. And that friend, the one I told you [about]… she wants to have children so much, it’s one of her aspirations in life, and she, half a kilo under target weight, doesn’t have it. And I wanted to get rid [of my period] – [but] it didn’t succeed.

Kinneret’s attenuated diagnosis coalesced with her perceived inability to cease to menstruate; yet, although she said her diagnosis was undefined, she did not embrace experientially undefined status. Having reached severe underweight, Kinneret explained that it was her body that did not let go of this physiological process – whereas she was hoping to relinquish it. Her diagnosis, therefore, was dependent on a physicality that was beyond her control, unreflective of her own wishes and experiences, an idea she underscored through the example of a friend who wishes for a healthy reproductive life but nevertheless suffers from amenorrhea. Juxtaposing her menstrual status with that of her anorexic friends, Kinneret implied that there were inexplicable differences between equally disordered bodies, rather than meaningful differences between the women labeled ‘anorexic’ and those labeled ‘not otherwise specified’. 

This designation of ‘bodily’ versus ‘real’ criteria for anorexia nervosa also appeared in Zoey’s narrative. Zoey, who had been ill for more than a decade when she sought hospital treatment, described her diagnosis with a sense of revelation:

So that was when I went to [psychiatric hospital] when they finally said, you know, you really do have anorexia, just the type that’s undefined, because I had my period at forty kilos, then, I mean, I don’t lose my period. My body is so strong. That’s when it came back. I didn’t go below forty kilos.

With her eventual diagnosis of (undefined) anorexia, Zoey received an affirmation that her prolonged undefined status relied on her menstrual status only; in all other respects, the clinicians acknowledged, she ‘really’ had anorexia. The reality of this anorexia, however, contrasted with the reality of her body, which, Zoey explained, ‘is so strong’. The strength that Zoey attributed to her body, with its continuous, ‘normal’ physiological functioning, was in opposition to her experiential reality, wherein she was severely underweight and felt so seriously ill that she turned to hospitalization. Thus, her EDNOS diagnosis was unreflective of the actualities of her eating disorder. However, the implication of continuing to menstruate transcended an ‘undefined’ diagnosis. At another point in her narrative, Zoey told me, ‘I got caught up in being sick as an alternative to the life I had. I thought it was a better alternative’. When I asked her, ‘What did you get out of being sick?’, she replied,

I got – [pause] I want to say I got nothing out of it, nothing – I tried very hard to get attention, by being sick, but I wasn’t sick enough ever, I was never thin enough, I was never – I never lost my period enough times, all my blood tests always came back perfect. I didn’t get very much out of it at all.

In over twenty years of eating disorder, Zoey had amenorrhea for two consecutive years, after which her menstrual periods resumed. Along with the other measurable criteria of illness –body weight and blood tests – this very continuity of menstruation undermined Zoey’s attempts at illness, making her insufficiently disordered, her claims of illness delegitimized. In attesting to ongoing health, Zoey’s menstrual periods not only demonstrated the supposed fruitlessness of her efforts, but also led to attenuated social and clinical recognition – as a menstruating woman with anorexia, she perceived herself as a lesser, undefined anorectic, illegitimately claiming a title that clinicians would not award. 

Similar frustration over an attenuated diagnosis appeared in Grace’s narrative. Due to the appearance of a single menstrual period after many years of illness, Grace was diagnosed with EDNOS, despite having been anorexic by all other criteria for several years. She deeply resented her diagnosis:

It’s very irritating. It’s very irritating. You know that when I was hospitalized at the [psychiatric] ward I weighed forty-three [kilos], I think, but umm, very strangely I had a period once – and that’s why I, to this day, I’m listed at the ward as like, anorexia restricting type, but hospitalized in a non-specific condition, like, I – they acknowledged that I’m anorexic but – wrote that when I was hospitalized I was in a non-clinical condition. Like, what’s their problem?

Grace’s frustration with her diagnosis, however, reveals much more than indignation over her relegation to the lesser category of EDNOS. As Grace contended, having ‘acknowledged’ her as anorexic, the clinicians nonetheless deferred to the authority of diagnostic categories, thus paradoxically labeling her condition as ‘non-clinical’ while clearly determining that she was sick enough to require psychiatric hospitalization. Moreover, in describing the appearance of her period as occurring ‘very strangely’, Grace positioned her menstrual status as irrelevant – an inexplicable physiological event, incommensurate with her true eating-disordered experience. After Grace mentioned her subclinical diagnosis, I decided to pursue the topic further.

KE: I know it’s a terrible question, but there’s a stereotype that anorexics hate getting their periods, so I wanted to know if it’s true?

Grace: Of course. It’s like one day, I was laughing because I talked with my friend and she says to me, like, tells me, ‘Everything’s bad and in addition to everything the son of a bitch arrived yesterday’. I told her, I understand. And then I talked with my [healthy] friend, I told her, ‘You know what, I just thought about it that if I had told you, in addition to everything, the son of a bitch arrived yesterday, what would you think?’ She tells me, ‘Who arrived, what arrived?’ I said ‘It’s so beautiful, the secret language of eating disorders’. Umm – why? Because it symbolizes that the body is healthy, and you don’t want the body to be healthy.

KE: And what about the sensation?

Grace: And also femininity, sexuality, and blah blah blah.

KE: And what about the sensation? Like, it reminds you that your body’s there, no?

Grace: Yes. First, it’s also an unpleasant sensation because sometimes, there’s edema, swelling, all sorts of disgusting things, but it’s not because of that, but the very idea that like, your body is healthy enough, because – because a period is really something that you get when you’re really healthy, it’s not something that the body – it’s luxury.

This was, I felt, a very telling exchange, with menstruation pulled in numerous discursive directions: my own anthropological orientation, attempting to ground amenorrhea in embodied experience; the psychological emphases on femininity and sexuality, to which Grace paid a dutiful, if unenthusiastic, homage; and Grace’s own insistence on the ‘insider’s truth’.[note 12]Here, Grace consciously referenced Peggy Claude-Pierre’s (1997) popular psychology book, The Secret Language of Eating Disorders. As Grace explained to me, she was not a devotee of Claude-Pierre’s ideas; she felt, however, that the book’s title reflected a central element of her experience within the eating disordered community.
 For menstruation doescarry all the meanings that authoritative discourses attach to it: it is acknowledged as a symbol and embodied sensation of femininity, of sexuality, of adulthood, of being. But for Grace, as for other women with anorexia, menstruation was, beyond all things, a manifestation of health, unruly and unwanted, positioning the body outside the realm of control, belying one’s anorexic practice, and countering her anorexic subjectivity.   

Conclusion: Amenorrhea and anorexic subjectivities

Amenorrhea was one of several (non)vital signs this study’s participants used in defining and communicating their conditions. Hypothermia, bradycardia, hypotension, fainting spells, fatigue, weakness, and constant nausea all made appearances in the narratives. Yet amenorrhea was the most pervasive and evocative sign of all. Laden with meaning, amenorrhea, even when seemingly mentioned in passing, was invoked at crucial points in the narratives, capturing the essence and gravity of illness, and identifying the narrator as truly anorexic.[note 13]In the narratives of the participants who had been diagnosed with both anorexia and bulimia, or with bulimia alone, mentions of menstruation and amenorrhea appeared in relation to the severity of their illness while underweight, or, for the former group, within discussions of their anorexic subjectivities; amenorrhea was not part of these women’s constructions of bulimic identity.
 In all eighteen narratives, amenorrhea was unequivocal; no participant spoke of it as existing alongside, or despite of, processes of recovering. Menstruation, however, was deeply contested. Though amenorrhea was narrated as a sign of illness, menstruation did not assume its opposite role as a sign of health. The participants actively negotiated the roles of menstruation: some spoke of their menstrual periods as contradicting their still-disordered, deceptively recovered selves, while others challenged their diagnoses of EDNOS, arguing that ongoing menstruation belied their anorexic identity – an identity independent of amenorrhea.    

As the participants’ narratives revealed, menstruation was a life force, not in attesting to the fertility yet to come, but in indicating the present-tense health of the body. Given this study’s setting in Israel, a nation characterized as profoundly pronatal (see Berkovitch 1997), the infrequent appearance of fertility in discussions of menstruation and amenorrhea was surprising. Above all, amenorrhea was a touchstone for illness, and the participants did not cast the resumption of menstrual periods as an achievement or a goal. To menstruate, in many ways, was to fail: to fail in controlling one’s body, in crafting an embodied anorexic identity, and in reaching the conspicuous outer absence that would validate the disorder within.

This conceptualization of amenorrhea as a marker of identity was not the idiosyncratic expression of a ‘disordered’ community. While illness narratives speak of individual experience, they also reveal the social environment in which personal stories are constructed and told (Good and DelVecchio-Good 1994; Skultans 1997; Steffen 1997). Arthur Kleinman (1978) argues that biomedicine is a historically and socially informed ‘cultural system’, in constant interaction with the society in which it operates. And, as the narratives of this study’s participants conveyed, it was biomedicine that provided the central context for the experience and interpretation of anorexic amenorrhea. The participants’ narrative emphasis on amenorrhea was in dialogue with the clinical discourses that reified it to the position of disorder- and identity-defining indicator. To be a menstruating woman with anorexia transcended the individual embodiment of (false) health, for it also meant receiving lesser recognition from the authorities vested with diagnosing, and thereby ‘naming’ a patient socially. Through medical diagnostic criteria, menstrual fluid was imbued with the property of defining praxis and personhood. It therefore became, for the participants in this study, a substance of health, to be negotiated and employed in the quest towards self-definition.

In her ethnography of anorexic embodiment, Warin (2010, 147), writing about amenorrhea, suggests that ‘[w]omen who think or hope they have anorexia also weigh this diagnostic marker as a defining signal that they have the credentials to join the club’. This need for official ‘credentials’ in naming oneself ‘anorexic’ arose time and again throughout my own fieldwork research. Anorexia, as one participant, Vered, explained, was a ‘title’ that had to be earned ‘with blood, sweat, and tears’. Within that framing, amenorrhea emerged as the most evocative of ‘credentials’ – at once biologically salient and socially meaningful, a physiological marker of anorexia that (unlike weight) was beyond the person’s control. Drawing on Paul Rabinow’s (1992) concept of ‘biosociality’ – but reframing the ‘bio’ and the ‘social’ it connotes – I suggest that the term ‘embodied sociality’might help to elucidate the centrality of amenorrhea for women with anorexia. Attending to embodied socialitywould allow us to place the experiencing body at the center of social being, to ground the dynamics of subjectivity-making within the lived experience of bodies, and to bring to the fore a subjective understanding of biology rooted not in the quest for a scientific ‘truth’ (see Rose and Rabinow 2006), but in sensory negotiations, unique to a community of sufferers. The case of anorexic amenorrhea suggests a particular form of ‘biosocial’ membership – a membership often unspoken and unorganized, hidden from view and experienced alone, where the boundaries between diagnostic naming and self-naming, between physiological processes and the making of subjectivities, are blurred. In the often-isolated existence that marks eating disorders, sociality, for the women who participated in this study, was experienced in the right to identify and name oneself – in the ability to claim a certain lived subjectivity as one’s own. 

With the removal of amenorrhea from the DSM-V definition of anorexia nervosa, it remains to be seen whether, and how, the role of menstrual cessation might be recast in eating disordered women’s definitions of themselves, their disorders, and their membership in the ‘club’. From the perspective of access to care, the DSM-V revision could carry important implications. As some proponents of the revision have suggested (Attia and Roberto 2009), the exclusion of amenorrhea from the diagnostic guidelines entails the inclusion of more women, previously diagnosed with EDNOS, in the defined category of anorexia nervosa, and thus may increase these women’s clinical visibility and access to care. At the same time, however, the exclusion of amenorrhea might have a reverse effect: in those cases where amenorrhea is a main indicator of illness (for example, where a woman has amenorrhea but is borderline underweight), de-emphasizing it as a diagnostic indicator might lead to reduced access to care.[note 14]I acknowledge one of the anonymous reviewers for highlighting the important issue of access to care.
 From an embodied socialityperspective,would amenorrhea, designated as consequence or correlate of anorexia, rather than as a diagnostic criterion, become only a narrative footnote, to be replaced with another, more diagnostically salient characteristic? Would the evocative power of amenorrhea transcend the vicissitudes of formal diagnosis, such that it would continue to define anorexia, at least in narrative? While the future significance of anorexic amenorrhea is currently unknown, the findings presented here strongly suggest that biomedical definitions gain expression in the lived and storied realities of women with eating disorders.

About the author

Karin Eli is a postdoctoral research fellow at the University of Oxford’s Institute of Social and Cultural Anthropology. Her research focuses on the subjective experience of eating disorders, and she has conducted extensive narrative-based research among people with eating disorders in Israel and the UK.


Abraham, Suzanne F., Bianca Pettigrew, Catherine Boyd, Janice Russell, and Alan Taylor. 2005. ‘Usefulness of Amenorrhoea in the Diagnoses of Eating Disorder Patients’. Journal of Psychosomatic Obstetrics and Gynecology 26: 211–15.
American Psychiatric Association. 1952. Diagnostic and Statistical Manual of Mental Disorders: DSM-I. Washington, DC: American Psychiatric Association.
American Psychiatric Association. 1980. Diagnostic and Statistical Manual of Mental Disorders: DSM-III. Washington, DC: American Psychiatric Association.
American Psychiatric Association. 1987. Diagnostic and Statistical Manual of Mental Disorders: DSM-III-R. Washington, DC: American Psychiatric Association.
American Psychiatric Association. 2000. Diagnostic and Statistical Manual of Mental Disorders: DSM-IV-TR. Washington, DC: American Psychiatric Association.
American Psychiatric Association. 2010. ‘DSM-5 Proposed Diagnostic Criteria for Anorexia Nervosa’. Accessed 27 December.
American Psychiatric Association. 2013. Diagnostic and Statistical Manual of Mental Disorders: DSM-V. Washington, DC: American Psychiatric Publishing.
Attia, Evelyn, and Christina A. Roberto. 2009. ‘Should Amenorrhea Be a Diagnostic Criterion for Anorexia Nervosa?’ International Journal of Eating Disorders 42: 581–89.
Bemis, Kelly M. 1978. ‘Current Approaches to the Etiology and Treatment of Anorexia Nervosa’. Psychological Bulletin 85: 593–617.
Benjamin, Orly, and Hila Ha’Elyon. 2002. ‘Rewriting Fertilization: Trust, Pain, and Exit Points’. Women’s Studies International Forum 25: 667–78.
Berkovitch, Nitza. 1997. ‘Motherhood as a National Mission: The Construction of Womanhood in the Legal Discourse in Israel’. Women's Studies International Forum 20: 605–19.
Birenbaum-Carmeli, Dafna. 2004. ‘“Cheaper than a Newcomer”: On the Social Production of IVF Policy in Israel’. Sociology of Health and Illness 26: 897–924.
Bordo, Susan. 2003. Unbearable Weight: Feminism, Western Culture, and the Body. 2nd edition. London: University of California Press.
Bruch, Hilde. 1974. Eating Disorders: Obesity, Anorexia Nervosa, and the Person Within.London: Routledge and Kegan Press. 
Brumberg, Joan Jacobs. 2000. Fasting Girls: The History of Anorexia Nervosa.New York: Random.
Buckley, Thomas, and Alma Gottlieb, eds. 1988. Blood Magic: The Anthropology of Menstruation. Berkeley: University of California Press.
Cachelin, Fary M., and Brendan A. Maher. 1998. ‘Is Amenorrhea a Critical Criterion for Anorexia Nervosa?’ Journal of Psychosomatic Research 44: 435–40.
Claude-Pierre, Peggy. 1997. The Secret Language of Eating Disorders. Sydney: Bantam.
Copeland, Paul M., Natalie R. Sacks, and David B. Herzog. 1995. ‘Longitudinal Follow-Up of Amenorrhea in Eating Disorders’. Psychosomatic Medicine 57: 121–26. 
Eli, Karin. 2014. ‘Eating Disorders During Compulsory Service in the Israeli Defense Forces’. Military Behavioral Health 2: 89–97. 
Fairburn, Christopher G,. and Kristin Bohn. 2005. ‘Eating Disorder NOS (EDNOS): An Example of the Troublesome “Not Otherwise Specified” (NOS) Category in DSM-IV’. Behaviour Research and Therapy 43: 691–701.
Gendall, Kelly A., Peter R. Joyce, Frances A. Carter, Virginia V. McIntosh, Jennifer Jordan, and Cynthia M. Bulik. 2006. ‘The Psychobiology and Diagnostic Significance of Amenorrhea in Patients with Anorexia Nervosa’. Fertility and Sterility 85: 1531–35.
Golden, Neville H., and I. Ronald Shenker. 1994. ‘Amenorrhea in Anorexia Nervosa: Neuroendocrine Control of Hypothalamic Dysfunction’. International Journal of Eating Disorders 16: 53–60.
Good, Byron, and Mary-Jo DelVecchio Good. 1994. ‘In the Subjunctive Mode: Epilepsy Narratives in Turkey’. Social Science and Medicine 38: 835–42.
Gooldin, Sigal. 2002. ‘Anorexia in Israel or “Israeli Anorexia”? A Few Comments on “Culture-Bound Syndrome” in a Glocal Connection’. Israeli Sociology 4: 105–40.
Gottlieb, Alma. 1988. ‘American Premenstrual Syndrome: A Mute Voice’. Anthropology Today 4: 10–13.
Gull, William. 1874. ‘Anorexia Nervosa (Apepsia Hysterica, Anorexia Hysterica)’. Transactions of the Clinical Society of London 7: 22–28.
Kleinman, Arthur. 1978. ‘Concepts and a Model for the Comparison of Medical Systems as Cultural Systems’. Social Science and Medicine 12B: 85–93.
Lager, E. Grace, and Brian R. McGee. 2003. ‘Hiding the Anorectic: A Rhetorical Analysis of Popular Discourse Concerning Anorexia’. Women's Studies in Communication 26: 266–95.
Laségue, Charles. 1873a. ‘On Hysterical Anorexia’. Medical Times and Gazette,6 September, 265–66.
Laségue, Charles. 1873b. ‘On Hysterical Anorexia’. Medical Times and Gazette, 27September, 367–69. 
Lester, Rebecca J. 1995. ‘Embodied Voices: Women's Food Asceticism and the Negotiation of Identity’. Ethos 23: 187–222.
Lock, Margaret M. 1993. Encounters with Aging: Mythologies of Menopause in Japan and North America.London: University of California Press.
Malson, Helen, and Jane M. Ussher. 1996. ‘Bloody Women: A Discourse Analysis of Amenorrhea as a Symptom of Anorexia Nervosa’. Feminism and Psychology 6: 505–21.
Mantzoros, Christos, Jeffrey S. Flier, Michael D. Lesem, Timothy D. Brewerton, and David C. Jimerson. 1997. ‘Cerebrospinal Fluid Leptin in Anorexia Nervosa: Correlation with Nutritional Status and Potential Role in Resistance to Weight Gain’. Journal of Clinical Endocrinology and Metabolism 82: 1845–51.
Marcus, Marsha D., Tammy L. Loucks, and Sarah L. Berga. 2001. ‘Psychological Correlates of Functional Hypothalamic Amenorrhea’. Fertility and Sterility 76: 310–16.
Milos, Gabriella, Anja Spindler, Ulrich Schnyder, and Christopher G. Fairburn. 2005. ‘Instability of Eating Disorder Diagnoses: Prospective Study’. British Journal of Psychiatry 187: 573–78.
OECD. 2013. OECD Factbook 2013: Economic, Environmental and Social Statistics. Paris: OECD Publishing.
Pike, Kathleen M. 1998. ‘Long-Term Course of Anorexia Nervosa: Response, Relapse, Remission, and Recovery’. Clinical Psychology Review 18: 447–75.
Rabinow, Paul. 1992. ‘Studies in the Anthropology of Reason’. Anthropology Today 8: 7–10.
Rabinow, Paul, and Nikolas Rose. 2006. ‘Biopower Today’. BioSocieties 1: 195–217.
Remennick, Larissa. 2000. ‘Childless in the Land of Imperative Motherhood: Stigma and Coping among Infertile Israeli Women’. Sex Roles 43: 821–41.
Riessman, Catherine Kohler. 1990. ‘Strategic Uses of Narrative in the Presentation of Self and Illness: A Research Note’. Social Science and Medicine 30: 1195–1200.
Roberto, Christina A., Joanna Steinglass, Laurel E.S. Mayer, Evelyn Attia, and B. Timothy Walsh. 2008. ‘The Clinical Significance of Amenorrhea as a Diagnostic Criterion for Anorexia Nervosa’. International Journal of Eating Disorders 41: 559–63.
Skultans, Vieda. 1997. ‘Theorizing Latvian Lives: The Quest for Identity’. The Journal of the Royal Anthropological Institute 3: 761–80.
Sperling, Daniel. 2010. ‘Commanding the “Be Fruitful and Multiply” Directive: Reproductive Ethics, Law, and Policy in Israel’. Cambridge Quarterly of Healthcare Ethics 19: 363–71.
Steffen, Vibeke. 1997. ‘Life Stories and Shared Experience’. Social Science and Medicine 45: 99–111.
Strober, Michael, Roberta Freeman, and Wendy Morrell. 1997. ‘The Long-Term Course of Severe Anorexia Nervosa in Adolescents: Survival Analysis of Recovery, Relapse, and Outcome Predictors over 10–15 Years in a Prospective Study’. International Journal of Eating Disorders 22: 339–60.
Waldman, Ellen. 2006. ‘Cultural Priorities Revealed: The Development and Regulation of Assisted Reproduction in Israel and the United States’. Health Matrix 16: 65–106.
Warin, Megan. 2003. ‘Miasmatic Calories and Saturating Fats’. Culture, Medicine, and Psychiatry 27: 77–93.
Warin, Megan. 2005. ‘Transformations of Intimacy and Sociality in Anorexia: Bedrooms in Public Institutions’. Body and Society 11: 97–113.
Warin, Megan. 2010. Abject Relations: Everyday Worlds of Anorexia. London: Rutgers University Press.
Watson, T. L. and A. E. Andersen. 2003. ‘A Critical Examination of the Amenorrhea and Weight Criteria for Diagnosing Anorexia Nervosa’. Acta Psychiatrica Scandinavica 108: 175–82.
Young, Allan. 1995. The Harmony of Illusions: Inventing Post-Traumatic Stress Disorder. Princeton, NJ: Princeton University Press.