In what follows, it is the appearance of a woman who self-starves and the potentially
fatal outcome of her activities that are deployed in the construction of her
and her practices as deviant. In contrast to previous accounts, anorexia is produced
on the opposite side of the eating disorder binary, as more severe and more
pathological than bulimia.
Alison: I think anorexic women, they can kill themselves. None of us would, that have
these compensatory things are gonna – are trying to kill ourselves. We just
. . . yeah, we, we don’t . . . and we generally do feel quite good about ourselves.
I mean, when I’m, I’m slim, I really do feel quite good about my
figure and I don’t really want to lose anymore.
Pip: I think they’re [anorexia and bulimia] both severe – they both have enormous
implications. If you regard death as the ultimate then the anorexics [sic] the
one that is terrifying.
In these extracts Alison and Pip (a doctor) reported that anorexia is more serious
and more extreme than bulimia, due to it being potential suicide, which is the
gravest or ‘ultimate’ implication. Both portrayed anorexia as severe due to its
potentially fatal outcome; however, unlike Alison, Pip also constructed bulimia
as having very serious implications. In contrast, Alison represented her body
management practices as quite reasonable through her justificatory statements
about being in control, being able to stop losing weight, her ability to be happy
with her slim body and her (minimizing) reference to ‘compensatory things’. She
therefore distanced herself from women with anorexia who go too far and risk
death. Described in both accounts as not potentially fatal, bulimia was portrayed
as the less severe and less serious condition.
In the next extract it is the observable outcomes of starvation that contribute to
anorexia’s construction as worse than bulimia.
Becca: Anorexia . . . I think the real reason it’s considered – with – more with
distaste is because, you really, really do see the results of it, and it’s just so
horrifying to see these skinny, skeletal, you know, skeletal figures, and it’s,
it’s more like the shock value where – with anorexia, and I think, more, more,
more people would consider anorexia . . . as, as a worse . . . as a worse thing
In this account the ‘anorexic’ woman’s emaciated figure existed as a ‘shock[ing]’
and ‘distasteful’ sign of her disorder. In cultures where identities are read off the
surface of the body, one’s physical state is understood to represent both moral and
mental health. As Becca described, the shocking evidence of anorexia invites
judgements about the severity of a woman’s illness. Indeed, the emaciated figure
of a woman who starves can more readily be incorporated into notions of
psychopathology than the average and unremarkable body of the woman with
bulimia who just wants to be slim and beautiful.
In the last three excerpts, it is the potential consequences of anorexia (emaciation
and death), not the practices themselves (restriction), that are constructed as
deviant and pathological. This produces something of a paradox. Initially, as
demonstrated in the first section of this article, restriction is valued in terms of
control, success, and discipline. It seems that this valuing only occurs up to a
point however, beyond which such qualities are viewed as obsessive and extreme.
This shift is delineated by changes in the self-starving woman’s physical appearance,
which evokes associations with death. As Brown (1993) has pointed out
about anorexia, ‘[p]rior to the emaciation her behaviours and psychological
stance is likely to be encouraged and rewarded’ (p. 58). This differs markedly
from representations of bulimia where there is no such initial valuing of the
‘symptoms’ (Squire, 2003). The practices themselves (bingeing and purging) are
consistently rendered out of control and deviant. Unlike the body produced by
anorexia however, the ‘end product’ of bingeing/purging is a visibly normative
body, which escapes public scrutiny.
Having identified how accounts of anorexia and bulimia are organized according
to a dualistic logic, it is useful to examine the ways in which discourses of femininity
might be involved in these constructions. This analysis enables us to
uncover how binaries afford certain moral (albeit shifting) identities and work to
privilege certain body management practices and identities over others.
In the first five extracts, Bruch’s quote, and the examples from popular culture
and psychological literature, anorexia was positively associated with the mind
and with success, control, strength, and discipline, whereas bulimia was represented
as a bodily disorder linked to greed, promiscuity, capitulation, impulsivity,
and weakness. The portrayal of anorexia as the mind being in control of physical
urges fits with age-old Cartesian gendered conceptualizations of the self, not as
an embodied entity but as a mental entity or self – gendered male – housed in an
unpredictable body (Shildrick, 1997). Heywood (cited in Lupton, 2000) has
described the anorexic aesthetic in these terms as ‘privileging reason over emotion,
mind over body and because the feminine is associated with emotion and the
body, the masculine over the feminine’ (p. 215).
Indeed, when ‘femininity is coded as corporeality’ (Lupton, 1996: 110; Grosz,
1994) and as female bodies have historically been ‘associated with instinct, irrationality,
unpredictability, sensuality, uncleanliness, and evil’ (Hutchinson, 1994:
154), alluding to the physicality of bulimia and describing it as a greedy condi-
tion, as Rosie does in Extract 4, produces it as a gendered disorder. Bulimia
as ‘lett[ing] your appetite control you’ (Fran) therefore exemplifies entrenched
cultural anxieties about the unpredictable nature of female bodies, and the consuming
or excessive woman (Brooks et al., 1998; Cooper, 1992; Squire, 2003).
Becca’s statement in Extract 1, about bulimia being due to an inherent fallibility
and natural tendency to indulge fits within this framework of inherent female
Bulimia as a ‘greedy’ and uncontrolled disorder can be conceptualized negatively
as rejection of idealized notions of virtuous, self-denying, and moral femininity
characterized by generosity towards others (Lupton, 1996). For example,
in her discussion of the threat posed by eating disorders to the environment,
Riebel (2001) has described bulimic practices in ways that construct such women
as transgressing a femininity predicated upon selflessness, warning that during a
binge a woman with bulimia ‘might consume what could literally feed a whole
village’ (p. 38). Additionally within such a framework, bulimia can be understood
as conforming to pathologized notions of a negative type of femininity constructed
as voracious, animalistic, immoral, uncontained, and uncontrolled, as
Rosie’s account (Extract 4) exemplifies. Either way – conceptualized as a rejection
of a type of valued, selfless femininity or conformity to negative, labile, and
uncontrolled femininity – bulimia is firmly embedded on the derogated side of
the eating disorder binary. This contributes to its marginalization and characterization
as abnormal relative to women who self-starve and women without an
eating disorder diagnosis.
The notions of compulsory heterosexual femininity identified in the examples
of psychological discourse demonstrate how ideologies around acceptable
womanhood are intimately connected to dominant ideologies of consumption,
indulgence, sexuality, and weight.7 Clearly, a woman can be both ‘too sexual’ and
‘too hungry’, with each existing as a metaphor for the other (Bordo, 1993). This
is reflected by the psychologist in Extract 5, and in the fourth edition of the
Diagnostic and Statistical Manual of Mental Disorders (American Psychiatric
Association, 1994), where notions of the appetitive ‘bulimic’ woman are
deployed in ways that account for indulgence in both eating and sex. McKinnley
(1999) has questioned whose interests are served by agendas of normative feminine
(hetero)sexuality, pointing out how constructions of women’s unrestrained
sexuality as ‘dangerous and loathsome’ (p. 105) stem from the potential that an
uncontained female desire for its own sake has to undermine or ‘threaten male
virility’ (p. 105). Dichotomous constructions of sexual abstinence/passivity and
indulgence/promiscuity, embodied in psychological discourse around anorexia
and bulimia, mirror the age-old dualism of the Madonna (‘good girl’) and the
Within many of the examples it is clear that anorexia can be portrayed more
positively than bulimia and can be reflective of an admired identity. A cultural
privileging of control (especially for women when ‘femaleness’ is paradoxically
constructed as out-of-control) and popular definitions of mental wellbeing as
‘rational’ (Rose, 1996) operate powerfully to position bulimia and bulimic behaviours
on the derogated periphery of the eating disorder binary. Bulimia is expressive
of cultural concerns about losing control whereas anorexia (until the point of
emaciation which can also be characterized as out of control [Malson, 1998])
indicates mastery over one’s femaleness. It is the apparent loss of control
expressed physically, rather than pathology per se that invites judgements of
deviance. This makes sense in cultures where the ideal self is the site of discipline.
Lupton (2000) has argued that ‘the symbolic basis of anxiety is the threat
of disorder, the loss of control over our bodies and the negative cultural meanings
associated with this, rather than more manifest concerns about the threat of illness
or disease’ (p. 213).
Indeed, despite anorexia’s medical status as the more dangerous eating disorder
(Zipfel et al., 2000), its practices appear to exaggerate some positively
valued (and potentially contradictory) requirements of both historical and more
contemporary femininities. In addition to representations of anorexia as the
embodiment of a more traditional femininity characterized by abstinence,
restraint, and passivity, self-starving is also paradoxically privileged as a signifier
of those qualities that have historically been associated with ‘masculinity’, such
as self-control, persistence, transcendence of the (labile feminine) body, and
strength. It thereby brings into relief the way contemporary (western) ideals of
womanhood circulate as a somewhat contradictory mixture of both feminine and
masculine ‘qualities’. Importantly, however, although anorexic practices might
be (regularly) positively represented (either as reflective of ‘good girl’ femininity
or of positively valued, more traditionally, ‘masculine’ qualities), this is not
always the case.
The remaining extracts (6, 7 and 8) demonstrate how anorexic and bulimic
bodies can represent a multiplicity of socially meaningful positions. Anorexia
and food restriction cease to be privileged over bulimia and instead are constructed
as deviant and dangerous. This shift is not a simple reversal of the
binary, however, as this construction of anorexia occurs independently, or at least
not directly in terms of its relation to bulimia. Due to the problematic nature of
construing the (potentially) dying anorexic woman positively (see Malson and
Ussher, 1997), we see in Extracts 6, 7 and 8 that a woman who restricts no longer
exemplifies the characteristics of a desirable femininity. She becomes, as Becca
describes, a ‘horrifying’ skeleton. To engage in practices that potentially invite
death cannot be incorporated within dominant discourses of femininity characterized
by selflessness and care-giving. Furthermore, unlike a woman with
bulimia who is able to remain an object of the heterosexual male gaze, a woman
with anorexia undermines the ‘gendered performance criteria’ and becomes a
‘spectacle stripped of pleasure for the public spectator’ (Spitzack, 1993: 3). In
effect, her body dramatically exceeds the slender ideal. This results in its exclusion
from representation in terms of the various femininities signified by the kind
of slenderness that is culturally sanctioned as, for example, healthy (see Burns
and Gavey, in press), beautiful, successful, among others. A construction of
anorexia as deviant therefore relies on bearing witness to its physical outcomes
whereas representations of bulimia as deviant rest upon cultural ‘fears’ around
‘feminine’ impulsivity, greed, and indulgence.
Indeed, the ambiguity embodied by the anorexic woman is not matched in constructions
of bulimia, which is fairly consistently disparaged (as the extracts in
this study demonstrate). This, I believe provides the foundation which allows for
the hierarchical dichotomization of women’s over- and under-eating and which
overlaps across various sites, as I have shown using cultural examples, ‘eating
disordered’, psy, and expert accounts. In the latter two instances, as others have
also argued, by separating and listing personality characteristics and behaviours
in the diagnostic criteria sets for anorexia and bulimia (thereby constructing these
as deviant), psychiatry and psychology participate in the promulgation of
dichotomizing cultural assumptions regarding what is and is not acceptable
womanhood (e.g. McKinnley, 1998; Saukko, 2000).
In her analysis Saukko (2000) has suggested that narratives of anorexia and
bulimia often leave us with two ‘bad girls’, one who is ‘too frigid/rigid’ and the
other who is ‘too easygoing’ (p. 306). This assessment seems to be an oversimplification,
however. My research indicates that while women with bulimia
are repeatedly portrayed negatively, anorexia and its practices are characterized
in more complex and contradictory ways. These include representations that are
consistent with a type of valued contemporary femininity that includes control,
discipline, and strength. This conclusion extends Bordo’s (1993) influential
analysis that has suggested that anorexia is freakish given that, as the embodiment
of an abnormal capacity for self-denial, it represents ‘the incorrect management
of impulse and desire’ (p. 187). She describes bulimia, on the other hand, as a
predictable resolution of western culture’s ‘schizophrenic’ relationship with food,
where women are encouraged to indulge and are simultaneously chastised
for their consumption. This ‘take’ on anorexia as extreme, and on bulimia as
unremarkable and less extreme, is certainly exemplified in the final part of the
first section of my analysis.
I suggest, however, that an overemphasis on this particular framing of eating
disorders belies the complexities contained within constructions of women’s food
refusal and ‘over’-consumption/compensation. It also focusses upon the physical
bodies of women with eating disorders without sufficient attention to the meaningful
practices in which they engage. Concentrating (only) upon the emaciated
body of the anorexic woman renders any positive construction of her self-denial
and control untenable and therefore renders many of the ‘positive’ accounts in
this study ‘uninterpretable’. Similarly, focussing upon the (visibly) normal body
of a woman who practises bulimia enables attributions of her, and the disorder,
as unremarkable or even normative (as parts of Bordo’s  analysis suggest).
However, if we turn our attention to the practices of these so-called disorders,
then the ‘conditions’ and the women who practise them can be imagined quite
differently. In contexts in which individual self-control, discipline, and containment
are prized because they are considered simultaneously difficult and desir-
able, transcending the needs of the body for sustenance and applying rigorous
self-restraint are valued can be positively constructed. In contrast, capitulation to
the body’s appetites (especially when ‘bodyliness’ is aligned with a derided and
immoral femininity), reflected by overeating and emptying out again, is judged
to be weak, disgusting, and wasteful, and is therefore routinely reviled.
My analysis in this article suggests that this latter construction is one that
appears to permeate (although not in a totalizing fashion as evidenced by simultaneous
references to the pathology of self-starvation) accounts of bulimia and
anorexia, which inform the hierarchically organized binarized linguistic constructions
located in the accounts of women who practise eating disorder, health
experts, the psy literatures and in popular culture.
Implications for Subjectivity and Practice
I now consider the implications of these binaries for subjectivity and practice.
The following extracts demonstrate how women work with, and ‘take up’, positions
in relation to the dualistic logic I have identified and discussed in the first
and second sections. They demonstrate the potentially problematic effects of such
a categorization of eating disorders in which anorexic practices are portrayed
MB: You’ve alluded to this already but do you think that um . . . women who say have
anorexia are similar to or different to women who have bulimia?
Kay: Mm. Totally different . . . ends of the scale to me.
MB: So how would you explain that to yourself then, when you sort of feel like have,
have periods of doing both?
Kay: That I’m a very confused person, that’s one thing I would put it down to . . . um
(long pause) I think that . . . I’m taken to be a very strong person by a lot of
people . . . that aren’t right inside my life . . . I appear very strong-minded,
strong-willed. I go after what I want . . . but those people don’t really know what
I’m like inside (quietly) and they don’t know that I’m actually quite a weak person
. . . and as far as the way I eat . . . goes, I see that a little bit as . . . what I
am . . . and . . . what I try to be. It’s like two . . . sides of it. Like when I’m
bulimic, I’m very vulnerable and weak and . . . that’s probably the real me.
Being the other side of it where I’m not eating . . . it’s a wonderful . . . cover or
. . . face (long pause) It’s very strong. Mm. So it’s very conflicting and um, it’s
actually quite bizarre that it’s – can be – that I can be both ways ‘cos then I often
think too, when I’m like this, why aren’t I always like this, that I have control of
what I put in my mouth. Like I can’t understand that I can actually lose control
In this excerpt Kay deploys a familiar discourse that produces women with
bulimia as different from women who have anorexia. This dualistic construction
raises difficulties for her in terms of accounting for why she experiences episodes
of both anorexic and bulimic type behaviours. Rather than drawing into question
the legitimacy of the diagnostic labelling and popular understandings that tend
to pit bingeing against self-starving, Kay accounts for her ‘bizarre’ sequential
participation in these apparently oppositional behaviours in terms of her own
shortcomings and confusion.
Rather than considering the gendered social construction of attributions of
strength and control or fallibility and failure to various eating practices, Kay
reproduces dominant psychological theory and popular representations that posits
these behaviours as arising in part from personal characteristics originating within
individuals (Rose, 1996). Restriction is therefore discussed in terms of control
and strength and binge/purging in terms of weakness and vulnerability. The
results of this, combined with psychology’s assumptions about the stable and
internal nature of pathology, are, at least, threefold. First, eating disorders are
seen to originate within individual women, with a coinciding attribution of
responsibility that ‘acts to deny the social and discursive context of women’s
lives’ (Ussher, 2000: 210). Second, the binary sets up anorexia as the more
acceptable/‘desirable’ eating behaviour reflective of, and arising from, inner
strength and resolve, whereas bulimia represents the antithesis of the idealized
self-controlled subject. Finally, a focus on the self as responsible for eating
‘pathology’ results in Kay reporting the problem as originating within her. She
effectively becomes her diagnosis and her identity and behaviour become defined
by her ‘illness’. Because, as I have demonstrated in the first two sections of
analysis, being self-controlled and abstinent are viewed positively (especially
for women), and because Kay describes struggling (often unsuccessfully) to be
this way, she concludes that her real self is a bulimic self with all the negative
associations this conclusion confers. Kay says that ‘when I’m bulimic, I’m very
vulnerable and weak and . . . that’s probably the real me.’
A dichotomous construction of bulimia as weak and anorexia as strong informs
Kay’s account of who she is and forecloses a representation of herself as strong
minded and strong willed whenever she is in a bulimic ‘phase’ (see Dickerson
and Zimmerman, 1995). Consequently, Kay describes not only being miserable
about her behaviour, but also about the version of the self that this entails.
This representation of her real self as ineffectual is informed by psychiatry’s
(re)production of a liberal humanist idea of selfhood as rational, unified, fixed,
self-contained, and stable. Kay is prevented from representing herself as mutable,
plural, and contextual (e.g. Gergen 1991; Henriques et al., 1984) and therefore
capable of being, for example, weak, strong, helpless, and powerful at different
times and in different contexts, or even simultaneously.
Lyn: I don’t know how the atmosphere had been created but I know that it felt
right away that there was some kind of hierarchy between those labelled with
anorexia and those labelled with bulimia and that anorexia represented this more
kind of achievement of perfection and it was a cleaner disorder because you
weren’t throwing up and there were just all of these things that um made – made
that category. I mean that’s the ultimate achievement of anorexia is to kind of
have it perfectly / MB: mmm / um . . . and I was labelled with bulimia and right
away um the problem grew really really strong and difficult for me and decided
that it needed to be called anorexia and um I went rapidly down hill in all ways
and I I was in this supposedly very supportive environment but things took such
a turn for the worse that I was um you know an inpatient in a matter of months
in a couple of months / MB: right / and that was not being addressed at all in the
program I felt like I couldn’t speak about it at that time but it was there was some
kind of competition happening there that was really unhealthy.8
In this reflexive account, Lyn uses externalizing language9 to deconstruct her past
experience of eating disorders. She offers an explanation for the development of
anorexia that is an alternative to the legitimate frames of knowledge we would
find in the clinical literature, and which dominate the field in which Lyn works
and was herself treated as a (in)patient. Lyn’s report that there was an implicit
atmosphere of competition in the treatment facility, between the ‘achievement of
perfection’ and ‘throwing up’ deploys (in a critical way) the hierarchical binary
discussed earlier, in accounting for why she developed anorexia. Rather than
describing her move into the category of anorexic and her declining health in
terms that imply that her psychopathology worsened, the development of
anorexia is portrayed as a reaction to the implicit valuing of this category and the
corresponding derogation of the failure and uncleanliness represented by bulimia.
Her move into anorexia can be seen as becoming a ‘good girl’ in this treatment
Lyn’s use of a less conventional way of constructing the onset of anorexia is
useful in terms of theorizing the potential ‘extra-discursive’ implications of
privileging certain eating disorder behaviours over others. It offers insights that
are useful for understanding interactions in eating disorder treatment/support
settings where women with bulimia and women with anorexia both participate.
In order to protect against the unhealthy competition that Lyn describes, (and
which has been articulated by many women in this and other research10) it is
important in these settings to challenge the implicit hierarchy that privileges
anorexia above other types of eating distress. Anecdotal evidence and Extract 10
suggest that failing to take this into account can result in further isolation for
women with bulimia, a potential worsening of their binge eating and purging
behaviours and the development of self-starving.
CONCLUDING COMMENTS: DESTABILIZING THE BINARIES
In this article I have sought to highlight and problematize the ways that we have
become accustomed to thinking about the categories of anorexia and bulimia and
women identified as anorexic and bulimic. My goal was to demonstrate that constructions
of eating disorders located in the accounts of women with bulimia,
health professionals, psy literatures, and in other cultural examples are organized
according to a dualistic logic and are intimately connected to historically contingent
discourses of femininity. These dualistic discourses (re)produce patriarchal
dichotomies in which femininity is associated with corporeality, excess,
weakness and irrationality (Bordo, 1993; Malson, 1998). The associations
between bingeing and compensating, and this extremely negative construction of
female lability and danger, render the self-starving body as transcendent of this
derided femininity and produce a strong, controlled, contained, disciplined, and
successful subjectivity in opposition to that signified by bulimia and its practices.
While anorexia is ostensibly the more deadly condition, I have shown that
women who identify as bulimic, or who are labelled bulimic by the health professionals
they seek help from, are at risk of being constructed as the eating
disordered ‘other’ to women with anorexia and the practice of dietary restriction.
This has implications for treatment practices, the way in which we understand
the ‘condition’ and for the person who is the subject of this definition and the
practices arising from it.
In order to conclude and to raise the possibilities presented by different ways
of conceptualizing eating disorders, I have included an extract that troubles the
more common hierarchical and dualistic construction of them. Although such
expressions were rare in the interviews, Lyn consistently eschewed dominant
psycho-medical conceptualizations that characterized anorexia and bulimia as
discrete diagnostic categories. As a strategy this is interesting to consider as it has
the potential to disrupt the fictions that have built up around the categories of
anorexia and bulimia and the women who are identified in this way.
Lyn: I kind of think of them as interchangeable and I would / MB: right / just say that
all as one word if I could.
MB: Right OK right so
Lyn: I really don’t think of there be – as being any difference between the two / MB:
right/ I mean there are some different behaviours but I don’t / MB: mmm /
generally I would talk about it as like anorexia and bulimia as one word.
This representation of anorexia/bulimia, not as fixed, separate and stable but as
practices that can be engaged in simultaneously or sequentially, problematizes
diagnostic categorizations of eating disorders and blurs the boundaries between
them, reframing them as a continuum or as ‘interchangeable’. This conceptualization
is potentially deconstructive, as is the reported desire to refer to anorexia
and bulimia with one word. Referring to eating disorders with a generic term
disrupts the associations that have become crystallized around each disorder and
around the women who populate these categories. A more fluid and contextual
conceptualization of disordered eating in all its forms is one that is excessive to
the current binary I have identified in my analysis. It does not simply replace the
dualism identified in this article by its reversal, still maintaining but inverting
patriarchal dichotomies, but displaces this structure altogether. By exceeding
these dualistic oppositions, such a move troubles the very binary around which
eating disordered behaviours and identities are separated, and organized, in terms
of their ‘value’.
In terms of the dualistic constructions that have been identified in my analyses,
this tactic could potentially disrupt attributions of weakness, failure, greed, sexual
promiscuity, loss of control, and deviance that are often made to women who are
diagnosed or who identify as bulimic. It disrupts the ‘double’ pathologization or
‘othering’ that women who practise binge eating and purging arguably encounter.
Furthermore, a notion of fluidity between the eating disorders and a construction
of ‘over-’eating and ‘under-’eating as inseparable responses to a (western)
slenderness imperative (that are intimately tied to cultural assumptions around
appropriate womanhood), might also function to break down the dangerous
idealization of abstinence and restriction that have become exemplars of feminine
control and strength.
1. Bulimia is derived from boulimia, which literally translates from the Greek, meaning
2. Ninety-five percent of people diagnosed with eating disorder are women (American
Psychiatric Association, 2000).
3. The term Pakeha refers to people born in Aotearoa New Zealand of European descent.
4. Aotearoa is a Maori term for New Zealand.
5. Research in Aotearoa New Zealand indicates that, at 2 percent of the population, rates
of bulimia are similar to those reported in the UK and America (Bushnell et al., 1994).
6. Interestingly, for a woman to be sexually active (rather than sexually over- or underactive)
is construed as problematic.
7. Given their transgression of normative scripts of feminine restraint and heterosexuality,
it is possible that lesbian women who practise bulimia would be even more likely
to be subjected to discourses linking their ‘deviant’ appetites with disorder.
8. Although Lyn was recruited to this study as a health professional, here I have utilized
extracts in which she draws upon her own eating disorder experiences.
9. This involves using narrative to separate the ‘self’ from the experience of a problem
by objectifying the disorder rather than constructing it as an inherent or stable condition
(White and Epston, 1990; Zimmerman and Dickerson, 1994).
10. For critical analyses of this competition in inpatient facilities see Gremillion (2003)
and Segal (2002). In her ethnographic study of an eating disorder unit Segal (2002)
describes how ‘women needed to establish themselves as pure anorexics rather than
bulimics, who rank lower than anorexics in the eating disorder hierarchy’ (p. 7).