Introduction

The term anorexia nervosa was effectively invented in 1872 when it was first used to describe a disorder occurring chiefly in young women between sixteen and twenty-three years of age and manifesting itself in self-imposed starvation (Gull, 1896:305). Little has changed in terms of anorexia's definition in the medical literature; it is now defined in terms of a cluster of symptoms. Craighead & Nemeroff (2002:110), typical of this approach, describe it in terms of the refusal to maintain normal body weight, a fear of weight gain, a disturbance of body concept that often results in cessation of the menstrual cycle. The medical model goes on to explain that its causes are rooted in an interaction between psychological, biological and cultural factors.

There is much less emphasis, within the medical model, however, on what it means to be anorexic. Turner (1992:214) talks of anorexia being a talking disease, the idea that anorexia involves the self in the joining of the community of anorexia. It is the idea, largely an alternative to the medical model that sees anorexia as, at least partly, a linguistic construction. As such it cannot be understood without understanding how anorexics talk about their condition. There is no attempt here to denigrate the medical model or challenge its validity; it is more the attempt to add the required levels of meaning to anorexia. Established discourses on anorexia rely on research carried out that averages populations, making its comparisons, drawing its conclusions at a population level. Instead, the feminist approach has brought the individualism into the study of anorexia.

Personal accounts of anorexia such as that provided by Chernin (1981) in her book 'Womansize: The Tyranny of Slenderness', provide testimony of the experience of anorexia as well as placing the condition within its social and cultural context. Kim Chermin explains that the body contains wide realms of meaning, and nowhere are they more clear than in women's obsessions about food and weight. Through the body a woman may express how she feels about the culture she lives in and the desires and appetites of the body that she 'inhabits' - her body effectively becomes the physical manifestation of her emotional life (Chermin, 1981:1).

The quest for what it means to be anorexic is not only how that meaning is constructed but also where those meanings come from. An analysis of the dualistic philosophies starting with Plato, through Descartes and others has found a striking resonance with the experience of anorexia. These are examined along with their manifestations in consumer culture. Finally, these ideas are connected up with evidence coming from the empirical research examining sociocultural factors.

Plato and Descartes

Plato's 'Symposium' is identified by Heywood (1996:19) as one of the first philosophical treatments that resonates with anorexic experiences. Women, for Plato, are defined by their bodies, by 'material', while men aim to transcend this level and reach outside their bodies. This transcendence is achieved, for Plato, by love, where thought is seen to transcend love, mind transcends the body, and masculine transcend the feminine. Moreover, Plato saw the body as a perennial deceiver attempting to hoodwink us into mistaking permanence where there is none.

It is Descartes who is called on by Bordo (1993:144) to show the most obvious parallels with the anorexic experience. What is known as Cartesian dualism is, at its most fundamental, the idea that there is a separation between the world of material objects and the world of the mind (Gadrey, Baker & Morris (2002:1). The world of the mind is understood by a person through the process of introspection, the objects of these introspections being the internal 'states of consciousness'. Descartes believed that this power of introspection was infallible - that whatever was discovered there by this method must be true. The outer world, in contrast, is understood through the five senses: taste, touch, seeing, hearing and smelling. The senses, though, are fallible and cannot be trusted.

This way of thinking separated mind and body into two classes of existence. This provided a fundamental split between the ways in which mind and body were amenable to understanding and truth, a split between methods of knowing. The body, therefore, became subject to scientific examination, understanding and description, sometimes through the realm of mathematics. In contrast, the mind could be seen as a completely separate area which, in Descartes' analysis, involved all those things that were not bodily. Aside from definitional problems this sets up a most fundamental problem: how do the mind and body interact?

One of Descartes most important accomplishments was to bring the mind and the body closer together, rather than the reverse, as is commonly believed (Lowry, Shapiro & Walsh, 1982:8). He brought forward the idea that the actual connection between the mind and the body was mutual, that each had an effect on the other. Rather than the mind sitting outside the body, controlling it from beyond, untouchable, as it were, by the physical realities of daily struggles; Descartes introduced the idea that there was a fundamental connection. Apart from this, Lowry, Shapiro & Walsh (1982:7) explain, the body gained a certain sense of agency instead of merely being a vessel within which the mind travelled. Most importantly, though, for Lowry, Shapiro & Walsh (1982:40), Descartes introduced the idea that the vitality, the driving force of the body was not, in fact, located in the soul, as had previously been thought, but was located in the body. Descartes saw the difference between animals and humans as being a result of the connection between mind and body. The animal body was seen by Descartes as mindless, although it could function perfectly well on its own. It is the mind that brings rationalism, it is the mind that directs the body and exerts controls over it so that it can be made to function more effectively than it can in its animal state. It is the mind that interrupts in, what a behaviourist would understand, as an automatic link between the stimulus and response, and exerts its agency over the body. Without the mind, the body, therefore, becomes a robot, unable to break out of its endless cycles of eating and sleeping.

As Cottingham (1992:119) points out, there are a number of problems with Cartesian dualism, but despite this, one of the most important consequences of Cartesian thought was the idea of first-person privilege (Scruton, 1995:38). While a person is fundamentally uncertain about the things outside their body, indeed outside their mind, a person can be certain about the contents of their own mind. It is this component of accurate knowledge and control that will become more important as it is discussed in connection with anorexia nervosa. It should not pass without mentioning that dualistic notions of the world in philosophy have now been largely destroyed within modern theorising (Scruton, 1995:40) although they still hold considerable importance, especially in the discussion of anorexia.

Anorexic Dualism

The intersections between Cartesian dualism and anorexia nervosa are placed in their historical context by Heywood (1996) in her book, 'Dedication to Hunger'. Heywood (1996:19) introduces anorexia by drawing out some of the similarities between it and the Cartesian approach. The way in which an anorexic experiences the separation between mind and body is very similar to the way in which Descartes described the separation. In this analysis it is as though each is separate from the other, but also that there are aspects of the mind that are particularly privileged. Anorexics feel that they must achieve a distance and a control over their bodies, that is the very essence of what Descartes called for in order to understand the world in which we live. The senses, including bodily sensations like hunger, are to be rejected as inaccurate and untruthful ways of experiencing the world. For the anorexic, the body is ultimately deceitful, and the mind should dictate its actions; just as Descartes thought the body, without the mind, cut adrift and left to its own devices, was just as surely an animal.

In their body imagery, Bordo (1993:144) argues, anorexics show a number of commonalities with philosophical dualistic thought. There is a sense of alienation from their bodies described by anorexics. It is as though their bodies are not quite part of them, and in this separateness there are echoes of Descartes' ideas about the 'otherness' of the body, its distance from the mind and the difficulty of understanding the connection. The body is seen by anorexics as a confining force, something that holds them inside, prevents them being free. Many anorexics claim that they want to be free of their bodies; indeed some wish their bodies to be entirely absent. For this reason the body becomes the enemy, just as Descartes counselled that, in the attempt to think, the body with its base desires for food, was simply a distraction - a distraction to be fought against and subdued. Plato similarly associates the body with the baser elements of society's products such as wars and famine, products of its mindless, directionless desires.

Most fundamentally, however, Bordo (1993:145) maintains that the idea of control is central in both the anorexic experience and the dualist approach. Moving control, then, to the centre of experience, creates a dualism in the mind, an us and them mentality, with the mind and body pitted against each other in a fight, quite literally, to the death. Hunger is seen as the enemy, just as Descartes wanted to escape bodily sensations, so the anorexic wants to escape the feeling of hunger. It is the bodily sensation of hunger that haunts the anorexic and creates fear and obsession in the mind. There is a fear that once the eating begins, it will never stop, even when the anorexic person has had their fill of food and is not currently feeling hungry.

It is not just the body, though, that can be thought of dualistically in this context, it is also the idea of fat. According to Malson (1998:129) fat can have a dualistic construction as it is the fat that is the enemy and seen, by the anorexic, as that which must be eradicated. It is the fat, like the body, that must be purged to display the control which the anorexic has over it. Again, like the body, Malson (1998:129) finds that anorexics feel that the fat 'shouldn't be there', is somehow 'other' than themselves. Causing the fat to be lost, then, may constitute a kind of penance for the anorexic, a means in itself to purify the body, rather than a means to becoming more attractive. In this way, the body also represents the mind or soul and its appearance is a representation of the mind's purity - or otherwise. Malson (1998:130) finds this moral tone as a theme amongst her anorexic interviewees, they see thinness as equated with goodness and fat as evil. The only way to become a good person is to lose all their fat. Then, if the fat must be eradicated, then so must the body. Malson (1998:133) finds that the ideal body for some anorexics is no body at all. Not only must the body and its fat be reduced to the minimum, it must be completely removed from existence. For the anorexic it is possible that the thin body represents the non-body.

If self-starvation is the means, then thinness is the end, so what does thinness represent for the anorexic? Bordo (1993:148) explains achieving thinness in the anorexic as proving that she or he can exercise control over their own body, that the will of the mind can transcend - as Plato might also have seen it - the desires of the body. Anorexics use this control over their own bodies as a way of coping with the failure to control external aspects of their life. A typical anorexic, explains Bordo (1993:149) has parents who have controlled many aspects of their life for them, and so it is only in their body that they can exercise control. Bordo (1993:149) explains the typical way in which the affliction starts - simply as a casual attempt to diet. Having discovered that control can be exerted effectively and that the results are gratifying the dieting starts to become an obsession.

The idea of control is clearly important for both the anorexic and dualist viewpoints - there are striking similarities. Like both Plato and Descartes, the anorexic sees control of the body by the mind as paramount in achieving their ends - indeed, as Malson (1998:121) states, the Cartesian viewpoint considerably over-emphasises the importance of control. It is only through the privileged knowledge gained from the mind that the subjugation of the body can be carried out. To where, though, is the anorexic trying to escape through exercising this control? Malson (1998:122) argues that the anorexic is attempting to ally her or himself to the idea of a controlled life. Through displaying the thin body ideal, which is clearly visible to the outside world, a person displays their control over life. Control also means power, and it is here again that the anorexic discourse converges for Malson (1998:135). Here the idea of 'dieting' is seen as separate and qualitatively different from 'starving' by the anorexic. While dieting is more associated with beauty, it is 'starving' that is associated with notions of power. The fat body means that its owner is out of control and has no power, whereas the thin body means power, it means the anorexic has a stronger identity. To gain weight, then, is to lose identity.

The importance of food in the dualistic nature of anorexia is central (Malson, 1998:125). When the anorexic eats, it is argued, food becomes the object of the body's desire and so it is in the construction of food that importance lies. Again the dualistic attitude is seen towards food, as the body desires it and the mind rejects it. It is both loved and hated at the same time. Malson (1998:126) finds that anorexics, while admitting it is life-sustaining, still construct food as something that is actually life-threatening. This can be seen as a rejection of that to which everyone is dependent - especially as femininity is frequently associated with dependency.

The relationship with food is seen by Malson (1998:127) as providing a parallel with the daughter's relationship with the mother. Bruch (1973) is cited by Malson (1998:127) as finding that this relationship is often disturbed in anorexics. The way in which daughters construct their own identity is tightly bound-up with their relationships with their mother. The mother, as the original source of all food, and then the provider of food, has a vital role to play. Malson (1998:127) warns against blaming the mother entirely as the cause of anorexia, but acknowledges that she must be included at some point.

A consistent thread in the literature is the use of anorexia to transcend femininity (Malson, 1998:136; Bordo, 1993:36; Riley, 1988:6). Why is it that the anorexic wants to escape, to transcend? The construction of what it means to be female, especially with reference to the body is highly problematic in western culture. Indeed it is the woman's body that is most closely attached to what it means to be female, much more so than for a man. The female body is, for example, associated with menstruation and the inability to control this process amongst others. This uncontrollability can be seen in a variety of contexts: pre-menstrual stress and the menopause can be seen as medical constructions. Apart from these medical ideas, there is the obvious second-class treatment that women have received, and still receive in many areas of life. This poor treatment spans across educational and occupational opportunities and causes women to be excluded and marginalised from many parts of society.

Malson (1998:137) emphasises how the female body has become associated with the site of excess. The attractive heterosexual woman is a thin woman, but there are many ways in which the fat female form is seen as purely sexual and uncontrollable - the femme fatale, the nymphomaniac and the man-eater (Malson, 1998:138). The idea of the 'mother' as well takes on a variety of contradictory, almost dualistic, forms. The mother is associated again with her body and in Malson's (1998:140) interviews is found to resemble a cow and to be stupid, almost without a mind. Here then, the mother stands at one end of the dualistic spectrum, in front of the daughter, the epitome of all things bodily: fat, stupid and uncontrollable. It is precisely this (imaginary of course) construction of the mother that the daughter is trying to escape from by enacting the exact reverse process on her own body and so (in the imagination) reversing all the connotations contingent on the body shape. In this way, the daughter can reject the domestic role of the mother, along with all the negative connotations that go along with it.

Returning to transcendence, then, it is the Cartesian dualism that appears to provide an escape from the problems of embodiment, of motherhood, of all the cultural problems inherent in femininity (Malson, 1998:136). By splitting the body from the mind the anorexic exerts, and displays, control over the uncontrollable body, finds a way of reaching above the prevailing construction of women as weak. The anorexic can also be seen as a parody of the cultural requirement for thinness in women. By denying their bodies through starvation, the anorexic rejects all these abhorrent parts of femininity and creates for themselves a new identity that is constructed outside the boundaries and limits set by gender. In this new form, and amongst this dualistic discourse, the anorexic finds power - but this power is ultimately dependent on the subjugation of the body to the mind and so the anorexic is locked in an endless struggle with the body, with its uncontrollable urges and alien desires.

Hegel and Freud

The arguments about the similarities between dualistic philosophy and the experience of anorexia draw, to a large extent, on culture. This is not just the culture that has produced the stance of dualism but the all-encompassing cultural life that surrounds every one of us. What is this culture telling us about the sorts of people that it is 'good' or 'acceptable' to be? How does culture inform our behaviour and affect the ways in which we relate to others as individuals and as members of society? Some of the sociocultural factors have been touched on earlier in the discussion, but, in order to understand anorexia, it is necessary to take a closer look at culture and society to examine its effects.

The origins of anorexia in the general public consciousness are located in the 1970s as explained by Brumberg (1988:41-125). At this time there were concerns about feminist politics, dieting, fashion, displaying the self, excessive consumption and ultimately the rise of a new narcissistic attitude. Anorexia's emergence is often associated with this time and this place in western society. In the mind of the intellectual and philosopher, as seen through the eyes of Heywood (1996), the anorexic condition has long been discussed. As well as tracing back the anorexic state of mind to Plato and Descartes, Heywood (1996:17) sees important early cultural influences from Hegel and Freud.

Picking up from Plato's theory of love and how it can transcend, Hegel posits that it is, in fact, love itself that must be transcended in order for man to obtain a higher level of consciousness. For Hegel, unlike Plato, love is associated with the body, with the feminine and the family. It is these things that man, and only a man - must transcend to reach the ultimate goal, as Hegel (1977) saw it, of self-consciousness. This self-consciousness can, therefore, only be achieved by a man removing or disconnecting himself from the family. Hegel also delineates between the types of thinking of which male and female minds are capable. The female mind, being particularly interested in emotion and concrete thought, whilst the male mind deals with the abstract and ineffable (Hegel, 1942). The associations with being female go on and are typical of a patriarchal society: the man takes control, goes out into the world, makes the decisions, and decides how the community should operate. The woman stays at home, looks after the family, which in turn 'sends out' the man - the man representing a 'being' while the women has an explicitly unconscious spirit. With their place in the family, then, Hegel defines women as non-beings. For Hegel there is no sense of struggle between men and women. Women are simply relegated to a lower status of existence and their being is seen as the necessary sacrifice which must be made. Not only are women devalued in this sense, they are also used up, consumed and erased.

It is Freud, Heywood (1996:25) argues, that signals an important change in the way that the relationship between men and women is seen in the eyes of intellectuals. Heywood (1996:26) locates this change in Freud's explanation of how males and females cope with their respective Oedipal and castration complexes. The castration complex in the male arises after he has seen a female and, discovering that she has no visible genitals, assumes that they have been removed. This only happens, though, at a later stage of development as the male is deemed to ignore or reject this information in the first instance. This leads to the partial resolving of the Oedipal complex as the male begins to reject the mother and move away from the family. Here, then, is seen a similar movement to that which Hegel outlines - the male moving away from the family - and in this process the superego is formed. The female, meanwhile, accepts her apparent castration immediately and this leads to one of three possible results: a masculinity complex is formed, sexuality is rejected, or the father becomes the object of love. The last of these options represents the Oedipal complex in its female form. Whichever of these routes the female takes, Heywood (1996:27) argues, they will not be able to enter into the public world of civilisation. Each of these routes is described by Heywood (1996:27) as a dead end for the female, never stepping outside the confines of the body, never escaping into the civilisation.

Both Freud and Hegel, then, build up a number of unflattering associations about what it means to be female. These writers associate femaleness with a kind of absence in some senses - literally in the case of Freud where it is the lack of a penis that causes the castration complex. The only way for the anorexic, within this discourse, to take part in the apparently superior calling of being male is to renounce the body, to become detached from it and rise up. This is seen as a political message by Heywood (1996:28) and through the enactment of the necessary processes - anorexic processes - a woman reduces herself, paradoxically to nothing and loses her power and control.

Perhaps the analysis that places one of the strongest emphases on the dichotomy of gender is that provided by Chernin (1981). Chernin (1981:124) sees men as fundamentally afraid of the female form. Unlike Freud who thought that the female was primarily envious of the penis, within this analysis it is the male envy of the female form that is primary. What men have done through their fear of the female form is, literally in Chernin's analysis, to attempt to convince women to deny their own forms so that they are not forced to confront that which most upsets them. Parts of the growing pubescent female body that were once filled with mystery and allure, such as the rounded belly and the filling breasts, are now treated, because of men's censure, as objects of disgust and revulsion to be eradicated through fasting. Not all of the blame is placed at men's door, however, it is acknowledged that, in a sense, women have accepted this construction of their own body. The gender difference in what weight is considered acceptable, though, was obvious and striking when Chernin was writing in 1981, and it is still today, although slightly less so. A fat man means a powerful man, a man who can afford to eat, a man flaunting his power, probably a millionaire. A fat woman is an object of mischief and fun, of amazement that anyone could 'let themselves go' that badly. A fat woman is powerless. Again, the dichotomies of control played out on the same battlefield of weight. Chernin (1981:124) argues that men have effectively transferred their shame onto women so that they do not have to carry it themselves.

Consumer Culture

While cultural critics point to the lineage of thought that sets up the dualistic fallacy, they are also keen to examine how modern consumer culture intersects with the desire to be thin. Bordo (1993:165) explains that culture is written on the body, which is itself an extremely potent symbolic form. Part of the problem, however, for the sociocultural critique is that it is extremely difficult to measure the pressures that culture puts on a person in terms of which forces are acting on them. Both culture and society act in many different ways on a person. Despite this, feminists have been keen to exploit the ready-made cultural evidence that continually bombards us through, the written, visual and aural forms. Advertising, Hollywood films, novels and television: each of these and many more forms have cultural values embedded within them. To what extent, then, is thinness an important locus and how do advertisements traverse the dangerous territory of gender politics and secret obsession?

There is an interesting dichotomy in the way that advertisements place their spin on control, an important subject for anorexics. In the analysis of Bordo (1993:105), adverts aimed at women tend to emphasise how their products will help them control their lives while products aimed at men tend to emphasise control over other people. Some of the examples provided include adverts for mascara and deodorants. It seems, then, that advertisers are aware that, in some way, the lives of women are out of control: specifically through their obsession with eating. Women are tacitly acknowledged to be continually hungry, although perplexingly, continually slender in the adverts' imagery, externally in control of their bodies but internally disturbed. It is somehow a secret that women have strong appetites, while men's hearty appetites are celebrated in adverts. To display the dark obsession that women have with eating, it is, therefore, necessary to displace the female figure in an advert for that of a man - despite the fact that the target audience is clearly female.

It is precisely this reluctance to show women enjoying food that makes Bordo (1993:108) suspicious of how adverts construct female desire. A woman cannot be seen to be enjoying or abandoning herself to desire unless this desire is measured, is toned down, is not allowed to reach the levels of abandon which men are seen in when eating. Female desire for food becomes entangled, in this analysis, with sexual desire - these desires cannot therefore be shown in their most obvious light, they can only be exposed obliquely or through demure imagery. On the other hand, male sexual desire can openly be displayed but is easily dissociated from the rapturous enjoyment of food. It is as though, for a woman in these images, the hunger for food and sex are one and the same, a woman is uncontrollable and unable to differentiate the two while the man is 'allowed' to enjoy both and can tell the difference. This apparent cultural confusion of women between sexual desire and desire for food is seen in anorexics who attempt to deny both their body and their sexuality through self-starvation.

Heywood (1996:30-40) draws attention to adverts in some of their many forms. A Nike campaign draws out many of the cultural assumptions made about women in society. One Nike ad (Nike advertisement, 1992) asks, "Did you ever wish you were a boy?" and then goes on to cajole the (female) reader into admitting that they had wanted to be a boy. This taps in to a powerful idea that anorexics and dualistic philosophies both point out and experience. The male gender is culturally associated with achievement, with athletic success, with control and mastery. For a woman to identify with this is to attempt to co-opt some of these associations for herself. For Heywood (1996:31) the subtext of the advert is that the gender boundaries are easy to cross, should a woman desire, and what it conveniently ignores, all in the name of marketing consumer goods, is that people are socialised into particular gender roles which cannot be dematerialised by wearing Nike clothes. In effect, the advert tells its audience that gender is there to be manipulated and played with, but anyone living in the real world knows this is simply not the case.

What the Nike advert plugs in to and resonates with is the cultural idea that, contained within the gym, as well as in the control over the body, is the answer to transcending the barriers of gender. The advertising sucks the viewer into a world where it might actually by possible to transcend these barriers, for a woman to step across and find out what it is like to be a man - only, though, with the crutches in the form of clothes and other accoutrements provided by Nike.

The anorexic position is also seen by Heywood (1996:34) in the phenomenon of female body building. While this looks, to the casual observer, as though it might be the opposite of anorexia, it does in fact embody, in a very real sense, many of the issues discussed here. Rather than reducing and ultimately causing the body to vanish, the female body builder is concerned with increasing the body's size, with attempting to subvert the idea of women as smaller, gentler and less powerful. Heywood (1996:35), in her analysis, takes the case of an article in Flex magazine (Flex, 1994:141). The title of the article, 'Blonde Ambition,' attempts to subvert the idea that women, as blonde, are seen as sexual objects and do have ambitions. The pictorial for the article shows a body-builder dressed in a variety of standard pornographic-type undergarments. In the text the pictured body-builder makes it clear that she wants to assert her sexuality as a muscular woman and she is proud of her achievement. Unfortunately, the very poses which she adopts and the clothes she is wearing point to the fact she is being defined by the magazine in exactly the same way women traditionally are - as sexual objects.

Ideas of strength covering weakness become, for Heywood (1996:38) veils which cannot cover the attempt to transcend, the same attempt to move into the cultural space occupied by men, to attempt to co-opt and own it, ultimately an obsession with the body, and a sexualised female body at that. Within the text the body-builder's references to herself and her body are still made to how attractive she is. Attractive to whom?

Modes of Investigation

The dominant discourse for the analysis of anorexia nervosa is that provided by medicine. Within this discourse the importance of biological, psychological and social factors are all important. While there is some emphasis in most of these types of discourses on the social and cultural factors, it tends to pale in comparison to the focus on having a disease. Bordo (1993:49) points out that anorexia is treated in much of the literature as something to be explained by categorisation, and the effect of culture is mostly sidelined - a large degree of the causation is assumed to flow from, for example, biological factors or from personality traits. Bordo (1993:49) argues that, for the anorexic, the cultural analysis provides the one unifying factor which cuts across a 'multidimensional' disorder and pulls the factors together. The incidence of anorexia is very much situated within its cultural and, particularly, within its historical context. The strong cultural analysis provides a way to stop the sociocultural factors disappearing from view underneath all the other factors that have been posited.

Like many pathologies that are deemed multi-factorial, the fight is on between researchers to stake their claim for the most important dimension. Bordo (1993:50) argues that the feminist critique of the contributory factors for anorexia has begun to emphasise the importance of the cultural critique - how anorexia is primarily learned rather than having its biological origins activated. The barriers are also being broken down between what is considered normal and what is considered pathological. Researchers have begun to show that the extreme methods which anorexics take in the control of their body is an extreme extension of how most women feel about their bodies. It might be argued that the anorexic is 'normal', only more so.

It is this breakdown of the line between what is normal and what is pathological that has started to create more space for the sociocultural factors. The importance of sociocultural factors can most strongly be established if there is evidence of many sub-clinical forms of anorexia. If only a few extreme cases of a disease exist, in a digital, on or off fashion, then it may be difficult to implicate sociocultural factors as primarily responsible. But, when a more analogue sweep between what is considered pathological and normal is seen with many shades of grey in between, the sociocultural critique becomes much more convincing. We are all steeped in society and culture so we should all be affected to some degree.

It is precisely these sorts of half-way houses that have been found in between anorexia and 'normality'. Bruch (1973:89) identified Body Image Distortion Syndrome (BIDS) in which a person is unable to assess the size of their own body objectively. This is directly related to eating disorders as anorexics will often complain that, despite being extremely thin, that they are actually too fat. What began to emerge from the research was that there were elements of BIDS amongst what were considered 'normal' people. Dolan, Birtchnell & Lacey (1987:513-520), for example, examined 100 non-eating disordered men and women and found that many of them consistently over-estimated their own body widths in comparison to their absolutely accurate perception of a neutral object. The majority of both men and women wanted to weigh less: 52% of men and 84% of women. Despite being of approximately average weight, the women, on average, wanted to be 3.25kg lighter while the men wanted to be 0.75kg heavier.

Bordo (1993:55) explains that the standard way in which this disparity between actual and desired, or even perceived, size was explained in the medical model was as a distortion of some kind. In other words, according to this model, women perceive themselves accurately but then are distorting this perception as a result of being self-critical. What studies showed, however, was that women feel bad about their bodies. Cash & Henry (1995:19-28) carried out a survey of 803 women in the US and found that around a half of women were dissatisfied with the major aspects of their bodies. The authors report that these results showed more dissatisfaction than had been recorded a decade earlier. What the sociocultural analysis does is rescue women from this label of strangeness, pathology and bizarre behaviour, and try to discover what it means, culturally, to be female. Instead, as Bordo (1993:57) sees it, of women being labelled as 'faulty thinkers', as though something were wrong with them, it shifts the emphasis onto the prevailing social attitudes towards our bodies and how we appear to others.

In the last decade there has been a large amount of empirical work done that attempts to draw out the connection between how sociocultural factors provide a catalyst for anorexia - often concentrating on body perceptions. Field, Carmago, Taylor, Berkey & Colditz (1999:1184-1189) looked at the effects of adolescent girl's peer groups and the role of the media. The study involved 6,982 girls between the ages of 9 and 14 who were initially free of purging behaviours. They were followed up a year later and the results showed that the effects of both their peer group and popular culture - in the form of magazines, TV and movies - were important. A variety of studies have tested the effects of short-term exposure to thin body ideals (Groesz, Levine & Murnen, 2002:1-16) but its effects have also been examined over a more extended time period. Stice, Spangler & Agras (2001:270-288), while finding difficulty in producing a longitudinal study with an experimental manipulation, report interesting findings. Half of 219 adolescent girls were given a subscription to a fashion magazine for 15 months, while the other half were not. Some evidence was found that those girls who were already vulnerable had an adverse reaction to the fashion magazines. It is, however, clear from this experiment - no main effects were found - how difficult it is to control media exposure such that a controlled experiment can be carried out. This kind of research highlights the problems that experimentally orientated researchers have found in collecting data on the sociocultural catalysts of anorexia.

One of the standard arguments that have been levelled against the cultural explanation of anorexia and other eating disorders is that, if we are all surrounded by these cultural influences then why don't we all develop eating disorders? This argument, while seeming strong to many critics of the cultural analysis, is obviously flawed. The criticism assumes that culture is like some kind of constant temperature and pressure controlled atmosphere that surrounds us at all times. Whereas, of course, culture is something that ebbs and flows, that interacts in different ways with us depending on many aspects of our lives and identities. So, not only is culture different across time and space, but our reaction and interaction to culture and society is fundamentally affected by our backgrounds, memories and attitudes. Despite these rebuffs, researchers have continued to look for more solid evidence of the effect of sociocultural factors and attention has turned to sub-cultural, intra-cultural and cross-cultural studies.

Anorexia Across Cultures

A major component in the argument for the sociocultural effects on anorexia is its supposed increase in prevalence over time. If we are now, with the increasing emphasis on the thin ideal in the media in the last three decades, much more obsessed with the idea of thinness then this should be reflected in an increase in the number of cases of anorexia reported. Nasser (1997:4), reviewing the prevalence studies finds that indeed there has been an increase. Szmukler, McCance, McCrone & Hunter (1984:49-58), for example, looked at the period between 1965 and 1982 to examine the number of people on the Aberdeen Psychiatric Case Register with anorexia. In this study a significant increase in the number of diagnoses of anorexia nervosa could be seen over the seventeen years. Similar findings were obtained by Jones, Fox, Babigan & Hutton (1980:551-558) who found there was increase of between 80 and 100 per cent in both the USA and Scotland over the period of twenty years. Nasser (1997:8) points out that it has been argued that this increase in apparent prevalence is the result of a greater effort to understand anorexia. This idea has not been endorsed, however, by more recent meta-analyses of the prevalence data on eating disorders (Hoek, 1993:61-74).

There has been an increasing level of discussion and research into eating disorders amongst athletes. As a sub-cultural group that often places greater emphasis on the body and, in some cases, the importance of thinness, if there is a higher prevalence here then the sociocultural factors can be clearly seen to be acting. Sundgot-Borgen (1993:29-40) examined the prevalence of eating disorders amongst 522 female Norwegian athletes. The study compared this group to 448 members of a non-athletic control group. The results showed that, while the actual incidence of eating disorders amongst the controls was 5%, the rate amongst the athletes was 18%. There were a particularly high number of athletes with eating disorders amongst those who competed in sports where leanness or weight was important. Not only were the levels higher amongst athletes, it was found by carrying out additional random clinical interviews that, in the screening, the athletes were probably under-reporting their eating disorders.

One of the first areas of study, however, was amongst ballerinas where it was hypothesised that the sociocultural pressure to remain thin was much higher than the general population. Druss & Silverman (1979:115-121) researched ballerinas and found that they felt overweight despite the fact they were clearly not, and that the fight to control their weight was related to a repudiation of their sexuality. Similarly, Garner, Garfinkel, Schwartz & Thompson (1980:483-491) found that, amongst ballerinas, as many as 30% showed some degree of anorexia with 7% displaying the full symptoms. This seemed to be a result of the social and cultural environment into which they had been placed as all of the ballerinas developed the condition after beginning their training.

An idea emerging in the literature is the importance of urbanisation. There have been some studies that have shown a connection between urbanisation and eating disorders. Nasser & Katzman (1999:34) theorise that cultural changes have affected mealtimes such that the meal is no longer a family occasion - this change having affected city-dwellers disproportionately. As the mealtime is often central to the structure and interaction of a family, when this changes it can cause a breakdown in family relations. Hoek, Bartelds, Bosveld, van der Graaf, Limpens, Maiwald & Spaaij (1995:1272-1285) found evidence of this in their study in the Netherlands. In this study, detected rates of bulimia nervosa in urbanised areas were more than six times higher than those in rural areas. The same result, however, was not found for anorexia.

Cross-cultural studies have looked at the varying prevalence of anorexia in different cultures. The established, and now ageing, formulation for cross-cultural analysis is that anorexia is a culture-bound affliction. It is something about modern western societies that causes (mostly) women to starve themselves. Indeed, Lee, Lee, Pathy & Chan (2005:275-281) compare the prevalence of anorexia in both western and Chinese and Japanese societies. In western societies the prevalence has been estimated as between 0.2% and 0.9%, while in Japan the rates have been estimated as between 0.025% and 0.03% and in China rates are estimated at 0.01%. At the closest point these figures are still a factor of ten apart, at the furthest the difference is a multiple of thirty-six.

Early findings from cross-cultural studies helped to contribute towards theories that situated eating disorders within ethnic bounds. Nasser (1997:11) finds that eating disorders are the best example, amongst psychiatric disorders, of the 'culture-bound disorder'. This has lead to a number of different diagnostic formulations of how they have arisen. Many theories have as their fundamental basis the early findings of eating disorders predominantly occurring in western societies. Gordon (1990), for example, bases the analysis of eating disorders within the following diagnostic terms. The frequent occurrence of eating disorders within a culture, compared to other psychiatric disorders; ideas that are pervasive within a particular culture become expressed as the core concepts of the condition; and personal problems and personal distress are expressed through the condition and represent an exaggeration of many aspects of the culture within which it arises. This view, however, has started to be challenged by more recent cross-cultural findings.

Nasser (1997:19) outlines some of the reasons why non-western societies were thought to have a much lower prevalence of eating disorders. Non-western societies are thought not to place the same emphasis on thinness as a cultural ideal, indeed are sometimes thought to view the opposite as desirable. Anorexia is thought to be a condition associated with affluence and a breakdown in gender roles - neither of which are common in non-western countries. A number of studies, reviewed by Nasser (1997:20-34) were carried out to assess eating disorders in a variety of ethnic minorities. What these studies found was that there was evidence that the risk factors for eating disorders could permeate across ethnicities when people moved from non-western to western societies. Furnham & Alibhai (1983:829-837), for example, compared British, Kenyan Asian and Kenyan British females for how they evaluated body shapes. The results showed that the Kenyans found the larger body shapes more acceptable than both the Kenyan British and the British females. These results clearly show the importance of sociocultural factors in anorexia, the assumption being that mere exposure to British attitudes and culture had changed the attributions of the Kenyan British. Similar results have been achieved in studies of other ethnicities, e.g. Lee, Hsu & Wing (1992:545-551) examined those from Hong Kong, while Silber (1986:121-128) carried out research in the US looking at Blacks and Hispanics. Comparable evidence was found.

These studies showed the importance of sociocultural factors in the development of eating disorders and lead to the idea that as non-western societies become more westernised in nature they will show a greater incidence of eating disorders, including anorexia nervosa. Nasser (1997:20-34) reviews the data that has accumulated from countries around the world. Apter, Shah, Iancu, Abramovitch, Weizman & Tanyo (1994:83-99) compared the attitudes towards eating of a group Israeli Jewish high school girls, a group of Arab high school girls living in Israel with a group of diagnosed anorexics. This study found that the girls who had the most similar attitudes to eating as the anorexics were those who had the most exposure to western culture and practices. Some of the highest results were obtained for the Arab girls - this was attributed to their greater affinity to western culture. Similar results were also found in other countries. Nasser (1994:25-30) screened 351 secondary school girls in Egypt for their attitudes towards eating. The rates of unusual attitudes were found to be similar to those obtained from a sample in the UK. This study claimed this was evidence that attitudes towards food were changing and the first signs of the development of eating disorders were being seen, despite the fact there had been little or no report of them in the past. Research from the Middle East, then, shows some support for the idea that as western cultural values seep into non-western societies, the prevalence of eating disorders rises.

Results of this kind have been obtained from many different countries around the world, although not always providing clear cut results. Studies amongst Japanese students, however, tended to show an increase. Suematsu, Ishikawa, Kuboki & Ito (1985:96-103) compared rates in 1976 to 1981 and found a doubling of the rates of anorexia over that period. Evidence is beginning to emerge from China (Lee & Lee, 2000:275-281) as well as Nigeria (Oyewumi and Kazarian, 1992:663-666) and Brazil (Nunes, Bagatini, Salvador, (1991). For a full review of these studies, including differences now being observed across Europe, see Nasser (1997:29-34). A note of caution should be attached to many of these studies, a point that Nasser (1997:35) makes. Much of the work that has been done has used the Eating Attitudes Test questionnaire and, to use it in different cultures with different languages has not been comprehensively shown to be valid. So far, in the main, straight translations of the questionnaire have been carried out without testing for cross-cultural validity. The questionnaire used was originally validated using Canadians - it is possible that for Chinese people, the questionnaire is not measuring the same constructs.

While this type of research is at an early stage and some of the methodologies have been questioned, it is fair to say that a uniform picture is emerging of the increase in eating disorders in non-western societies. While it was once thought that eating disorders were only to be found in western societies, this has now been shown to be untrue. Nasser (1997:36) considers some of the reasons that have been associated with this change. The role of the family has been thought to be important in the development of eating disorders. Family structures have been seen to move towards more western models, especially in the abandonment of the extended family structure in favour of the nuclear family. Eating disorders are often said to be a condition of wealthy societies and in the past it is western societies that have had the most money. But it is not only having money, it is the modernisation that is associated with greater wealth that is also thought to be important. Nasser (1997:41) draws attention to the changes in diet that have been brought about in western societies by changes in the way that we eat. This is exemplified by the McDonalds Corporation and the spreading US obsession with fast food. As westernisation spreads, these types of eating patterns are spreading and displacing indigenous food habits. Western eating habits tend to lead to obesity, which, has been shown to have an association with anorexia. It is in those countries that have the highest rates of obesity - the US - that the highest rates of anorexia are also seen. Bruch (1966:555-566) acknowledged that when life becomes stressful, many people turn to food as a form of comfort. Unfortunately the worship of the slender form, especially amongst women, leads to extreme body dissatisfaction.

Last, and by no means least, Nasser (1997:44) points out the importance of the global media. As western media have proliferated around the world so cultures have become more exposed to western weight norms (or rather ideals) of weight. Lewnes (1991:106-110), for example, found a relationship between American soap operas shown on Egyptian television and an increase in dissatisfaction over body weight. This kind of relationship between media representations and body weight dissatisfaction has been investigated in a large number of studies. Groesz, Levine & Murnen (2002:1-16) reviewed 25 studies that looked at experimental manipulations of the effect of media images on body image. In most of these studies it was found that, after viewing images of thin models, their own body image was significantly more negative. The authors of this meta-analytic review find that studies support that idea that there is a connection between media representation of thinness and thoughts and behaviours that are influential in eating disorders.

Other researchers on eating disorders are less happy to lay the blame for eating disorders entirely at the door of western culture as though it was pathogenic in and of itself. They argue that, while sociocultural factors are extremely important in anorexia, for example, to blame this entirely on western ideals is simplistic. Katzman & Lee (1997:387-88) call for a wider analysis of the causes of eating disorders outside the crude category of western culture. They provide three criticisms of focusing on the idea of western culture. Firstly, there is an assumption that being thin is one of the values that is taken in by a non-western culture. There has been little examination of how so-called western and non-western values are integrated with reference to the slender ideal. Secondly, assumptions are often made that geographic boundaries are representative of cultural boundaries. This problem is echoed by Nasser (1997:75-81) in her analysis of what it is that makes up 'a culture'. Thirdly, there is little examination of the meaning of what self-starvation means in both western and non-western cultures. In other words, assumptions have been made that those ideas related to body image are equivalent in different cultures and therefore related in the same way to eating disorders.

Katzman & Lee (1997:387-389) are, instead, keen to develop an analysis that is less ethnocentric. They draw on the analysis of Steiger (1993:347-357) who is critical of studies that see anorexia in terms of beauty ideas. Instead, greater emphasis should be placed on the importance of refusing food. To a certain extent, the analyses presented above represent a distancing from the importance of gender. It is gender, however, that has been central to the feminist perspective of anorexia, especially within dualist ideas. It is more universalist ideas that Steiger (1993:347-357) focuses on, such as the global importance of control and self-definition. It is here in the trans-cultural literature, then, that a call for a return is seen to the kind of feminist ideas with which this essay began. Issues of control and gender which sit outside a simplistic notion of the fear of fatness. Perhaps it is better not to think of anorexia as a condition that is culturally bounded but better to see it as a result of changes within cultures, with each culture experiencing it differently within its own social and historical context.

Conclusion

This analysis of the explanations of anorexia began with philosophical perspectives and, in particular, dualist notions. In this construction, common anorexic experiences of attempting to maintain control over the body were given meaning by their parallel with Cartesian ideas about the difference between males and females. It was seen that, in some sense, the control exerted over the body was an attempt to transcend gender boundaries to feel, and exert power. These Cartesian notions have a long history within the intellectual tradition running from Plato, through Descartes, Hegel and Freud. This analysis provides an extremely meaningful fit with the experience of many anorexics. As Bordo (1993:147) points out, while anorexics are not specifically taking up a metaphysical position through their illness, as clearly it is debilitating, there are striking parallels with Cartesian dualism to the extent that these ideas help to inform its analysis. It was then seen how the 'tyranny of slenderness' has permeated the culture such that many forms of the media are saturated with it.

The feminist critique of anorexia prefigured much of the empirical research that has since been carried out to examine how sociocultural factors affect the anorexic experience. Powerful arguments for the catalytic effect of sociocultural factors were shown in the intra-cultural, sub-cultural and cross-cultural research. Criticism has been made of the pathogenic view of western cultures, but there is evidence that anorexia is on the rise around the world and this is a result of changing sociocultural environments. While feminist analyses have concentrated on the personal level and the meanings of anorexia, those operating within a more medical model concentrate on population level analysis: prevalence and aetiologies. Each of these analyses, though, shows the power of the sociocultural analysis and its importance as a catalytic factor in anorexia nervosa. References

Apter, A., Shah, M., Iancu, I., Abramovitch, H., Weizman, A., Tanyo, S. (1994) Cultural effects on eating attitudes in Israeli subpopulations and hospitalised anorectics, Genetic, Social, and General Psychology. Monographs, 120(1), 83-99.

Bordo, S. (1993). Unbearable Weight. Los Angeles: University of California Press.

Bruch, H. (1966) Anorexia nervosa and its differential diagnosis, Journal of Nervous and Mental Disease, 141(5), 555-566.

Bruch, H. (1973) Eating Disorders, New York, Basic Books.

Brumberg, J. J. (1988) Fasting Girls, the Emergence of Anorexia Nervosa as a Modern Disease. Cambridge, Mass: Harvard University Press.

Lewnes, A. (1991) Fat chance: Cairo fitness centres surveyed, Cairo Today, 31(1), 106- 110.

Cash, T. F., Henry, P. E. (1995) Women's body images: The results of a national survey in the USA. Sex Roles, 33, 1-2, 19-28.

Chernin, K. (1981) Womansize: The Tyranny of Slenderness. London: The Women's Press.

Cottingham, J. (1992) Descartes. London: Blackwell.

Craighead, W. E., Nemeroff, C. B. (2002) The Corsini Encyclopedia of Psychology and Behavioral Science. Chichester: Wiley.

Dolan, B. M., Birtchnell, S. A., Lacey, J. H. (1987) Body image distortion in non-eating disordered women and men. Journal of Psychosomatic Research, 31(4), 513-20.

Druss, R., Silverman, J. (1979) Body image and perfectionism of ballerinas: comparison and contrast with anorexia nervosa, General Hospital Psychiatry, 115-121.

Field, A. E., Carmago, C. A., Jr., Taylor, C. B., Berkey, C. S., Colditz, G. A. (1999). Relation of peer and media influences to the development of purging behaviors among preadolescent and adolescent girls. Archives of Pediatric Adolescent Medicine, 153, 1184-1189.

Flex. (1994) Blonde Ambition. April, p. 141.

Furnham, A.C., Alibhai, N. (1983) Cross cultural differences in the perception of female body shapes, Psychological Medicine, 13, 829-837.

Gadrey, J. J., Baker, G. P., Morris, K. J. (2002) Descartes' Dualism, London: Routledge.

Garner, D.M., Garfinkel, P.E., Schwartz, D., Thompson, M. (1980) Cultural expectations of thinness in women, Psychological Reports, 47, 483-491.

Gordon, R. (1990) Anorexia and Bulimia: Anatomy of a Social Epidemic. Cambridge, Mass.: Basil Blackwell.

Groesz, L. M., Levine, M. P., Murnen, S. K. (2002) The effect of experimental presentation of thin media images on body satisfaction: a meta-analytic review. International Journal of Eating Disorders, 31(1), 1-16.

Gull, W. W. (1896) Anorexia Nervosa (Apepsia Hysterica, Anorexia Hysterica). In: T. D. Acland (ed.). A Collection of the Published Writings of William Withey Gull. London: New Sydenham Society.

Hegel, G. W. F. (1942) The Philosophy of Right, trans. T. M. Knox. Oxford: Oxford University Press

Hegel, G. W. F. (1977) The Phenomenology of Spirit, trans. A. V. Miller. Oxford: Oxford University Press.

Heywood, L. (1996) Dedication to Hunger: The Anorexic Aesthetic in Modern Culture. Berkeley: University of California.

Hoek, H. (1993) Review of the epidemiological studies of eating disorders, International Review of Psychiatry, 5, 61-74.

Hoek, H., Bartelds, A., Bosveld, J., van der Graaf, Y., Limpens, V., Maiwald, M., Spaaij, M. (1995) Impact of urbanisation on detection rates of eating disorders, American Journal of Psychiatry, 152(9), 1272-1285.

Jones, D.J., Fox, M.M., Babigan, H.M., Hutton, H.E. (1980) Epidemiology of anorexia nervosa in Monroe County, New York: 1960-1967, Psychosomatic Medicine, 42, 551-558.

Katzman, M. A., Lee, S. (1997) Beyond body image: the integration of feminist and transcultural theories in the understanding of self starvation. International Journal of Eating Disorders, 22(4), 385-94.

Lee, S., Hsu, G., Wing, Y. (1992) Bulimia nervosa in Hong Kong Chinese patients, British Journal of Psychiatry, 161, 545-551.

Lee, S., Lee, A. M. (2000) Disordered eating in three communities of China: A comparative study of female high school students in Hong Kong, Shenzhen, and rural Hunan. International Journal of Eating Disorders, 27(3) , 317 - 327.

Lee, H. Y., Lee, E. L., Pathy, P., Chan, Y. H. (2005) Anorexia nervosa in Singapore: an eight-year retrospective. study. Singapore Medical Journal, 46, 275-281.

Lowry, R., Shapiro, D. H., Walsh, R. N. (1982) The Evolution of Psychological Theory: A Critical History of Concepts and Presuppositions. Chicago: Aldine.

Malson, H. (1998) The thin woman: feminism, post-structuralism and the social psychology of anorexia nervosa. NY: Routledge.

Nasser, M. (1994) Screening for abnormal eating attitudes in a population of Egyptian secondary school girls, Social Psychiatry and Psychiatric Epidemiology, 29(1), 25-30.

Nasser, M. (1997) Culture and Weight Consciousness. New York: Routledge.

Nasser, M., Katzman, M. (1999) Eating Disorders: Transcultural Perspectives Inform Prevention. In: N. Piran, C. Steiner-Adair, M. P. Levine (eds.), Preventing Eating Disorders: A Handbook of Interventions and Special Challenges. London: Brunner/Mazel.

Nike advertisement (1992) Vogue, June 1992.

Nunes, M., Bagatini, L., Salvador, C. (1991) What to think of anorexia nervosa in Brazil, a country of hunger and undernourishment? Poster presentation at the International Symposium on Eating Disorders, Paris, April.

Oyewumi, L., Kazarian, S. (1992) Abnormal eating attitudes among a group of Nigerian Youth: 1, Bulimic behaviour, East African Medical Journal, 69(12), 663-666.

Riley, D. (1988) Am I that Name? Feminism and the category of 'women' in history. Macmillan, Basingstoke.

Scruton, R. (1995) Short History of Modern Philosophy: From Descartes to Wittgenstein. London: Routledge.

Steiger, H. (1993). Anorexia nervosa: Is it the syndrome or the theorist that is culture and gender bound? Transcultural Psychiatric Research Review, 30, 347-357.

Silber, T. (1986) Anorexia nervosa in Blacks and Hispanics, International Journal of Eating Disorders, 5, 121-128.

Stice, E., Spangler, D., Agras, W. S. (2001) Exposure to media-portrayed thin-ideal images adversely affects vulnerable girls: A longitudinal experiment. Journal of Social and Clinical Psychology, 20, 270-288.

Suematsu, H., Ishikawa, H., Kuboki, T., Ito, T. (1985) Statistical studies of anorexia nervosa in Japan, detailed clinical data on 1,011 patients, Psychotherapy and Psychosomatics, 43, 96-103.

Sundgot-Borgen, J. (1993) Prevalence of eating disorders in elite female athletes, International Journal of Sport Nutrition, 3(1), 29-40.

Szmukler, G.I., McCance, C., McCrone, L., Hunter, D. (1984) Anorexia nervosa: a psychiatric case register study from Aberdeen, Psychological Medicine, 16, 49-58.

Turner, B. S. (1992) Regulating bodies: essays in medical sociology. London: Routledge.

Ussher, J. (1997) Body Talk: the Material and Discursive Regulation of Sexuality, Madness and Reproduction. London: Routledge.