Whose Culture is it Anyway? Disentangling Culture and Eating Disorders – Part 4

The more I write about culture and eating disorders, the more I want to know. I keep finding more articles to add to the mix; I know I’m far from the first to be interested in how culture and eating disorders intersect, and for that matter, what counts as “culture.” Still, this has been a fascinating exploration so far! In case you’re curious, this is to be the second last post in the series, for now at least. There will be one more after this, about eating disorders in Ghana (from a Twitter request). In this post, I will continue to explore the “culture boundness” of eating disorders by looking at a study relating to eating disorders in Africa. In this study, Le Grange, Louw, Breen & Katzman (2004) illustrate how eating disorders have emerged in Caucasian and non-Caucasian adolescents in South Africa.


Le Grange and colleagues set out to complicate the package picture of the eating disorder/culture-of-thinness relationship by exploring the emergence of eating disorders in groups traditionally presumed to be “immune” to eating disorders (i.e., black adolescents in an impoverished context). This article begins by exploring several of the tensions I’ve written about so far in this series; first among them, the idea that (as Keel and Klump’s work so clearly highlights) anorexia nervosa might not be so cleanly tied to a culture that prizes thinness as one might expect.

Rather than being culture “bound,” Le Grange et al. suggest that eating disorders might be better understood as “culture reactive” (DiNicola, 1990). This would mean that people living in societies undergoing large-scale social and cultural changes (such as those introduced in prior posts–Fiji and Mexico) might become more prone to developing problematic eating. In this model, instead of the thin ideal, other aspects of culture shifts come to bear on individuals’ sense of self, potentially contributing to the development of eating disorders. What might those “other factors” be, you ask? The authors note:

Come to think of it, any of those articles could make for quite an interesting post about the political and economic sides of “culture.” In any case, the authors explore what this means in the South African context. Among developments they noted in South Africa at the time of writing (post-apartheid, i.e., 1994 and onwards), particularly for Black South Africans:

  • Shifting gender roles
  • Increased mobility between rural and urban settings
  • Increased presence of Western ideologies; e.g., the “Coca-Cola culture,” which Le Grange et al. describe as “an embracing of American individualism, competition, individualistic aspirations, and general worldview”
  • Shift from collectivist to individualist orientation to the world

These shifts, Le Grange et al. argue, may be more pronounced amongst adolescents, who are actively seeking to make use of the new discourses on identity in order to be competitive in society, but who may feel conflicted about where traditional values can factor into this new identity. This relates back to the point in part one of this series about the kinds of multiple pulls on individuals in this “in-between” space of “new” and “traditional” cultures: there is confusion about where one fits, which set of rules govern social life, and how to negotiate between conflicting ideologies one is confronted with every day.

To look at increasing rates of eating disorders in South Africa, the authors note two key tensions:

  • Are increases actually illustrative of increased prevalence, or are they more closely tied to methodological issues (e.g., how representative the sample is of the population studied, validity of questionnaires used in cross-cultural groups, as translation of words does not necessarily mean translation of meaning)?
  • Do eating disorders look different in the South African context?

To try to move toward a more nuanced cross-cultural analysis, Le Grange et al. conducted a two part study with surveys and follow up interviews.

The authors surveyed a mixed sample of students (mean age 16.8; 58% female) from high schools varying in rurality, race segmentation, and affluence, for a total of 813 responses to the EAT-26 and the Bulimic Investigatory Test, Edinburgh (BITE):

  • 362 white students
  • 230 black students
  • 221 mixed-race students

They then conducted follow-up interviews with students (5; all female, ranging in age from 15-19) from one school that historically had primarily black students enrolled. Here, they aimed to confirm a diagnosis of an eating disorder if students scored in range on the questionnaires. They also wanted to use this exploration to look at the Black South African students’ interpretations of survey items.

Perhaps surprisingly, they found that black students were more likely to score in clinical ranges on the EAT-26 and BITE than their white peers. Upon interviewing five young women (interestingly, 59 participants from the chosen high school scored in clinical ranges but only 5 were able to be reached and agreed to participate in an interview), none explicitly espoused what we might think of as “Western beauty ideals”; for instance, 3 noted that they expected older women to be larger, and all chose South African news media sources as preferable.

One of the main problems with the questionnaires for measuring disordered eating in this population emerged while conducting the interviews: the fact that participants were not able, in the context of the questionnaire, to expand on their reasons for the behaviours they engaged in. Le Grange et al. give some examples, including:

  • Preoccupation with food might be related to food insecurity; in other words, a concern about there not being enough food for a family to eat
  • Purging might follow a meal that included pork as an adolescent might have no other food to choose from, but the digestion of the pork would be against religio-cultural beliefs
  • Laxatives were found to be used relatively frequently to counter constipation upon eating some traditional foods

Though these 5 participants had scored in clinical ranges on the EAT-26, their responses to the EDE-Q, administered during the interview, did not indicate an eating disorder. Notably, the finding of more eating disorder pathology among black South African adolescents (versus white and mixed race South African adolescents) might be linked to other complex and intersecting factors, rather than only to an increase in “Westernization.”

What does it mean for our understanding of eating disorders as culture bound?

While these 5 participants’ interviews are obviously not representative of everyone who scored in the clinical range on the written questionnaires, the difficulties the researchers encountered as they conducted this research, including both logistical issues related to making contact with young people living in poverty and needing to clarify questions that did not easily translate, are indications of just how different this “cultural context” might be.

The authors offer some suggestions for future cross-cultural work, including:

  • Not assuming that scores on questionnaires reflect the same “reality” they might in the cultural context in which they were created
  • Working to translate instruments not only into other languages, but carefully attending to different cultural meanings tied into eating behaviours
  • Making note of how poverty and food insecurity can profoundly shape relationships with food in a way that challenges the divide between normal and abnormal
  • Considering that (especially young) people raised in other contexts may not be accustomed to reflecting on behaviours and cognitions in a way that will necessarily translate into a score on a questionnaire

I think that this last point is an especially interesting one. It speaks in part to the assumptions that underlie “Western science,” as it were. We might take for granted that participants will be used to the kind of (particularly psychological) research that requests an objective evaluation of a behaviour, thought, or emotion on a rating scale. This is very much a part of the “culture” of evidence-bases and quantification we live in.

I have to say, this was one of my favourite studies for this series thus far. The nuancing of culture and poverty contexts reveals a lot about our systems of categorization and ways of doing research cross-culturally that offers something beyond a simply movement of assessment strategies from one place to the next.


Le Grange D, Louw J, Breen A, & Katzman MA (2004). The meaning of ‘self-starvation’ in impoverished black adolescents in South Africa. Culture, medicine and psychiatry, 28 (4), 439-61 PMID: 15847050


Andrea is a PhD candidate focusing on individual, familial, and health care definitions and experiences of eating disorder recovery. She has an MSc in Family Relations and Human Development and a BA in Sociology. In her Masters research, she used qualitative and arts-based approaches (digital storytelling) to explore the experiences of young women in recovery from eating disorders. Andrea has recovered from EDNOS. She can be reached at andrea[at]scienceofeds[dot]org.


  1. So basically, as with most mental conditions the causes are multiple and personal. To me it just highlights how inadequate acute care based medicine is in dealing with modern medical issues.

    • Yes, causes are multiple and complex, and their interpretations (and the interpretations of their causes) depend partially on cultural context. And our classification schema can’t just be uprooted from one place and used in another, because of differing cultural meanings particularly tied to food and bodies. I’m not sure what you mean when you refer to modern medical issues?

  2. In my view, that last point you found particularly interesting is a problem when applying these questionnaires everywhere. When I read these questionnaires and think of myself trying to answer them, it’s hard to avoid the impression that the results are anything more than fiction. I mean, applying these questionnaires without subsequent, careful interview is a methodological no-no to my mind, even if they have been in some way validated. I know this will sound completely misguided, but I’m still not convinced of the contrary.

    I guess your series has been pointing out more and more this fundamental difficulty: we don’t have a lab test or exam that can absolutely confirm the existence of an ED and separe it from non-ED. We often rely on a diagnostic manual that is a vague, at times arbitrary description of symptoms that are themselves selected, explained, and grouped according to values that are not purely objective or scientific.

    Westerners want to find out about EDs in South Africa and yet our categories of EDs are vague, insufficient, and imprecise in the Western culture itself. It’s not as if we were absolutely sure about them.

    The fear of fat thing is a good example. Person A says: “I stick to the old school. No ED without fear of fat. People who lack this fear may still struggle tremendously with eating, but they have something else, or need help from other sources than medicine. We all struggle with things in life and experience some degree of impairment, it doesn’t mean we have an illness.” Person B says: “You are totally misguided. EDs can occur in the absence of a fear of fatness and, in fact, you are actually invalidating these people’s suffering.”

    To my mind, there’s no sure way of saying which one is right. No test to be run. Consensus may consider that B is right today but it may well shift to A’s position 10 yrs from now.

    This is in part the reason I liked Lester’s article a lot. In my reading, her reasoning makes sense of the fact that things like, I don’t know, “unduly influenced by body shape” are not objective parameters as blood sugar levels are. Cut-offs for diabetes may change for lots of reason, but blood sugar level in itself is purely quantitative, it’s not a parameter influenced by culture in the same way as “self-evaluation unduly influenced by body shape” is.

    I find myself inhabiting this uncomfortable intellectual space where EDs are surely neurobiological conditions, but also surely culture bound (because we still can’t think about EDs except through our cultural lenses) and even uncertain. EDs surely exist, but I’m not sure about what they are and how to safely define them. This tension is uncomfortable in the sense that is indicative of a fundamental shortcoming/insufficiency in my understanding.

    Your series has been helping me a lot to think more clear about these things. And I find it very rewarding. I’m learning so much too…and not only about EDs!

    • I think that’s one of the cool things about looking at a lot of the ED literature; you can learn so much that goes beyond “eating disorders” themselves and lots about the socio-cultural surrounds. I never thought I’d unearth things I learned in early undergrad (like political economy) when I began to study eating disorders in depth, but here I am… it all reminds me of how important it is to remember that nothing happens in a socio-cultural-historical vacuum. Obvious, I guess, but sometimes when honing in on minutae of disorders etc. we can forget that basic fact, I think.

      Yes, good point. Questionnaires are plagued with critiques, and it’s a bit of a bind- on the one hand, they’re efficient, can result in large sample sizes so lead to more robust results, help with large scale comparisons, and are often “what we’ve got” to make sense of phenomena. On the other hand, without accompanying rich and deep data to explore how people fill them out, we can never really be sure how good of a job they’re doing at explaining what they are supposed to explain. If I’m not mistaken, there is a whole branch of psychometrics-based social science based around the science of how people fill out surveys. Yep.

      You wrote: “Westerners want to find out about EDs in South Africa and yet our categories of EDs are vague, insufficient, and imprecise in the Western culture itself.” – Yes, this too. In some ways it feels like a futile exercise, but in another way I’m curious whether we can flip the lens and apply some of what we learn in cross cultural comparisons (particularly those that involve a good deal of rich-data generation) to our Western ways of categorizing things. Kind of a reverse view on whose knowledge is being imported where. Then again, I presume that would be a hard sell. Many in Western society don’t like to think of Western society being in need of any development.

      I don’t know that the liminal intellectual ground you’re living in indicates any insufficiency on your part- I think that’s a fundamental tension in the ED field! I think living in that tension (partially) unveils the need to avoid reducing EDs to an “either or” when it comes to culture/biology & helps to avoid reductionism. Or at least I hope often feeling in turmoil about such things is a good sign, because I live there too.

Comments are closed.