The Tarot and the Atkins Diet, or How I Learned to Stop Worrying and Love Structural Anthropology

The Hole in the Centerpiece

When I first learned CPR, the sequence was simple. Two long breaths, followed by fifteen chest compressions, rinse, lather, repeat. The instructor emphasized these numbers- we were even encouraged to count aloud, to ensure that bystanders heard us and understood we were competently following a plan. In the advanced class, you learned different numbers- one breath and five compressions, perhaps, or (after intubations) twelve breaths and one hundred compressions, asynchronously every minute.

That’s not how CPR is taught anymore. For one thing, the breaths are optional for non-professionals- repeated, fast chest compressions circulate some air, and build up pressure in the (elastic) aorta, and this pressure dissipates when the rescuer stops to give a breath. The added benefit of that full breath (over the “mini-breaths” that come with compressions) was less than the cost of losing that pressure, and there were concerns about people not attempting CPR out of fear of mouth-to-mouth contact with a dead person. Furthermore, it turns out, even after an airway has been secured, positive pressure ventilation increases intrathoracic pressure (normal inhalation decreases it) and flattens the vena cava, diminishing preload and compromising cardiac output. Therefore, even as a professional rescuer, I am now instructed to provide no more than eight-to-ten breaths per minute by ET tube, much less than the “hyperventilation” I learned back in… well it was a while ago. If you do alternate breaths and compressions, you now do twice as many compressions- thirty- for every sequence of two breaths. There are other changes too.

I tell this story to give some perspective to an article a fellow student forwarded me from the Huffington Post. Larry Dorsey cites a BMJ article evaluating “common medical treatments” and announcing that, under statistical review, only 13% are found to be beneficial and 23% are found to be “likely” to be beneficial. Of the remainder, the majority (46%) are insufficiently tested against, among other things, placebos, and 18% are “as likely to be harmful as beneficial” or worse. These are shocking numbers- only a third of medical treatments work?- and seem implausible at first glance. The CPR example, though, makes it clear how little we understand about what we do- for decades, doctors, nurses, and emergency workers have been diligently providing breaths that actually made patients less viable. Civilians were taught to check for pulses, before it was understood that without extensive training and experience, non-professionals are likely to mistakenly feel a pulse when none is present. Taking a wider view, there is considerable controversy about whether statin drugs- which inhibit the HMG-coA-reductase enzyme needed to synthesize cholesterol- prolong life in anyone other than middle aged men with a history of severe heart disease; studies in women, the elderly, and people who have not experienced a heart attack suggest that while statins do indeed lower cholesterol and prevent cardiac events, they don’t change overall mortality- you are correspondingly more likely to die of something else. Even hoary old coronary bypass procedures don’t seem to significantly extend patients’ lifespans: viability at seven years improves from 67% to 71%, at twelve years even this advantage has virtually evaporated.

These should not be seen as failures, however. Even strictly scientific biomedicine is a stepwise approximation, everything gets a little bit better every year, but not without early misapprehensions and assumptions. The changes in CPR and the past changes in inframammary ligation surgery (look that one up on your own) and the possible future changes in cholesterol management are interesting because of how they are wrong, and because of what this says about the cultural place of medicine.

Placebos For Everything

If you give people a pill that looks just like an aspirin, will it make their headache go away? Yes. Why? The common term is the placebo effect, named for inert substances passed off as medications, but anthropologist Daniel Moerman has argued that a better term is “meaning effect.” For one thing, although actual aspirin will make your headache go away better than aspirin-shaped starch pellets, fancier brand-name aspirin will work better than actual aspirin that looks like a starch pellet, starch pellets that look like fancy aspirin will work better than starch pellets that look like starch pellets, and all of these will work better than aspirin that you don’t see. Even though medication works, the experience of being medicated, either with real medicine or inert placebos, makes it work better.

This raises a number of questions, entertainingly posed by Moerman’s Meaning, Medicine, and the Placebo Effect, about what exactly the work of medicine really is. On the one hand, some drugs and therapies (approximately 36%) are known to be more effective than placebo, meaning that you could sneak them into people’s bodies and they would do better than if you did nothing. On the other hand, if you make a great big deal of treating people, either with well-tested active medicines, poorly-tested (or poorly-testing) medications, or with inert substances, they will also do better than if you gave the same substance secretly.

This significantly changes the discussion of the ethics of placebos. It is still clearly an ethical obligation for physicians to offer their patients the treatment that is known to offer the best possible benefit with the least possible risk. Moerman’s research suggests it is also an ethical obligation for physicians to administer this treatment in the manner than best activates the patient’s “meaning response.” This is where we start to run into strange, strange questions. How should a physician do such a thing? And how should they understand it?

Structural Anthropology and the language of symbols

My first academic training was in math and linguistics, so its a wonder I never fell into structural anthropology until recently. Cognitive linguists posit that human language is structurally determined by human cognition, that there are deep homologies between languages that seem to share no particular characteristics. The formation of speech (or sign) arises out of structural, syntactic, and lexical elements which are strongly, somewhat, and not conserved respectively. Languages contain a generative rule-set that creates grammatical utterances, and although these rule-sets tend to vary across different languages, commonalities and constants can be observed (syntax.) The words that mean particular objects, or concepts, however, are not conserved at all except between related languages, and neither are the concepts themselves (lexical elements.) There is no reason to expect the Chinese word for “anxiety” to resemble the English word for “anxiety,” but this doesn’t mean that the grammatical rules that govern the structure of Chinese sentences come from an equivalently different set of possible rules as the rules that govern English.

Structural anthropology posits that there is a similar “generative grammar of symbols” underlying culture. The symbols themselves may vary widely but the use of symbols and the ways that their interrelationships are structured are at least partly conserved. Thus it is possible to talk about “healing” as an intercultural phenomenon, provided we allow room for the specific content units of “healing” cultures to be as different as English and Chinese words.

What is universal healing? Axiomatically, structural anthropology considers the dominant culture (i.e. the anthropologist’s culture) to be subject to the same constraints as “world cultures,” and thus a definition of healing must be as applicable to your most recent trip to the dentist as to a nomadic sorcerer. Psychiatrist Jerome Frank, in his 1967 crosscultural contextualization of psychoanalysis, defined “psychotherapy” as

only those types of influence characterized by:
1. a trained, socially sanctioned healer, whose healing powers are accepted by the sufferer and by his social group or an important segment of it
2. a sufferer who seeks relief from the healer
3. a circumscribed, more or less structured series of contacts between the healer and the sufferer, through which the healer, often with the aid of a group, tries to produce certain changes in the sufferer’s emotional state, attitudes and behavior. All concerned believe these changes will help him. Although physical and chemical adjuncts may be used, the healing influence is primarily exercised by words, acts, and rituals in which sufferer, healer, and- if there is one- group, participate jointly.
Frank, J. Persuasion and Healing: A Comparative Study of Psychotherapy Baltimore, Johns Hopkins Press, 1973. emphasis mine

This to me scans very close to Moerman’s “meaning effect,” in fact it seems to encapsulate the sum total of non-surgical, non-pharmacological aspects of western (US) medicine.

Since Frank’s work was published, anthropologists have expanded significantly on the nature of the symbolic underpinnings of health. A recent textbook on ethnomedicine (Pamela Erickson’s) adds a “theory of disease causation,” an “organizational system for caring for the ill” and a “system for paying the healer” (this last I believe to be less universal- but anyhow.)  Of these, following Kleinman, the most critical is the theory of causation, which represents the link between the cultural world of symbols and the sense of illness or dis-ease that demands a medical explanation.

Causative theories are (crudely) separated into naturalistic- infectious agents, falling branches, misbegotten genes- and personalistic- taboos, transgressive acts, conflict with a magical enemy, etc. Biomedicine tends to abjure personalistic causes in favor of naturalistic ones, but from an anthropologist’s viewpoint, western medicine is simply overflowing with personalistic theories of disease. We are besotted by the idea of “will-power”- a social, moral and ethical virtue to whose absence we ascribe all manner of illnesses, from diabetes to lung cancer. We are entranced by “stress,” despite the continual reminders from psychologists that the concept cannot be isolated or quantified. We have powerful sanctions against certain behaviors that are contextualized as “illness” and legislated as crimes. To quote Erickson some more:

…our medical system reflects our core cultural themes of independence, individualism, scientific positivism, and capitalism. In most, perhaps all societies the threat of illness plays a powerful role in the moral order, and the threat of illness as an outcome of personal behavior is a powerful motivation to obey moral, social, and environmental norms. In Western culture and its medical system, personal responsibility for health is a paramount theme, and those who “choose” lifestyles that lead to illnesses are held responsible for those illnesses.

Consider, for instance, the Atkins diet. I have never read Dr. Atkins book, so my knowledge about the diet comes entirely from the folk sector- what I’ve heard from other people who heard from other people. The essence of the Atkins diet is to avoid carbohydrates, and focus on getting energy from proteins and fats. However, the popular understanding of these categories is not the same as what a nutritionist might say, and the resulting diet is certainly different from what a dietician would recommend. Certain foods are identified as “lower carb” when they may only be so by dilution- whole wheat bread, for instance, is considered virtuous, while bean salad, we are reminded, is actually a carbohydrate. In point of fact, beans do contain carbohydrates (so do meats) but they are significantly lower than whole wheat grain products which enjoy the Atkins stamp. Yet, bean salad lacks whole wheat’s symbolic meaning as a less-appetizing alternative to tastier familiar foods, and hence for the non-dietician public fits much better into the self-abnegatory ontology of “diet foods.”

It is worth pointing out- hell, its worth a whole other entry- that a major exception to this characterization is alcoholics anonymous and its imitators. Here is a well-tested, well-liked medical intervention that explicitly rejects the idea of “will-power” as a form of medicine against illness. Participants are encouraged to surrender their belief in their own individual exceptional strength- the key component of every other lifestyle intervention I’ve ever been taught- in favor of faith in the group or a higher power which, for some, is the same as the group. Compare this to the last conversation you heard a doctor have with an overweight patient about exercising.

Placebos- Giving and Misgivings

All this is fairly straightforward, at least within the poorly-mapped intersection of medicine and culture. The problem is, what do you do about it? How do you use these insights to provide people with care that both improves their medical outcomes (in a direction they value- longer lives aren’t always better) and reinforces their sense of belonging and identity within their culture at the same time? For at least fifty percent of doctors, their approach includes prescribing placebos- or at least drugs not known to be germane to the presenting illness. While this is officially frowned upon, its important to remember- a placebo cure is a cure. If your cholesterol goes down because you took a sugar pill as part of a controlled survey, guess what- your cholesterol is now lower, and you can enjoy the health effects of that. If the drug that induced your cancer to regress is shown to be no more effective than a placebo, your cancer doesn’t come back. Using placebos where better treatment is available (or general antidepressants for physical complaints- a common strategy that lives on the margins of the placebo effect) isn’t the most ethical thing, but neither is it worthless. People do get better.

Furthermore, they get better when the treatment they receive- the evidence-based, reproducibly-better-than-placebo treatment- is given in a context that best activates the narratives of meaning appropriate to the patient’s culture. Your best option, as a physician, is to provide treatment that has both verifiable scientific effects on a disease and a strong placebo effect as well. This means not just slipping the drug into a body, but making a big, confident deal of it- acting exactly like the patient’s idea of the Best Doctor in the World, giving the Best Drug Ever.

The problem is, I don’t particularly like the dominant narrative of western medicine. The Best Doctor in the World act strikes me as arrogant, reinforces the subservience and impotence of the “patient,” and fundamentally enables the massive transfer of cultural power from participatory, community-based traditions into a (profit driven) industry. Given what I know about the uncertainty of health science, it also strikes me as somewhat dishonest. I don’t want that person to be me, and I certainly don’t want anybody else’s life depending on it. It seems like too much hocus pocus to embrace.

And then I read Levi-Strauss’ analysis of Franz Boas’ famous anecdote about Quesalid.

Quesalid was a Kwakiutl man who believed that the shamanic healing ceremonies he had grown up with were faked- the shamans were using sleight-of-hand and prop tricks to simulate curing. When offered a chance to study to become a shaman himself, he “did not wait to be asked twice”- his intention, from the beginning, was to learn the tricks and expose them, and free the community of the trouble caused by non-healing healings. While in school, he found himself in a position of having to perform the tricks to heal a sick person whose family had sought him out- and he found they succeeded, wildly. In fact, despite a continuing cynicism about the motivations and skills of his colleagues, Quesalid became a renowned healer in his own right. Levi-Strauss uses the anecdote to underscore the cross-cultural equivalence of psychoanalysis and ritual healing, and places the acknowledged expert healer at the crux of both. “Quesalid did not become a great shaman because he cured his patients;” Levi-Strauss concludes, “he cured his patients because he had become a great shaman.”

Tarot- a Clinical Education in Bulls–t

How, then, to embrace the hocus pocus aspect of medicine, without necessarily co-signing for the problematic aspects of health care and doctor worship? I continued my reading into the next chapter of Levi-Strauss, a close textual analysis of a healing song sung to women experiencing difficult childbirth in South America. The song uses the symbolic elements of the Cuna culture to develop a narrative, strange to my etic perspective, that incorporates and explains sensations and phenomena in terms of a spirit conflict in which the healer is directing an army of good. More than this, though, it also suggests sensations and phenomena that rise above the limina or experience and become real- a phenomenon Levi-Strauss describes as the manipulation of representations. By changing what an experience means, the healer can “bring about a modification of the organic functions of the woman in childbirth”- a concise summary of Moerman’s analysis of the placebo/meaning effect.

Narrativizing experience, and then manipulating the narrative and the symbols contained therein to create a more positive meaning? This is a radically more flexible and self-aware framework for understanding healer self-presentation than merely acting confident and believing in your drugs. For one thing, Levi-Strauss’ healing model is interactive- it depends on an acute awareness of and response to the cultural meaning of things- the meaning to the patient. Ultimately the patient and physician work together to create a mutually acceptable story of how the prescribed interventions will lead to healing, the patient contributing the worldview and belief system, and the physician working within this system to reframe the “in play” symbols towards a healthful end. It doesn’t matter whether the healer, like Quesalid, disbelieves in the sick person’s symbolic worldview, and it certainly doesn’t matter whether the healer, like a standard issue western physician, can force the patient to adopt a biomedical explanatory paradigm. The final ritual, if you will, can be as idiosyncratic as our multifarious and individualist culture allows.

Unfortunately, this is neither an innate skill set, nor is interactive symbolic narratization taught in medical school*. In order to develop this ability, a medical student must find a way to practice co-creating meaningful symbolic interpretations on their own. For myself, I puzzled over this only briefly, and then a friend gave me a tarot deck.

Do I “believe” in the tarot? No, of course not. Its an interesting collection of cultural touchstones developed from late nineteenth century romanticism’s dim memory of earlier post-renaissance romanticism’s heroic mythology of the European medieval age. Even tarot apologists, who claim that the tarot encodes human “archetypes” are making the same mistake as Freud, or the early linguists who searched for an essential “Adamic” or “natural language.” Archetypes, like words, are inherently culture-bound, and culture varies with time and geography and even with the individual. The tarot is a symbolic lexicon, but it has no more validity than any other…

…except that people in my culture take it seriously, and I can get books on it.

The Modern Faith Healer

So this is where my analysis has led me. I may not be able, ethically, to embrace the Best Authoritarian Doctor in the World approach to inspiring patient confidence, but I can see the value in providing patients with an explanatory narrative of their illness, treatment, and prognosis that contains symbolically potent suggestions that contribute to healthful behaviors or a more powerful placebo effect- layered, of course, over the best available evidence-based medicine. I haven’t yet found a perfect way to develop those skills, but I have a plan which, while admittedly being rather silly, compares more-or-less-equivalently with the sleights-of-hand offered by millennia of less-well-equipped healers. If, someday, statins turn out to be no more effective than a bloody tuft of down spat from a cheek, I hope I can rest in the confidence that by then, like Quesalid, I will have figured out the important part.

A

*- If I were not being so strictly anonymous, I would explicitly thank a few clinical preceptors I’ve known who, possibly without being aware of it, respond rapidly to patients’ desire for meaning with individually appropriate narratives of exceptional explanatory and suggestive power. This is one of those “secret superpowers” that distinguish good physicians from the merely capable, and I feel honored every time I get to watch a good physician at work.

Also, thanks to AGK for sending me to Daniel Moerman