Endo, Meso, Ecto To your right are three examples of the infamous Ivy League “posture photos” (the linked story, uncredited, is by Ron Rosenbaum), which don’t offer much in the way of prurient interest from an official purveyor of pornography, but it’s the best I can do. For twenty years, from the late 1940s to the late 1960s, at the Ivy League and the Seven Sisters, freshmen were herded into the nurse’s office and snapped nude from three angles. Meryl Streep (Vassar). Hillary Rodham (Wellesley). My mother (Smith), to whom it didn’t occur until ten years later that there might be something untoward about it. That these same photographs were taken at the same time outside the Ivy League apparently bothered no one. It’s fine to take nude photographs of Army recruits and other such riff-raff, but Girls of the Ivy League? Quel horreur!

Behind the posture photos was one William Sheldon, a psychologist whose body-typing theories, highly respectable at the time, provided most of their rationale. He published many of them in his Atlas of Men and intended to publish many more in an Atlas of Women, unfortunately never completed. I happened upon a copy of one of his books, Varieties of Temperament, which shows his talents to lie more in the literary than the visual arts. Varieties of Temperament concludes with capsule summaries of Sheldon’s 200 subjects. Some are savage:

An expansive, popular, garrulous young man who wasted his time as a medical student, but has since become an oculist with quite a lucrative business building up. Has joined the Masons and one or two other fraternal organizations of that sort. He is married to a gluttonous, gleeful little 6-3-1. They set a fine table.

(6-3-1 is a “somatotype,” about which much, much more below.)

A weak, baffled undergraduate to whom college is like a bad dream. He wants to cry on somebody’s shoulder but is instead required to attend gymnasium classes.

Some are shrewd:

Has as unpleasant a disposition as is often met. He suggests a partly tamed native gray rat among a generally well-behaved colony of white ones. He continually bares his teeth, although he rarely bites… Many adopt the tolerant attitude toward him that is often shown toward a noisy terrier.

…he has been accused vaguely of homosexuality, but he is in fact strongly heterosexual, and now that his academic career is securely under way he devotes perhaps a disproportionately great amount of time to the sexual pursuit. He has no etchings but his collection of symphonies has begun to grow a bit notorious.

Occasionally he seems to have been gotten the best of:

A young rabbi studying psychology. Polite, ceremonious, yet watchful and critical. His fine black eyes are disconcertingly observant. He is hyperattentional and overly intent.

All make the best bathroom reading you could hope for. Do any other shrinks write like this when no one is looking? And who was William Sheldon anyway?

Sheldon was born in 1898, grew up rural in Rhode Island, and as boy became an expert enough shot to be able to hit a marble thrown 20 feet in the air, which skill he demonstrated to Annie Oakley. He took an M.D. and Ph.D at the University of Chicago in the 1920s, began his teaching career there and at the University of Wisconsin and moved to Harvard in 1940, where he did his best work, published in The Varieties of Human Physique (1940) and The Varieties of Temperament (1942).

He was divorced twice and was apparently not the easiest of men to get on with. At Harvard he served on a Ph.D. examination committee that passed a student he considered unprepared. Sheldon concluded that the term “Doctor” no longer had meaning, and for some time insisted on addressing everyone on campus, including the elevator man, as “Doctor,” embarrassing his faculty colleagues, and no doubt the elevator man.

Sheldon enjoyed a great vogue for a while. Life magazine devoted a cover story to him in 1951, and his disciples included Aldous Huxley, who credited Sheldon with much of his success. The rise of behaviorism did his reputation in. Skinnerian psychology made his theories glacially unfashionable, and he died, bitter and forgotten, in 1977, having published almost nothing for more than 20 years.

Folk wisdom has it that temperament and physique are related: fat people are reputed jolly, short people “Napoleonic,” tall and thin people “awkward,” and so forth. Sheldon set out to ground this scientifically. He distinguished among three “somatotypes,” as he called them — endomorphy, mesomorphy, and ectomorphy; the terms are still with us. Each body type has a corresponding temperament: viscerotonic (endomorph), somatotonic (mesomorph), and cerebrotonic (ectomorph). Sheldon ranked his subjects according to their endowment of each body type, and temperament, on a 7-point scale. Although we now tend to think of people as one type or the other, the pure endo (7-1-1), meso (1-7-1), or ecto (1-1-7) turns out, in Sheldon’s system, to be extremely rare. He concluded that temperament usually matches physique nearly exactly, with at most a point difference on each of the three scales.

Like Freud, Sheldon fancied himself a scientist; Varieties of Temperament is full of scales, indices, and standard deviations. As he describes his own procedure:

First a list of 650 alleged traits of temperament was collected… These were sifted, condensed, and described as systematically as possible. A few contributions from our own observation were added, and the list was finally reduced to 50 traits which seemed to embrace all of the ideas [emphasis his] represented in the original 650. The 50 traits were then incorporated into a simple 5-point graphic rating scale (later expanded to a 7-point scale)… Then began the tedious process of analyzing a series of subjects in order to rate them in these 50 traits…

We then proceeded to build up lists of such clusters of traits as showed consistently positive intercorrelations among themselves, much after the manner of building up suits in a game of cards…we soon found that three groups of traits showed positive intercorrelation among themselves, and negative correlation with all or nearly all of the other traits

…we set up two quite arbitrary criteria for determining the qualification of a trait within one of the nuclear groups. First, the trait must show a positive correlation of +.60 with every other trait in its nuclear group. Second, it must show a negative correlation of -.30 with every trait in each of the other two nuclear groups. Employing these criteria rigidly, we found that…22 of the original 50 traits had qualified.

This resembles hard science as numerology resembles mathematics. It has about it a distinct pre-scientific whiff of bacon — Francis Bacon, who believed that one could formulate useful scientific hypotheses by gazing steadily at an object and making lists of what one notices. The blithe way Sheldon throws out the 28 of his 50 traits that don’t meet the correlation tests adds a more modern, Johnnie-Cochran flavor: if the trait don’t fit, get rid of it. Since there are no hard-and-fast rules for classifying types, either physically or by temperament, the scheme groans under confirmation bias.

Still, this is not physics but social science, where there may be something to be said for simply looking around. Dubious methods may also serve a defensible thesis. One suspects that Sheldon has been “discredited” less for his technique than his belief that physique influences temperament, which, knotty cause-effect questions aside, is obvious to any sentient inhabitant of the planet.

Sheldon was a prescient critic of therapy, which he considered useless for most people and counterproductive for some. He was no fan of endocrine, the Prozac of the 1940s: “A long history of experimentation with endocrine therapy seems to have intensified [the subject’s] unhappiness without improving the situation.” He didn’t go in much for talking therapy either: “He has been having a long series of conferences with a psychiatrist whose point of view places great weight on the early intrafamilial relationships. This has turned the youth’s attention more than ever toward his parents and familial entanglements… He has learned to blame his mother and father for every disappointment.” Thus Sheldon casually hands Freud his head without even mentioning his name. Of his 200 cases in Varieties of Temperament Sheldon recommends psychoanalysis for about five. Anyone who has observed its whiny modern products will share his skepticism.

The customary accusations of “eugenicist” and “biological determinist” have been hurled at Sheldon, with the customary accuracy. Sheldon nowhere advocated selective human breeding, and although he sometimes lets his system carry him away (“2-2-6’s do not write novels, they only dream of it”), the general thrust of his work is away from determinism. (It is the extreme environmentalists, on the contrary, who tend toward it.) Sheldon takes pains to point up cases of identical somatotype where one succeeds and one fails. At one point he remarks of two cases, an extreme cerebrotonic and an extreme somatotonic, “it is useless to expect Christopher to become a heavyweight boxing champion, or Boris to learn cuneiform,” which may sound like biological determinism to some but sounds like common sense to me.

Sheldon preaches, insofar as he preaches, that human nature is not indefinitely plastic. “Know thyself” is how the Greeks put it. Today that makes him a quack; the modern psychiatric wisdom is “be what you want to be.” How will that look in fifty years?

(Update: Howard Owens comments.)

Aaron Haspel | Posted September 4, 2003 @ 1:37 AM | Culture

16 Responses to “Sympathy for the Devil”

  1. 1 1. Eddie Thomas

    "It has about it a distinct pre-scientific whiff of bacon Francis Bacon, who believed that one could formulate useful scientific hypotheses by gazing steadily at an object and making lists of what one notices."

    There is no method for formulating scientific hypotheses, so I’m not sure why Bacon’s approach isn’t as good as any other. (Or are you just arguing that Bacon was wrong to place such confidence in it?) The history of the study of electricity began in just this way, with William Gilbert carefully experimenting with the scope and conditions of what we would now call static electricity, seemingly to good effect. The approach may be pre-theoretical, but I don’t see how it can be said to be pre-scientific.


  2. 2 2. Michael Krantz

    Only someone who knows little to nothing of psychiatry, psychology and/or ‘talk therapy’ could blithely write that "the modern psychiatric wisdom is ‘be what you want to be.’" This is false. The modern psychiatric wisdom, if it’s possible to distill it down to one phrase, is "Know who you are," which I consider as good a starting point for self-improvement as any. The modern ADVERTISING wisdom is "be what you want to be," which I consider as good a starting point as any for selling something to somebody.


  3. 3 3. Bill Kaplan

    This is just a method of reinforcing your prejudice against fat people, isn’t it?

    By the way, my favorite Sheldon quote goes something like this: "A French poodle can’t become a German shepard by eating a lot."


  4. 4 4. Aaron Haspel

    Eddie: We may not know how to form scientific hypotheses. We do know that the hypothesis must precede the observation, otherwise you don’t know what you’re looking for. Bacon did not understand this, which is why his view is pre-scientific and his own investigations never got anywhere.

    Michael: Your distillation is generous. The topic of genetic influence on "who you are" remains largely taboo, in and out of psychiatry, which doesn’t strike me as much of a starting point for self-knowledge, or anything at all. Do you suppose that Sheldon is written off as a quack because of his methodology? It’s no worse than the general run of social science today; in some ways, being less influenced by scientism, it is better.

    Bill: Yes.


  5. 5 5. Bill Kaplan

    "We may not know how to form scientific hypotheses. We do know that the hypothesis must precede the observation, otherwise you don’t know what you’re looking for."

    Not really. That might be true for both special and general relativity, but it certainly wasn’t true for quantum mechanics. With quantum mechanics it was the divergence between the the expected results using classical theory and the orbserved results that gave rise to the need for a new theory.


  6. 6 6. Aaron Haspel

    Your example presupposes a rather precise idea of what you’re looking for to begin with. Giving rise to the need for a new theory is one thing, and formulating that theory so it can be tested is something else again. Quantum mechanics would have gone nowhere had it employed the Bacon method.


  7. 7 7. Bill Kaplan

    My point is this: In its earliest stages quantum mechanics didn’t need to be tested. Why? Because the results were already a given, only their explanation wasn’t. The "ultraviolet catastrophe" was well known. "Black box" radiation did not produce infinite ultraviolet light as classical theory predicted. It radiated different colors in limited quantities. The hypothesis clearly followed the problem.

    Now the Wheeler thought experiments wherein the past can be changed, explained in part here, would have to be confirmed by experiment. But how do you do that if the past is changeable?


  8. 8 8. Eddie Thomas

    "We may not know how to form scientific hypotheses. We do know that the hypothesis must precede the observation, otherwise you don’t know what you’re looking for. Bacon did not understand this, which is why his view is pre-scientific and his own investigations never got anywhere."

    Good luck trying to form a hypothesis without having observed anything at all. To be sure, having a hypothesis helps shape up a plan of observation, and gives the results of that observation more significance, but every scientific endeavor begins at some point from certain pre-theoretical experiences. Bacon’s approach makes perfect sense in fields of inquiry that are very young. In older fields, there is already a wealth of both theoretical models and experimental results, so getting acquainted with the phenomenon takes a different path. I’m guessing that if you look at the history of different fields, you’ll find that their origins are Baconian.


  9. 9 9. Bill Kaplan

    This Onion article is my favorite and shows what happens when you have observation only without a good hypothesis.


  10. 10 10. Will Duquette

    The whole tenor of this discussion is silly. Of course you have to observe before you hypothesize–until you’ve noticed something you don’t understand and asked yourself "Why is that?" you don’t need a hypothesis. And then once you’ve come up with the hypothesis, then you go back and observe again, to see whether your hypothesis pans out or not.


  11. 11 11. Alan Sullivan

    I read Sheldon and Skinner during college in the late 1960’s. The latter was assigned; the former I found on my own. Sheldon appealed to me a lot more than the rat-man. I haven’t reconsidered "Varieties of Temperament" in years. Thanks for shining the bright light of your blog into this dim place.


  12. 12 12. Sheila Lennon

    I just blogged about this too — "When freshmen were photographed nude" My own posture picture, Wellesley assures me, was destroyed within a year of being shot.

    The whole thing was so weird, but it was all part of the confusing, "Go here, do that" orientation week.

    And of course, all uses were to be clinical and noble.


  13. 13 13. Carl Jung

    I just finished reading a Robertson Davies novel, The Rebel Angels, which has a minor character based on Sheldon. Sadly, everyone else in the novel buys into the whole somatotype theory, stunting their development by informing them who they are without empowering them to become who they want to be. Somatotypes, if bad (social) science, allowed Davies to dream up some interesting characters.


  14. 14 14. Bill Kaplan

    Perhaps you can comment on the theory of somatotypes and "Moneyball".


  15. 15 15. mccaff

    Induction & Deduction seem two facets of the same thing. To have one, one must have the other. All thought contains theory. The thought: no theory — is a theory. All theory contains data. No data — no theory, and not a theory.


  16. 16 16. p

    i get a kick out of people that talk down about somatotyping and say that people can become what they want to. it’s as if they then forget to follow up these silly statements with the most important part: actually deeply observing numerous people for long enough periods of time to realize that those people really never end up changing much and that they do indeed fit the somatotypes pretty damn close. time and again. it’s very predictable, and i do it for people i’ve never even met to their friends and family. “He’s like this” and “she’s like that” and they say “Yeah, how did you know!?”

    no, it’s actually all about insecurity. everyone is just insecure about their own issues, own bodies, and lack of uniqueness…wondering just if everyone else will find out. that’s quaint since nobody really cares about their petty issues, their mediocre body, and the fact that they really are not unique. that’s because everyone else is fretting over THEIR own insecurities, mediocre bodies, and mundanity. Think about it: with 7 billion humans, and our limited amount of variation, there are DOZENS of MILLIONS of humans very very similar to you, and DOZENS of MILLIONS or people very very similar to me. right now. thinking almost identical thought patterns. feeling almost identical feelings. reacting in almost identical ways. factor in all the billions that have come and gone, and you in essence, my ridiculously common friend, have come and gone many many many times before…are now…and will continue to. you’re just one more ant in the endless farm. deal with it.


Add a Comment

Basic HTML acceptable. Two-link limit per comment.