Banner
    Born Evil: Is This The Rebirth Of Phrenology?
    By Oliver Knevitt | July 26th 2011 06:22 PM | 17 comments | Print | E-mail | Track Comments
    About Oliver

    In a nutshell: I like fossils. But even more than than that, I like arguments about fossils. Which is why my current occupation as a PhD researcher...

    View Oliver's Profile
    There hasn't been much of a debate about this paper at Science 2.0, so I thought I might briefly review it, and see what everybody else thinks. But, before we start, I should make one thing clear. I'm not a psychologist. I'm not even an evolutionary psychologist; I'm a paleontologist. And I should also make it clear that I am often very dubious of a lot of the findings of evolutionary psychologists, which often seem to me like pontificating on very banal things with very little actual science going on.

    Anyway, I happened to see the advance publication of this paper, published in Proceedings, which I want to hear some other peoples views on. Because, at its core, this paper implies that not only can people be born immoral, but that this predilection is written in the structure of our faces.



    From Flickr, by Gaetan Lee

    Phrenology is the belief that different aspects of your character manifest themselves in bumps and bulges in the skull, which correspond to the area of the faculty responsible for that trait in the brain itself. It was all the rage in Victorian England, and indeed around the world, and it was a reputable sister to the the field of psychology. Many authors of this period were enthusiasts, including Bram Stoker, the Brontes, and Herman Melville. This is an extract from the Hound of the Baskervilles, by Arthur Conan Doyle,



    Mortimer (centre) talking to Baskerville, Holmes and Watson

    Dr. Mortimer: You interest me very much, Mr. Holmes. I had hardly expected so dolichocephalic a skull, or such well-marked supra-orbital development. Would you have any objection to my running my finger along your parietal fissure?
    Sherlock Holmes: Please, Dr. Mortimer.
    Dr. Mortimer: A cast of your skull Sir, until the original becomes available?
    [Holmes bursts into laughter]
    Dr. Mortimer: It is not my intention to be fulsome, but I confess, I covet your skull.
    Sherlock Holmes: Behave and sit down, Dr. Mortimer.
    [Gesturing to the chair]
    Dr. Mortimer: [dog sits beside Mr. Mortimer] Spot. Yes, good boy.
    Sherlock Holmes: Well, I presume it was not your phrenological passion which drew you to Baker Street.
    Dr. Mortimer: Unfortunately it was not, Sir.


    Dolichocephaly is a long, thin head, characteristic of a typical Aryan head. The arches over the eyes are the areas, according to System of Phrenology by George Combe, that house the organs responsible for perception, such as weight, number, tune, and hence in Holmes' case, the moral faculties are exceptionally developed. Combe explains of this area that
    it is the faculty of consciousness [...] which produces the feeling of natural right on the part of one to demand, and of natural obligation on another to perform.

    Needless to say, it's all bunk. As we all know, the consciousness isn't compartmentalised like that. Phrenology's proponents completely ignored the complete lack of scientific evidence and continued to make what were little more than assertions about people's personalities. Eventually, it became widely seen as pseudoscience, like alchemy or necromancy. The real death knell was a real life Godwin's effect: the Nazis used phrenology to define their aryan master race. Have a look at this Google Ngrams for the popularity of phrenology in the literature over the past 200 years, showing an almighty peak in the mid 19th century before its subsequent preciptious fall from grace,





    Recently, a growing body of work has begun to show that many aspects of our morality,
    and the ability to detect unethical behaviour in others, may, in fact, be innate. But here is where this study treads on new grounds: can genetically determined physical characteristics be a visual indicator of an individual's morality? We already know that facial expressions like facial tension and posed smiling are interpreted as signifying ethically questionable behaviour. But could we be born with an innate, underlying bias, that is written all over our faces? So, a psychologist would be able to examine a skull, in the same way as the phrenology enthusiast Dr. Mortimer did above, and conclude "Hmmm... you seem to be an aggressive person, and your moral senses seem to be lacking..."

    My first reaction on seeing this paper title, I'll have to admit, was "Horseshit - surely this is just phrenology disguised as a scientific study?" Which makes me a bad scientist, obviously - I mean, you can't judge a paper by it's title, only by it's scientific rigour - and I can't really dismiss something in a subject I know little about. And, after all, it's published in the usually dependable Proceedings. So let's hear them out!

    Theoretically, of course, if we had a reliable and unique indicator of unethical personality, individuals that possessed it would be selected out of the gene pool - nobody likes unethical dicks. However, Haselhuhn and Wong physical propose an exception to this. Is is possible that physical associates of evolutionarily desirable characteristics, such as physical dominance, are highly correlated with other, less socially acceptable actions, such as cheating?

    The thing is, like it or not, aggression is consistently shown to be a preferable character in a mate (though not a requisite, clearly as with yours truly). But wouldn't aggression also be correlated with unethical behaviour? If so, then physical traits that have been selected as reliable signals of male dominance and aggression would, by proxy, serve as reliable predictors of morally questionable actions.

    Now, it is already possible to read people's faces to see how aggressive they are, through a parameter called the facial width-to-height ratio (WHR). Men with greater facial WHRs are more likely to retaliate to perceived slights by others, and are more likely to act in their own self-interest, even if it means violating another's trust. Basically, don't trust wide faced men.




    Don't trust this guy...



    ...but do trust this guy
    . I should point out, before any snarky commenter does, that really, these differences probably only manifest after puberty

    Here's the science bit, then. In the first study, WHRs of 192 male and female students were measured from photographs. The students were randomly paired to negotiate by email a fictional business deal over property. The "seller" was told not to sell the property if it was to be developed commercially, but the "buyer" was told to buy the property to turn it into a hotel, providing opportunities for deception during the negotiations. In the second study, the 103 participants played a dice rolling game, this time playing for a chance to win money. Both men and women reported dice rolls significantly higher than rolls expected by chance, but again, only high WHR in men emerged as a predictor of cheating.

    So, the results suggest that men who lied were three times as likely to have wide faces as thin faces. Maybe there's some truth in the ol' phrenology then? Indeed, Sherlock Holmes, with his long
    dolichocephalic head, would, according to this research, be inferred an epitome of a deeply moralistic person; something we know seats well with the "real" Holmes.

    Haselhuhn relates this marker to testosterone related changes during puberty, and state that

    Our analyses indicate that this effect was driven by men's sense of power. Men with relatively larger WHRs felt more powerful and, in turn, this sense of power directly predicted the overstatement of the reported dice rolls. Men with relatively wider faces are more aggressive and self-interested, which allows them to secure a greater share of resources when competing with other men.

    However, Paul Zak has pointed out that just 18 out of the 192 students lied during the negotiations, and of the 115 men in the study, only 13 lied. It could feasibly be just chance, then. He also makes the point that,

    Alpha males have less rather than more incentive to cheat because they are on top of the world, with the best looks and the most attractive females

    So, I might reserve judgement. I do have to say, though: it's not phrenology in the truest sense. It's certainly not suggesting that the cheekbones are bigger because that brain needs the room or something, and we certainly shouldn't dismiss it out of hand. But what does everybody else think?


    -------
    Citation

    Michael P. Haselhuhn and Elaine M. Wong
    Bad to the bone: facial structure predicts unethical behaviour
    Proc R Soc B 2011

    Other links,

    http://www.newscientist.com/article/dn20663-are-widefaced-men-rascals.html
    http://news.discovery.com/human/facial-bone-structure-crime-110705.html

    Comments

    I think this study, no matter how true, will be dismissed by the scientific community because it's politically incorrect.

    Ditto for racially-based studies and any research against the general American way of thinking "We're all born equal and you're fully responsible for your actions".

    Gerhard Adam
    Sorry, but I call bullshit. 

    First problem:
    The effect does not seem to apply to women, probably because they are "not subject to the same selection pressures,"
    Are we to conclude that women never cheat in anything?  If not, then what is the basis for the correlation in men?

    Second problem:
    Just 18 out of the 192 students lied during the negotiations, and of the 115 men in the study, only 13 lied.
    What does this mean?  What of the remaining six students that lied (out of presumably 77 women)?

    This doesn't even begin to address the problems of how ethics and behavior are taught and how one controls for the variations in people's attitudes. 

    Statements like this are also troublesome.
    ...only high WHR in men emerged as a predictor of cheating.
    Predictor?  In what sense?  If only 11% of the men actually cheated then how does one differentiate "prediction" from coincidence?  Why not brown eyes?  Long hair?  If the only criteria seems to be that I pick some trait and then correlate it to some behavior, does that establish an actual relationship?

    For example, in the study, there doesn't seem to be any indication of how many men did not have a high WHR (note: I didn't see the paper since it was subscription).  Did any of them cheat?  Were all the men high WHR?  If so, then the problem is obvious. 

    It simply seems that the researchers were looking for some testosterone related angle that they could exploit and this simply sounds like nonsense.

    One of the primary problems is that the focus invariably tries to establish some genetic component to behavior for which there is virtually no evidence.  This is easily demonstrated in other animals, where one can take a breed of dog (for example) that is genetically selected for a particular trait (aggressiveness) and yet they can be trained to behave differently.  This strongly suggests that whatever intrinsic disposition exists can be modified with training and learning.  I'm also a bit put off by the contrived notion of the "alpha male" in humans.  Since that isn't how alpha's actually operate in other species, this seems like a quaint rationalization for people that look to explain their behaviors as being somehow more advantaged than others. 
    Bonny Bonobo alias Brat
    Are we to conclude that women never cheat in anything?  If not, then what is the basis for the correlation in men?
    WHAT???

    Make love not war
    Gerhard Adam
    Simply that if if 11% of the men lied (to which high WHR is being attributed) but also 7-8% of women cheated (to which nothing is attributed), then why should we believe any correlation on the first part?
    Hank
    I reposted this with today's date - you must have saved it in draft mode and not changed the publication date because it's awesome and there is no way I missed this for 4 days.    If you use the scheduling option you don't have to change the publication date, it will be whatever date you pick for the article to publish on.
    Want more no-nonsense, independent science? Buy Science Left Behind
    First, the article is behind a pay wall, so I didn’t read it.

    Second, from your newscientist link, this quote:
    "When they compete for resources with other men, relative facial width is a strong sign of aggressive, self-interested behavior," he says. This means that other men defer to the "wide-faced", giving them more outright power, providing them with opportunities to abuse it through unethical behavior.

    It’s pretty obvious to me that whoever made the statement in the quote has never played medium-high stakes poker. Loads of aggression all around the table while the players are of every facial (and sexual) description.

    Third, whenever I see ambiguous terms such as ethics and morals used in the description of a supposed scientific topic, my BS detector fairly screams.

    Overall score == B.S. (and that ain’t Baloney Skins either)

    Hank
    Hey, atavism is a legitimate biology hypothesis - if it's 1880.   The founder of Cold Spring Harbor Laboratory said immigrants from southeastern Europe were "given to crimes of larceny, kidnapping, assault, murder, rape and sex immorality" which he was able to deduce by their appearance.

    I am glad science has escaped the miasma of the mid-2000s where this sort of topic would have been politically incorrect, even to debunk.    It creeps up every two decades or so because studies show some kind of heritable violence trait, it just has never been shown to be deterministic.
    Want more no-nonsense, independent science? Buy Science Left Behind
    we can actually trust everyone - the question is to define trust them to do what?

    and that is to act consistently with whatever behaviors they engage in

    so you can trust an aggressive person to be aggressive - you just shouldn't expect that they will act compassionatly or with consideration if they have never had a history to do so

    I find that the word trust is used to convey a meaning other than rely upon to be a certain way - as if trust is to beleive that a person will act in a way that we want them to, without consideration of whether the behaviour was a norm for them

    Gerhard Adam
    I think you're being a bit liberal in your definition of trust.  Certainly if you want to confine it to the narrow definition of "confidence", then your statement is generally correct.  However, trust usually involves a much broader definition relating to an implicit understanding that the individual being "trusted" will not take advantage even though you may be vulnerable.

    In short, trust generally refers to the absence of betrayal. 
    so you can trust an aggressive person to be aggressive - you just shouldn't expect that they will act compassionatly or with consideration if they have never had a history to do so
    Not true (even with your general disclaimer of "having a history of").  There may be many instances of where an individual is quite aggressive, and yet they may very well act in accordance with compassion or consideration towards teammates or comrades despite being aggressive to outsiders.  Your use of the term implies not just consistency or expectation, but almost a "hard-wired" response, which simply doesn't exist.
    ...and that is to act consistently with whatever behaviors they engage in...
    ... and what animal does that???

    In short, you cannot make any claims of such "trust" without defining the relationship of the individuals you're expecting to have interact. 
    having recently had much of my trust betrayal and undermined - I have taken a new meaning for trust - and it's basically like having faith the sun will continue to rise and set.

    I trust that people will behave in a manner consistent with their past actions

    if those actions are bad, then that's what I trust them to continue to do.

    I don't understand how trust came to ever include the concept that the person had to behave in a particular positive way.

    we deny people parole because we trust based on past behaviour and resistance to rehabilitation that they will continue to act in an anti-social manner.

    Gerhard Adam
    Sorry, but subverting definitions for your own personal reasons, simply isn't a very good reason.
    I don't understand how trust came to ever include the concept that the person had to behave in a particular positive way.
    That's just your cynicism showing and isn't particularly relevant.
    There's probably something to this study. If increased levels of testosterone cause both fat faces and a tendency to cheat, then by looking at someone's face, one could guess if they are likelier than normal to cheat.

    In these sorts of studies, just like in epidemiology, it's hard to separate correlation from causation due to confounding factors. But, the authors appear to have a reasonable hypothesis explaining all this.

    Gerhard Adam
    ...a tendency to cheat
    Sorry, but I don't count 11% males a "tendency", and to discount females renders it almost completely without explanatory power.  After all, if females are capable of cheating without a "testosterone" hypothesis, then why not consider that the same motivations might be applicable to males? 


    You're not following the epidemiological logic behind the study. If one group has an 11% likelihood of cheating and another group has 4% (or whatever the numbers actually are), that's probably a statistically significant difference and also probably worth reporting. All it means is that men with fat faces are 3 times LIKELIER to cheat than thin-faced men. It certainly does not mean all men with fat faces are cheaters and liars, and it's not even all that predictive, either.

    Your point that it does not reproduce in women is probably a fair point, but then their testosterone levels are a lot lower than men. A natural next step in this line of investigation would be to test the blood of cheaters and non-cheaters to see if cheaters have more testosterone. If they don't, then their hypothesis is wrong.

    Gerhard Adam
    Sorry, but comments like the following from the article leave me baffled:
    The researchers determined that unethical behavior in the exercise could be predicted by the WHR of men, but not women, although women also cheated.
    In addition, the difference in percentages was 11% for the men and 6.5% for the women.  Of course, since they didn't balance the genders, we can't actually know if these percentages would've changed with more subjects.  So out of 192 people, we have about 9.4% that will cheat, however if they're women, we don't have an explanation, but if they're men, it must be testosterone.  However, they also admit that men with low WHR's cheated, so it appears that out of the total percentage that actually cheated (men and women), it was an even match between those with high WHR's and men with low WHR's (and women).

    This statement from the article simply sounds like evolutionary psychology bullshit:
    The effect does not seem to apply to women, probably because they are "not subject to the same selection pressures," he said. "Men with relatively wider faces are more aggressive and self-interested, which allows them to secure a greater share of resources when competing with other men."
    This quote sounds almost like a modern form rationalizing the divine right of kings:
    ...which "found that men with larger WHR's are better leaders. Specifically, the facial structure of Fortune 500 CEOs predicts firm financial performance, such that CEOs with relatively wider faces achieve greater financial success for their firm."
    All in all, it simply seems that there are far too many agenda-driven conclusions being made for something that has such a weak correlation with no back-up data.  Just the simple fact that they assume that being a CEO "predicts firm financial performance" is so obviously filled with bias, I can't imagine anyone silly enough to have made the statement.

    I'm also not clear on what this kind of statement is supposed to mean:
    ...he says, "have captured a genetic component of unethical behavior and also nicely demonstrate that it is driven through a psychological variable: a sense of power."
    In one simple study they've managed to connect genetics with behavior and psychology.  A "genetic component of unethical behavior"?    .... Really?  So they're claiming they found the "cheating" gene?

    In general, this statement from the abstract should be sending up red flags everywhere:
    ...provide additional support for the view that evolutionary forces shape ethical judgement and behaviour.
    Such a statement is simply stupid.
    Just a comment on photo from The Hound of the Baskervilles (1939):
    The photo shows (left to right):
    Dr Mortimer (back to the audience - Lionel Atwill)
    Sherlock Holmes (Basil Rathbone)
    Dr. Watson (Nigel Bruce)
    Sir Henry Baskerville (Richard Greene)

    Hank
    You should also clarify it is the best Sherlock Holmes story ever.
    Want more no-nonsense, independent science? Buy Science Left Behind