Bygone brunette beauty: Fashion in hair color

Share on FacebookShare on Google+Email this to someoneTweet about this on Twitter

Long-time readers know that one of my beliefs that I’ll stop at nothing to prove is that blond women are not sexier than brunettes, whatever other appeal they obviously have for many men. Point-estimates of the current popularity of blond hair neglect the fact that standards of beauty can change over time — within boundaries, to be sure, but still. Perhaps we only live in a blond-obsessed world today, while brunettes may have ruled in the past. Indeed, I will show just that. Furthermore, the shift toward blonds parallels several other shifts toward a more masculine ideal of female beauty since the early / mid-1960s.

The data come from Playboy Playmates of the Month (“Playmates”) from 1954 to 2007. We need to look at sources that pander to popular demand in sexual tastes, which excludes runway fashion magazines (not used by males for fantasy purposes) as well as data on high-ranking Hollywood actresses (who are esteemed only in part based on their looks). We also need comparable data that stretch over decades, and that provide us with many data-points for each year — in a worst case scenario, we might look at something like Miss Universe winners, but estimating the value of blond-obsession for a given year with only a single data-point is hardly ideal. Playmates, though, yield 12 data-points per year.

In the name of scientific discovery, I looked at pictures of every Playmate [1], and coded her hair color as either 1 for blond or 0 for non-blond. Dark blonds counted as blond, light browns as non-blond. Redheads counted as blond if they had very fair, strawberry blond hair, and as non-blond otherwise. The point is not to measure the popularity of the full spectrum of hair colors — just blondness. A small handful of Playmates had several hair colors within the single issue they appeared in. I coded these as 0.5 because their pictures were split pretty evenly between blond and non-blond hair — maybe due to wigs, I don’t know.

I then took the fraction of blonds in a given year and plotted these over time. Here is the raw scatter-plot, together with 3-year and 7-year moving averages that smooth it out:

The scatter-plot suggests an increasing trend, and this is true: Kendall’s tau for the correlation between year and percent blond is +0.27 (p = 0.01, two-tailed). [2] However, because each year’s value can only take on roughly 12 values (1 / 12, 2 / 12, etc.), there are a lot of tied years, which may underestimate the true correlation. Kendall’s tau for the correlation between year and the 3-year moving average of percent blond is +0.47 (p = 2 x 10^(-6), two-tailed), and is +0.64 (p = 2 x 10^(-10), two-tailed) when the 7-year moving averages are used. Using a moving average gives us a better idea since they can take on far many more values, and so produce fewer ties.

Whichever one we choose, it is clear that blonds have increased quite a bit in popularity over the decades. At the same time, the trend is clearly not linear: there is a decrease in blond-obsession at least from the mid-1950s, when the data begin, to the early / mid-1960s. There follows an increase, and an apparent reversal since the turn of the millennium — please god, let it be so. This looks periodic, like a fashion cycle.

In trying to account for this trend, we should try to be as general as possible. What other trends in female beauty show an increase after the early 1960s? I didn’t look at other aspects of the Playmates, but someone else has tabulated data on Playmates of the Year from 1960 to 2006 — again, estimating the popularity of some trait in a given year based on a single data-point is worst-case, but I’m relying on it here because I’ve already spent enough time collecting hair color data. The links in footnote 1 provide all the anthropometric data, though, so if you want to collect an analyze it, we will link to your analysis.

I calculated the Waist-to-Hip Ratio and BMI of Playmates of the Year from the above data, and Kendall’s tau for the correlation between year and WHR is +0.53 (p = 4 x 10^(-7), two-tailed), while between year and BMI it is -0.24 (p = 0.02, two-tailed). So, these sex symbols are increasingly losing their feminine hourglass shape and fatty softness — nearly all BMI points are below 20, so it’s not like they used to be fat but are now healthy. They are also getting taller: Kendall’s tau is +0.31 (p = 0.004); and smaller in the chest: Kendall’s tau is -0.35 (p = 0.001).

Someone else has also done an analysis of Miss America winners, and the exact same trends emerge there as well (see his graphs).

The common factor of all these trends is that the ideal of female beauty has become increasingly masculinized. Recall that males are more likely to be blond, so the hair color trend is part of the larger masculinizing trend. I didn’t look at eye color, but if it’s part of the overall trend, the earlier Playmates should be less blue-eyed than later ones, as blue eyes are also more typical of males. Skin color would be tougher to analyze; if it’s part of the same trend, it should get darker over time. Anecdotally, these two guesses seem to be true, but someone should look at the data to check.

It therefore appears that a preference for blonds should also correlate with a preference for taller and less curvy women. Again, someone else can look that up in the psychology literature and post in the comments. But the words “tall,” “thin,” and “blond” usually co-occur, don’t they? Whatever appeal such women have, raw sex appeal is unlikely to be among the top reasons. Blond hair correlates with something like introversion, and that makes sense since men on average are more introverted than women. So, maybe guys start digging blonds when they become more marriage-minded, or if they are inveterate monogamists. A blond will be less likely to be bouncing off the walls and being constantly out and about in search of social stimulation.

Bang a brunette, bank on a blond? It would fit with the trend toward lower sluttiness in recent times, which we expect to weed out the sex kitten types from popular culture. This suggests that dark hair is part of that highly sexualized image — something that was always obvious to everyone but the blond-lovers.

[1] For years 1954 to 1992, I used this source that contains the full shoot for each Playmate, and for 1993 to 2007, Playboy’s official website (if the single picture available on the Playboy site was ambiguous, I did a Google image search to get a better idea).

[2] You can easily calculate Kendall’s tau with this website, which I used here.

Labels: ,

62 Comments

  1. I wonder how the same biometrics of the average American woman have changed in the same time period?

  2. assman, did you photoshop out the mustache?

  3. I wonder how the same biometrics of the average American woman have changed in the same time period? 
     
    That could account for the slight increase in height, but it’s not as if women have become more beanpole-like in shape, or that we’ve been invaded by blonds. 
     
    assman, did you photoshop out the mustache? 
     
    Her name is Nancy Jo Hooper, and she’s from Alabama, so little chance she’s from a mustachioed race. Even the Anglos back then were pretty nice.

  4. Maybe men just like those that stand out. Being blond may be sexy in the United States but in Germany it may be boring. Those that stand out may be best.

  5. This girl is much better looking than the generic Slavic blondes who dominate the tennis beauty ranks.

  6. Maybe men just like those that stand out. 
     
    This doesn’t make sense unless they stand out in a good way, though, right? What counts as “good”? Signals of health and femininity, basically. But recent decades have seen a rise in the popularity of females who stand out for being more masculine — blond hair just being one aspect of this. 
     
    This girl is much better looking than the generic Slavic blondes who dominate the tennis beauty ranks. 
     
    Ah, that’s because she’s a Southern Slav! Yes, they are incredibly hotter than the Eastern Slavs, with Western Slavs being in between.

  7. Also, re: standing out — if you’re positing that as an unchanging reason, it cannot account for the drastic increase in blond-obsession.

  8. If any psychology grad students are reading this, you could easily test what hair color is the most sexually appealing: 
     
    Get one of those ultra-hot, doesn’t really exist, composite faces from an already done study. Keep the geometry the same, but alter her hair, eye, and skin color. Hook the male subjects up to one of those devices that measures how engorged their junk is, and see if they get more sexually excited from different hair, eye, or skin colors. If not that, at least something like their heartbeat — something involuntary and physiological, rather than a self-report (maybe do that too to see if there’s a discrepancy). 
     
    You should also do this with the prototypically ugly face. That way, you can see whether a certain hair, skin, or eye color is more attractive regardless of what their face looks like, or whether it acts as an amplifier — e.g., if dark hair makes beauties look even more breathtaking but makes uglies look even more repulsive.

  9. Thanks. I didn’t realize there was a difference between Slavs. But I’m no expert. 
     
    I think your study idea is a good one. 
     
    As a woman, I just don’t understand the hysteria over blonde women. I think natural blonde hair is beautiful, but a nice thick head of brunette ahir with all sorts of glints in it is to me much more beautiful. Esp. combined w/blue or green eyes.

  10. It therefore appears that a preference for blonds should also correlate with a preference for taller and less curvy women. Again, someone else can look that up in the psychology literature and post in the comments. But the words “tall,” “thin,” and “blond” usually co-occur, don’t they? 
     
    If I understand you correctly, it sounds like you are describing a preference for blondes resulting in a changing from a Mesomorphic body type to a Ectomorphic body type. 
     
    (The mesomorphic body type being, for women, the curvy hourglass figure. And the ectomorphic body type being, for women, “thin” and much much less curvy figure.) 
     
    – Charles Iliya Krempeaux 
        http://changelog.ca/

  11. This talk of blondes and brunettes is Eurocentric. I would have added another category (or several) for non-Europeans. I don’t know how big a fraction they comprise but I know there have been some among the PlPl ranks.

  12. question: is lighter skin considered more attractive on average? whence tanning? personally i think the best looking skin tone on a chick is light olive.. am i an outlier or is that a common preference?

  13. Almost certainly there has been another hair-related change in Playmate appearance over the years. And it’s far, far worse :((((

  14. Hook the male subjects up to one of those devices that measures how engorged their junk is, and see if they get more sexually excited from different hair, eye, or skin colors. If not that, at least something like their heartbeat 
     
    They haven’t done this, but men already subjectively rate the same face as more attractive when it is paired with blond hair. 
     
    This and a number of studies were linked in the comments to this post. 
     
    Blond waitresses get higher tips, blonds get more attention on dating sites, and blonds are rated more attractive by (white) men in a number of studies. Blonds are also rated as more sexual and more promiscuous, and they may actually be overrepresented among prostitutes.

  15. Yeah, but that’s all a snapshot of right now — the Playboy data show clearly that blonds are really popular now, but ’twasn’t always so. 
     
    I guess that study design might not resolve the matter, then.

  16. diana has alluded to this before, but to some extent i think today that non-blonde european women who are very good-looking are using the handicap principle. that is, they don’t need the bottle to improve (soften) their looks because they’re so smokin’ that there are no marginal returns. OTOH, a 5 can easily improve their looks by a dye job to a significant degree, or at least distract from the plainness of their features. the prediction here would be that in professions which place a premium on physical beauty all things controlled besides hair color the brunettes would have better faces because they’d have to have them to compete.

  17. “As a woman, I just don’t understand the hysteria over blonde women.” 
     
    As a man, I’ve never understood why women must cover their eyelashes with mascara. I actually don’t care about the color of hair on the head, but there’s nothing more attractive to me than blonde or red eyelashes. Which is unfortunate, since it was already a rare female phenotype from before, and it now seems to be on its way to extinction. Sissy Spacek would be the last actress that I can recall having lite eyelashes (except maybe for the actress that played the White Whitch in that ‘Narnia’ movie -but I’m guessing that she was supposed to look freakish).

  18. Hair dye has improved over the decades — Jean Harlow destroyed her hair dying it blonde in the 1930s — so it’s easier to be blonde today. So now it’s become mostly a question of whether you want men to notice you instantly.

  19. today that non-blonde european women who are very good-looking are using the handicap principle. 
     
    The handicap principle requires that the signal (here, non-blond hair) be costly. Today, where there’s a premium on blondness, that may be true, but it wouldn’t have been true before the mid-1970s, when brunettes ruled. So as a general principle, it doesn’t work. 
     
    they don’t need the bottle to improve (soften) their looks because they’re so smokin’ that there are no marginal returns. 
     
    Ah, that’s not a handicap then — if they don’t gain by switching to blond, that means they’re equally attractive as a brunette, i.e. their attractiveness does not suffer by having dark hair = dark hair is not costly.

  20. Hair dye has improved over the decades — Jean Harlow destroyed her hair dying it blonde in the 1930s — so it’s easier to be blonde today. 
     
    That’s a non-starter, though, since Playboy and Miss America would’ve simply selected pretty girls who were naturally blond in previous decades — it’s not as if there weren’t any to choose from. 
     
    This account would only make sense in Japan or India. In a population of mostly northwestern Europeans, light blonds and dark blonds make up around 40-45%.

  21. And were blond wigs so unconvincing in the 1950s and 1960s, when blonds were not very popular in Playboy? (To me, all hair from that era looks a bit fake and processed.) These Playmates were not celebrities who would’ve had to maintain the wig at all times — just throw it on for a photo shoot. That wouldn’t have destroyed their hair or been an inconvenience at all.

  22. Perhaps blonde hair signals young? Children may have tow-head hair, then become brown haired as adults. 
     
    Thus, blonde hair might have that advantage of signalling youth. 
     
    Other aspects might be standing out from other women, see also red hair.

  23. I’m not a statistician, but is it valid to take a correlation with a moving average? By definition, the moving average throws away some of the scatter, so the correlation will be likely to be larger. And you’ve already thrown away a lot of the scatter by taking yearly averages. 
     
    FWIW, I generated a 0/1 random variable for each month in the range and carried out the same analysis as agnostic to obtain Kendall’s tau for the raw data and 3 and 7 year moving averages. Correlations appeared to increase with the size of the averaging interval, and p values decreased (although most were above 0.1). Here are the scatterplots for the first two runs: 
     
    Run 1
     
    Run 2
     
    By definition, there is no trend in these datasets, yet spurious patterns can clearly be seen in the 7-year averaged scatterplots, even more clearly than for the Playmate data.

  24. Children may have tow-head hair, then become brown haired as adults. 
     
    Thus, blonde hair might have that advantage of signalling youth.
     
     
    Well, that’s a bit more than just “younger than you look.” No one wants to look like a pre-pubescent child — maybe a teenager at the earliest, but blondness has started to erode by then. 
     
    Also, I didn’t mention it in this post, but will do so in a post tomorrow, but Playboy Playmates have grown steadily older since the mid-1960s. The increase in popularity of blonds tracks the increase in popularity of older females. 
     
    I’m not a statistician, but is it valid to take a correlation with a moving average?  
     
    I’m not one either, so I don’t know. I reported the correlation without using moving averages, though, and it is significant.

  25. I’m hesitant to use the raw month-by-month data-points since that doesn’t seem like the time-scale over which fashion in hair color would change. It would be like measuring one stranger’s skirt length each month for 10 years to see if skirts were becoming longer or shorter. It seems like the “pulse of the times” is closer to a year. 
     
    However, since the criticism was brought up, I broke apart the yearly data-points into the raw monthly data-points. Kendall’s tau is +0.07, p = 0.025 (two-tailed). Recall that I dichotomized the hair color variable, so it should be nearly impossible to detect a trend in 0s and 1s on a month-by-month basis. The fact that even my incredibly crude coding, measured on a probably unrealistic time-scale, still shows a significant positive correlation suggests that the pattern I refer to is real.

  26. For example, when the Playboy people are choosing each month’s Playmate, they probably have a rough target number of blonds for that year’s subscription, depending on the perceived demand for blonds. That’s one reason why I think it’s more realistic to look at the level of a year rather than the month.

  27. I don’t know why blond hair is so valued at present; but I note that it is specifically the blondness of hair that is valued – and not the whole naturally-blond women. i.e. the ideal is blond hair on a brunette’s face and body.  
     
    The ideal (in the US and Western Europe) seems to be blond hair on a natural brunette – virtually all the famous blonds from Marilyn Monroe onwards are peroxide blonds who were natural brunettes.  
     
    There are exceptions among Scandinavians, for example, but generally natural blonds lack well-defined eyes and eyebrows, sculpted nose, outlined full-but-firm lips, clear skin, and shapely and toned physique.  
     
    Indeed, many natural blonds are somewhat ‘syndromal’-looking, in the same sense that some natural pale-freckled-redheads look syndromal so that these redheads look more like each other than they look like their families. Perhaps natural blonds (outside of the Scandinavian focus) share a particular genetic mutation, in the way that redheads often do?  
     
    [NB: the first red hair gene was discovered by some friends of mine!  
    http://www.nature.com/ng/journal/v11/n3/abs/ng1195-328.html;jsessionid=666C2E763D32D300EB4DC1866C4BB333] 
     
    It doesn’t answer why blond hair should be valued in the first place, but I would guess that the increased popularity of blonds is mostly due to the huge improvement in natural-looking hair-lightening technology over the past 25 years.

  28. “diana has alluded to this before, but to some extent i think today that non-blonde european women who are very good-looking are using the handicap principle. that is, they don’t need the bottle to improve (soften) their looks because they’re so smokin’ that there are no marginal returns.” 
     
    With that in mind…Jason – here’s a thought – if men do tend to notice and favor blondes qua blondes – then nature has created a way to make brunettes stand out: by making them more beautiful! :):) 
     
    Chew on that. 
     
    (Maybe blonde waittresses get higher tips ’cause they evoke a protective (pitying?) instinct, whereas a man served by raven-trsssed food server feels….intimidated….just a thought.) 
     
    My point is that Agnostic is making a case here for intrinsic beauty, as opposed to fashion. I vigorously agree with him. 
     
    It’s impossible to create a truly objective panel of judges of beauty, so the issue will never be settled.  
     
    BCG,  
     
    Is the blonde craze really so prevalent in Western Europe? Any data on hair dying? French and Italian models and actresses (the most outstanding samples I can think of) don’t strike me as overstocked with blondes. They are overwhelmingly fine-featured brunettes (Tatou, Belluci, etc.)

  29. PS, totally anecdotal, but the dark-hair-with-light-eyes combination seems to attract a lot of attention. People always seems startled by the juxtaposition. I don’t know why. It’s not an uncommon combination (is it?).

  30. Agnostic, it’s not just the puberty that causes hair, skin, and nipples to darken—the hormonal changes associated with pregnancy may affect a woman’s hair follicles as well.

  31. but I would guess that the increased popularity of blonds is mostly due to the huge improvement in natural-looking hair-lightening technology over the past 25 years. 
     
    Here’s another reason why the “greater ease of dying your hair” argument is unpersuasive: think of all the time, money, and effort that women invest in improving their appearance. Make-up and other cosmetics, clothing, etc. Dying your hair requires no effort — you just sit there — only takes up a couple hours per month (in one sitting), and costs maybe $100 per month if you go someplace in the upper-middle price range. 
     
    If this had a noticeable effect, almost all women would do it — at least those who were middle income and above. Yet outside of Texas and southern California, you tend not to see brigades of blonds. 
     
    Compare this with removal of body hair (by whatever means), using moisturizer to firm the skin, exfoliants to clear away dead skin, teeth-whitening products, cosmetics that make the lips and eyes look larger, etc etc etc. Imagine if increasing your breast size, or keeping them the same size but firming them up to their peak value of defying gravity, required a similar investment of time, money, and energy (and was non-invasive, like dying your hair).

  32. diana, Star Trek DS9 had the Dax clan: pale dark-brunettes with very blue eyes. What that meant was six seasons of the tall, athletic Terry Farrell and then one season of the small, page-boyish Nicole de Boer. The Daxes were symbionts (“joined Trill”), who bore all the memories of previous Dax hosts. 
     
    The intention of casting these two women in that role was so to strike the audience with the contrast of blue eyes on a brunette. Sort of an “eyes as window to the soul” theme.

  33. “It’s not an uncommon combination (is it?).” 
     
    Non-brown eyes are uncommon worldwide, of course. It is the norm with black hair: China, Africa, Southern Europe, South America, Japan, the Middle East- that leaves northern Europe and the Slavic countries. Even there light eyes are more common with non-brunette hair, and people with black-brunette tend to have brown eyes. So rare enough worldwide, yes. So much so that people will mention it in these places, even guys will have it pointed out. I think that Brunettes with light eyes are really missing a trick here. Liv Tylor is the way to go.

  34. Missing a trick by dying their hair blonde, I mean.

  35. i don’t see how you could do any experiment at this point that wouldn’t be confounded by culture. 
     
    also, to the person who said that white people don’t frizz out their hair to look black – what about dreadlocks?

  36. It turns out the Playboy hair color analysis (along with other magazines) has already been published. Looking from 1950s-1980s they find blonds are overrepresented in all decades. (lowest during the 60s and highest during the 70s) 
     
    ——————- 
     
    if men do tend to notice and favor blondes qua blondes – then nature has created a way to make brunettes stand out: by making them more beautiful! 
     
    This kind of pleiotropy is unlikely, and the preferences are doubtfully this stable. For instance, Peter Frost referenced an important study in the last thread showing how hair color preferences shift in response to novelty.

  37. the dark-hair-with-light-eyes combination seems to attract a lot of attention. 
     
    This color combination is also correlated with mental illness.

  38. Wow, that is quite amazing. 
     
    Why do you think this is? Or rather, how did it come about? Do you think it’s also correlated with other personality traits? Intellect, etc.? Any ethnic relationship??

  39. Looking from 1950s-1980s they find blonds are overrepresented in all decades. (lowest during the 60s and highest during the 70s) 
     
    Does anyone have access to this article? It claims to show the same temporal trend that I showed here — low popularity during the 1960s, then a rapid shift to high popularity during the 1970s. But what I’m curious about is what they think the baseline level of blondness is in the population. 
     
    Returning to data on the prevalence of hair types in northwestern Europeans: 
     
    http://www.gnxp.com/blog/2007/11/brown-eyed-girl.php 
     
    Take the blonds and half of those in the “dark blond / light brown group” (since I put dark blonds with blonds), and you get 40 – 45%. Then go to my graph of percent blond by year. No way are the 1960s above 40 – 45% — they’re either within that range, or a bit below. Only after 1980 do the data clearly exceed 40 – 45%

  40. correlated with mental illness.  
     
    seemingly debunked 
     
    http://psycnet.apa.org/index.cfm?fa=search.displayRecord&uid=1979-33764-001 
     
    D. B. Cohen (see record 1979-23735-001) reported that dark-haired/light-eyed college females may be more susceptible to psychopathology than other groups. Data from a similar survey in an inpatient psychiatric hospital fail to confirm his findings; there was no significant difference between male and female distributions, nor was there an obvious skewing of the color combinations in favor of the dark/light combination. (1 ref) (PsycINFO Database Record (c) 2007 APA, all rights reserved)

  41. Good catch. Google Scholar didn’t link to the follow-up citation.

  42. seemingly debunked… 
     
    Too bad, just when I though Diana had Dark Triad Skillz

  43. Phew, I’m relieved. OTOH, Dark Triad Skillz sounds interesting…..

  44. But what I’m curious about is what they think the baseline level of blondness is in the population… Then go to my graph of percent blond by year. No way are the 1960s above 40 – 45% 
     
    This is an important question, and you’d think the answer would be easy to find. The authors state they couldn’t find this data, so they construct their own norm sample from undergraduates. The rates are 68.1% brunette, 26.8% blond, and 5.1% red. So blonds are overrepresented in every decade, with Playboy featuring more overrepresentation than the women’s magazines. 
     
    Another study on blond CEOs in the UK faces a similar problem as the last study. The authors assert that there is hair color data for the UK on the CIA Factbook website. I simply can’t find this kind of data anywhere at that website. They say this CIA data is supported by another study… of American undergraduates?? 
     
    Anyway their numbers are 68% brunette, 25% blond, 1% red, and 6% black. So pretty much the same as the other undergraduate study. 
     
    The supporting study also shows that blond hair and blue eyes often go together. This suggests (natural) blond hair has been declining even as it is has become more popular, consistent with the Frost study linked above. 
     
    “About half of Americans born at the turn of the century had blue eyes, according to a 2002 Loyola University study in Chicago [Actually 57.4% – JM]. By mid-century that number had dropped to a third. Today only about 1 out of every 6 Americans has blue eyes…” 
     
    Quote. Study.

  45. Maybe there has been a change in Playboy’s philosophy. Wasn’t the original idea to feature the “girl next door” — i.e., a normal, healthy-looking non-stripper. Perhaps over time the philosophy has evolved toward featuring fantasy figures.

  46. Steve, 
     
    Or maybe today’s “girl next door” is a dyed blonde :)

  47. Perhaps over time the philosophy has evolved toward featuring fantasy figures. 
     
    Men these days must have some pretty androgynous fantasies, then, since sex symbols have become less hourglass in shape and smaller in the chest (pictures of the ’50s and ’60s Playmates are pretty eye-opening). Angelina Jolie is pretty, but she has a pretty masculine skull — and even more masculine personality. 
     
    The authors state they couldn’t find this data, so they construct their own norm sample from undergraduates. 
     
    Their article is from 1993, and if blue eyes are steadily dwindling from 55% to 34% to 17% (early 20th C, mid-20th, late 20th), which the authors say is probably due to less assortative mating by ethnicity, then blond hair must be following a similar trend — and hence, the prevalence of blonds in the 1950s and ’60s would’ve been noticeably higher than the 27% estimate from 1993. 
     
    It’s not unreasonable to project backward and get 40 – 45% blond at mid-20th C., as per the estimates in the “brown eyed girl” post. Those data are from samples with over 1000 people, rather than a smaller convenience sample.

  48. Wow… I am seriously disappointed in physical anthropology. This is really the best information we’ve been passed down on hair and eye colors? Poorly sourced second and third-hand maps with mysteriously obtained and quantified information? 
     
    Why are we left with no real data? How hard is it to count hair colors?? Not one anthropologist cared? This is pathetic.

  49. jason, that’s why you have to look to the victorians. people like galton actually were interested in counting….

  50. What about the decline in the social stigma of cosmetics/hair dying? My grandmother remarked to me that during her youth (1930s-1940s) the bottle blonde was seen as something of a tart, and the practise of lightening hair was strongly associated with suspect catagories: Actresses, prostitutes, etc. Even now, with hair lightening a much more socially acceptible practise, extreme dye-jobs carry with them lower class connotations.

  51. Someone suggest blonds are “syndromal”? What about Tiny Tim? Didn’t he have some kind of brunette syndrome?  
    http://www.youtube.com/watch?v=skU-jBFzXl0 
     
    I often see people here and there, totally unrelated, who share a certain “look”. I see people who sort of look like me, and find they think sort of think like me too. 
    It would be fascinating to trace one of these “looks” back to their owners’ early genetic pool. They might be examples of extended families within the “extended family” that is race/ethnicity. 
    Despite a bit of bias towards the mediteranean type, I have to admit my niece and nephew are blond and looked like the proverbial angels when they were little. 
    Nothing “syndromal” about their looks, but then again, maybe there is. The rare “sunny blonds with natural peachy tans, full lips, etched eyebrows and dusky lashes syndrome.” I think that’s the one Hollywood had in mind. 
     
    Nevertheless, my niece disliked her blond hair and threatened to dye it red or black, but piercing her finely etched eyebrows with metal rings distracted her from that ambition.  
    Around 14 the once chubby tow head startled everyone by becoming a Boticelli beauty resembling the girl who disappeared forever in Picnic at Hanging Rock. Yet still she doesn’t think she’s pretty.  
    I blame the 1965 Immigration Act. Just kidding. 
     
    I’m sort of a blond convert though. Blonds are not intrinsically more beautiful than brunettesand arguing is pretty pointless. This debate goes back centuries, but somehow those blond wigs won’t go away.

  52. “My grandmother remarked to me that during her youth (1930s-1940s) the bottle blonde was seen as something of a tart” 
     
    Because they were? If something is socially proscribed then those who do it tend to be non-conformists.  
     
    Even the phrase “bottle blonde” has a disapproving ring.

  53. I know that dyed blonde hair is big in Eastern Europe, and here in New York pretty much every Greek woman I’ve seen has dyed blonde hair – or as the Aussies like to call it, “suicide blonde”

  54. It is difficult to get good anthropological data on hair color for a number of reasons: 
     
    1. Hair color varies continuously, so it is difficult, for instance, to draw the line between ‘blonde’ and ‘light brown’. 
     
    2. Many people dye their hair, particularly those who are prematurely gray. Such people may not always tell the investigator. 
     
    3. Hair color changes with age. At 1 year of age, 50% of male white children from Louisville, Kentucky had light hair. At 6 years of age, only 26% of the same sample had light hair. (Matheny & Dolan, Am. J. Phys. Anthrop. 42: 53-56). Hair color continues to darken from 6 to 18 years of age (Steggerda, J. Heredity, 32: 402-403). This darkening seems to proceed faster in males than in females. 
     
    4. Physical anthropologists have shunned this field of research since WWII. This is partly because it is seen as having Nazi connotations and partly because physical anthropologists became convinced that genetic markers (e.g., blood groups, enzymes, etc.) provide a more scientific picture of human variability. 
     
    On the other hand, I’ve seen a number of studies that peg the percentage of blondes among ‘old-stock’ Americans and English peoples at around 20-25%. Since this ballpark figure comes up so often, there must be something to it.

  55. Yes, that figure is consistent with the newer numbers above (from 1974 and 1993) and the earlier hair data in John Beddoe’s The Races of Britain (1885). 
     
    The early 20th century blue eye figure I cited above is not consistent with what we should expect from old stock Americans, and such a decline is not plausibly a result of Italian immigration (the only immigrant population significantly different in eye color). The numbers are suspect. 
     
    I think this suggests blonds have been favored and overrepresented in media since at least the 50s.

  56. The early 20th century blue eye figure I cited above is not consistent with what we should expect from old stock Americans, and such a decline is not plausibly a result of Italian immigration (the only immigrant population significantly different in eye color). The numbers are suspect. 
     
    Suspect meaning you think it should be higher or lower? In Icelanders, 80% of men and 70% of women have blue eyes. In the Dutch, it’s 70% and 52%. 
     
    Since blue eyes are mostly a recessive trait, less assortative mating will preserve the allele frequencies but diminish the genotype frequencies. The authors of the study you quoted suggest just that: people assort less on appearance than they used to, and more on education, social status, etc. Again, that’ll affect the prevalence of blond hair by the same logic.

  57. Since blue eyes are mostly a recessive trait, less assortative mating will preserve the allele frequencies but diminish the genotype frequencies.  
     
    assman, this would be more persuasive if consanginuity among americans around 1900 were like among many arab groups. but that’s no so. i mean, that would imply that across america there were blue-eyed towns and brown-eyed towns. but again, this is a new country and though there are cases of whole villages transferring en masse (i’m thinking of instances where a village in norway simply relocated over a generation to minnesota town, etc.), usually towns were founded by disparate groups of newcomers not closely related to each other. compared to other pre-20th century peoples my understanding is that americans were especially mobile (the frontier and all). the model doesn’t pass the smell test, it seems that it is way more likely that it’s an artifact of measurement or something.

  58. p.s. australians are not that different from the dutch and icelanders from what i recall, and there you have a population of roughly the same geographic origin as the USA, but fewer germans and more british isles ancestry, with a non-trivial but minority southern european quanta.

  59. Miss Maxim shows only 1/6 blonds, even though about 1/2 of the countries could easily find blonds. 
     
    http://www.maximonline.com/miss_maxim/ 
     
    I’ve looked at the “Maxim Hot 100″ list before, and so has Audacious Epigone, but we’re probably wrong to do that. We should look at the Maxim counterpart of the “centerfold girl,” like these girls.

  60. Blonde hair is shinier, so it catches the eye more. It’s the same reason that most women’s jewelry is gold or silver.  
     
    Women who want to be noticed tend to dye their hair the most noticeable color.

  61. I just did an analysis of all the girls who’ve appeared on the cover of Maxim from 1997 to present, and will post more detail soon, but the average is 2.8, where 1 = light blond, 2 = dark blond, 3 = light brown, and 4 = dark brown. So, women who want to be noticed today are light brown, not blond. 
     
    There is no trend over time in Maxim as in Playboy, though Maxim’s only been around 10 years. 
     
    Maxim targets younger audiences, while IIRC the average Playboy reader is 33. So again, we likely have a co-hort effect — readers brought up during the heyday of blonds want more blonds, while those brought up as blond popularity was plateauing want darker hair. 
     
    As far as overrepresentation goes, light blonds are about 40% more common than in the general population, but dark browns are about 60% more common. It looks like the middle is wiped out in favor of emphasizing the two extremes, which appeal to blond-lovers and brunette-lovers. Evidently, among Maxim’s target audience, there are more brunette-lovers.

  62. Steve, 
     
    “Blonde hair is shinier, so it catches the eye more.” 
     
    I’m not going to go all gnxp on you and say, “cite the source.” I’ll just say I disagree.  
     
    But I do agree that the main component of beautiful hair IS shininess. If you have dull hair, no matter how thick & silky, it is ugly. 
     
    I have seen dull blonde hair (“dishwater blond”); dull brown hair but never dull black hair.  
     
    The thing about shiny blonde hair as opposed to shiny dark hair is that it’s unusual. And anything unsual catches the eye more.  
     
    BTW I just got an “e-lert” from the Estee Lauder company advertising that their four models (Hurley, Paltrow and two others whose names I forget) are coming to Bloomingdales NYC at the same time.  
     
    ALL FOUR have blue or green eyes. Two blondes, two brunettes. So, heavy overrep of blondes and completely unrep eye color! 
     
    But a very unique and small sample size….

a