Did Boys Use To Wear Pink? Revisited

By Neuroskeptic | July 2, 2017 5:37 am

Five years ago I blogged about the debate over whether the blue-for-boys, pink-for-girls color convention used to be the other way around. My post focused on a 2012 paper by psychologist Marco Del Giudice arguing that the idea of a cultural “pink–blue reversal” in the English-speaking world in the early 20th century is a myth.

Now, Del Giudice has published an ‘update’ revisiting the issue. Based on text data from late 19th and early 20th century American newspapers and magazines, Del Giudice says that there’s still no evidence that pink-for-boys, blue-for-girls was ever the dominant pattern.

Del Giudice admits however that the periodicals do reveal that in the period before 1920, both blue-for-boys, pink-for-girls and the reverse scheme were mentioned, roughly equally often. In other words, the genders of these colors were inconsistent before this point.


However, American books from the same period (and afterwards) overwhelmingly mentioned blue-for-boys, pink-for-girls. There were negligible mentions of the reverse scheme.


British books showed a similar pattern. Del Giudice says that he’s not sure how to interpret this discrepancy between books and newspapers/magazines. I wonder if it reflects differences in the demographics of periodical vs. book authors during this period?

Del Giudice concludes that overall, there is still no evidence for a “pink–blue reversal”, although the situation before 1920 may have been somewhat complex. This seems sensible to me.

In the second part of the new paper, Del Giudice discusses cross-cultural evidence of a female preference for pinker hues, although he acknowledges that the evidence on this topic mostly comes from small studies.

Del Giudice goes on to review possible evolutionary explanations of a sex-based color preference, though he says that these hypotheses are “rather speculative and may well be incorrect.” I have to say that they do seem like just-so stories to me, taking a historically contingent fact and working backwards to explain how it plays a biological function:

The color red is implicated in sexual choice in many animal species; in humans, skin redness is a cue of health and attractiveness and – intriguingly – has a sexually dimorphic distribution, with men showing higher average levels of redness than women… This raises the possibility that women may be more sensitive than men to skin redness in potential sexual partners…

The skin of newborns and infants also shows elevated redness compared with that of adults and takes a pink coloration in areas that are especially rich in capillary loops (e.g.,cheeks). Preferences for red and pink may be partly linked to female preferences for babies.

  • smut clyde

    What I liked about these two images is that they’re on the same wall at the Wallraf-Richartz Museum in Cologne.
    1. Renoir (1898), “The Artist’s Son”
    2. Cassatt (1901). “The Artist’s Daughter”.

    • Erik Bosma

      Well, he looks like an artist’s son alright.

      • smut clyde

        Jean went on to make some damn fine movies. So the pink smock didn’t harm his development too badly.

  • Michael Pettit

    Unfortunately, this study (like the original) is rubbish. As a historian, I am quite familiar with Chronicling America. It is an amazing resource, with uneven coverage, and some glitchy OCR. According to the article itself, the search term “blue pink boy girl” appeared all of 62 times (27 unique times) in over 12 million digitized pages over a hundred year span. This is statistical noise. To put it in perspective, the now obscure psychologist “Joseph Jastrow” gets 251 hits in the same time. Americans newspapers cared more about him than the colors associated with babies. I would be wary about making any inferences about a medium like newspapers (let alone a culture or a species) based on this meager data. I am kind of shocked a peer-reviewed journal accepted this (again).

    There may or may not have been a reversal in the color associated with different genders, but the methodology used was an inappropriate one to detect it. Indeed, the quotations from the sources the author cites demonstrate a high level of ambivalence and indifference, that the colors associated with different genders wasn’t fixed or certain.

    I tried to warn behavioral scientists about the challenges of using this kinds of corpus, alas to no avail.

  • Asia Jędrzejewska-Szmek

    I was under the impression that color reversal happened mostly in catholic countries, where girls had been wearing blue, because blue is the color of Virgin Mary (or one of the colors). (But it’s a home grown ethnological theory, not really researched.)

  • http://www.mazepath.com/uncleal/qz4.htm Uncle Al

    The only deeply saturated stable (though not to acid) historic blue was ultramarine (from lapis lazuli, from one mine in Afghanistan until 1928 and Jean Baptiste Guimet). Woad, indigo, blue bice, smalt, Prussian blue were unstable and disappointing, God in His blue heaven was a very expensive boy. Brilliant red was never difficult (cinnabar to alizarin lakes). Reddened lips were definitely feminine.

    Spanish New World pillage missed formidable Maya Blue, indigo intercalcated into palygorskite. Modern stable hyper-blues are YInMn, DOI:10.1021/ja9080666

  • smut clyde

    the second half of the new paper, where he discusses possible evolutionary explanations why females would prefer pink.

    I read Del Giudice’s Letter and am still struggling to see how and why he jumped from one contentious issue (colour-coding of gender in infants’ and children’s clothing) to a second contentious area of research (colour preferences in adults). Parents do not think “the colour-preference pattern of women is skewed slightly towards pinks (relative to men); therefore we should dress our new-born daughter in pink clothes”.

    Even if the preference differences turn out to be real and based in biology, and even if one of the many evo-psych explanations turns out to have some validity, that has nothing to do with the cultural origins and history of gender codes in clothing. It is as if Del Giudice thought “I am writing about gender differences, so I must include multiple paragraphs of evo-psych speculations, however irrelevant they might be.” Or “I must include paragraphs of evo-psych speculations, and they must be relevant somehow, because they’re about colour”.

    • http://blogs.discovermagazine.com/neuroskeptic/ Neuroskeptic

      Quite! Given that in most cases (at least until recently) it is the mother who picks the clothes for both girls and boys, wouldn’t the “female pink preference” suggest that mothers would prefer to see all of their kids in pink?

  • OWilson

    So there long standing “debate” about color reversal? I hadn’t heard. :)

    Maybe it is just a debate for the usual social engineers who either want to eliminate differences between the genders, or create many genders to hide the real ones. (I hear New York recognizes 31 gender descriptors :)

    If you take a young boy and girl from the same family and offer them crayons and paper and leave them to it, when you come back, in most cases you will be able to tell which color crayons were chosen by each and which picture each had produced by subject.

    When they grow up they will tend to chose those colors for their children.

    • smut clyde

      No-one seems to have noticed this debate except Del Giudice!

  • Debora Weber-Wulff
    • smut clyde

      The Smithsonian author sums up Paoletti’s original thesis of a situation in flux:

      the two colors were not promoted as gender signifiers until just before World War I—and even then, it took time for popular culture to sort things out.

      Del Giudice exaggerated that into this idea of a colour/gender convention reversal, the “Pink-Blue-Reversal”, presenting it as received wisdom and himself as the sole contrarian, non-conformist voice…

      it is usually taken for granted that a remarkable cultural shift took place in the United States during the first half of the twentieth century. In the span of a single generation, gender-color associations underwent a sudden reversal

      Then he cherry-picked a few examples of authors who exaggerated Paoletti’s position for the popular press, as if they represent a consensus. Before claiming that Paoletti’s examples of “blue-girls, pink-boys” are few, and cherry-picked, and probably misprints (or the work of 1910 feminists trying to subvert the norm). Evidently not noticing that his own evidence for the universal acceptance of a PBR “scientific urban myth” is even thinner. I think he’s fighting a strawman.

      Anyway, the evidence he finds in the written literature seems to vindicate Paoletti’s claim… that colour/gender conventions only developed in the early 20th century, and initially lacked consistency. I am puzzled how he can jump from that recent and inconsistent development, to an argument that the conventions must be an expression of evolutionary psychology.

  • Not_that_anyone_cares, but…

    I think it was probably 1959 that I was student body president and my duties consisted of almost nothing except introducing the Tuesday morning’s ”assembly” activities. I had a pink shirt, pink trousers with a pink belt, pink shoes and pink socks. I have no idea why I had these things but apparently stepping out through maroon drapes to introduce the program was impressive

  • Jenny H

    WHY are the graphs all designated as male??????

    • http://blogs.discovermagazine.com/neuroskeptic/ Neuroskeptic

      I added the colors to the graphs. Each line shows the occurance of a pair of colors in the texts e.g. “blue for boys pink for girls.”

      I decided to make the lines correspond to the “male” colors of each pair. You could equally well flip the colors over and have them show the “female” colors.

      • OWilson

        What about the other 29 genders the government recognizes?

        Aren’t they entitled to a color too? :)

  • Erik Bosma

    Another million dollar study of an important subject paid for with someone’s hard earned tax dollars. I feel so enlightened now.

    • smut clyde

      Del Giudice’s methodology of consulting archives of newspaper word-usage has its drawbacks, but at least it can’t have required much money, only a few hours writing his report on the results.

    • http://blogs.discovermagazine.com/neuroskeptic/ Neuroskeptic

      I doubt a single dollar was spent on this paper!

  • Jim Fosnaught

    It helps if you actually go BACK TO WHEN BOYS wore pink, instead of starting at 1880 after the switch.. but here is aphoto from PRE 1880 showing a boy all in pink

    • Affoo

      That’s not pink.



No brain. No gain.

About Neuroskeptic

Neuroskeptic is a British neuroscientist who takes a skeptical look at his own field, and beyond. His blog offers a look at the latest developments in neuroscience, psychiatry and psychology through a critical lens.


See More

@Neuro_Skeptic on Twitter


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar