All berries were pink in the Pleistocene, while meat was blue

They just won’t let it go. Some evolutionary psychologists are determined to salvage the idea that “pink is for girls, blue is for boys” has a biological basis. Marco Del Giudice goes digging with Google’s ngram viewer to collect data on whether pink and blue actually have undergone a consistent shift in preferences by sex (something no one has claimed), and thinks he has found evidence to overturn an idea he imagines that EP critics hold. It’s an amazing miss.

The role of pink and blue as gender markers is a source of endless fascination for both academics and the broader public.

Dude. No. We don’t find this pink and blue nonsense fascinating at all. We find evolutionary psychologists constant struggle to find biological significance in cultural phenomena exasperating. What is it with your bizarre obsession?

Five years ago I documented how a narrative that I labeled the “pink–blue reversal” (PBR) had become entrenched in contemporary culture (Del Giudice, 2012).

First, every true American knows that PBR stands for Pabst Blue Ribbon.

But secondly, there is no entrenched “pink–blue reversal” narrative. What is confirmed is that some people have insisted that there is an absolute, biological difference in how men and women percieve the world based on no evidence at all, and they were routed by observations of cultural variations that reveal that these color preferences are not hard-coded by evolution at all, but are conditioned responses to social signals.

There are biases. Visit a toy store; they all have the notorious pink aisle, where toys intended for girls are an eye-burning wash of hot pink. There has been no reversal. The question is whether girls are biologically programmed to prefer pink, the better to pick ripe fruit or respond to blushing or fevers, and that’s been shown to be a hypothesis without any good evidence, and a lot of counter-evidence.

The entrenched narrative is that evolutionary psychologists are full of shit. This paper does nothing to show that’s wrong. Quite the contrary: it demonstrates they’re even more full of shit than we imagined.

The PBR maintains that, in the U.S., pink was associated with males and blue with females until the 1940s, when the convention underwent a rapid and complete reversal. At the time, the PBR was treated as established fact in the media and the scientific literature. However, its originator—American Studies researcher Jo Paoletti—never argued that the convention was reversed prior to the 1940s, but only that it was inconsistent (Paoletti, 1987, 1997, 2012).

Oh, look. The point is sailing over the author’s head. I think it’s achieved escape velocity.

Again, that’s the goddamned point. Evolutionary psychologists want to claim a perceptual bias honed by millennia of hunter-gatherer selection on the African plains; everyone else points out that color fads in fashion fluctuate on a time-scale of years or decades, so you don’t get to invoke genetics as a basis for them.

Evolutionary psychologists come back to claim that the inconsistency makes their opponents wrong.

So what does his irrelevant data look like? Here’s a plot of his discovery of pink/blue color references by sex in books, over the last 140 years. He multiplies the frequency by 107 because the numbers are really tiny, but there is an initially small but steadily rising preference for claiming pink is a girl’s color and blue is a boy’s color over that time.

Clearly, this is evidence of a selective sweep for a pink gene in colors over the course of five generations. (No, it’s not). He argues that the UK was much more consistent in claiming that “pink is for girls”, and it’s just a few instances among those weird American books that claim “blue for girls”.

But wait. That’s from books. What about newspapers and magazines?

In total, the database of quotes from newspapers and magazines comprised 34 instances of standard coding and 28 instances of reverse coding. The combined data are plotted in Fig. 2. While the number of occurrences in the figure is too small to draw confident conclusions, the distribution of standard versus reversed gender coding looks approximately even, at least until about 1920.

(I love the way he labels “blue for boys” as standard coding and “blue for girls” as reverse coding, despite the fact that his own data shows that they’re approximately equal in frequency. Let your biases hang out!)

So the color assignments are basically equal by sex until about 1920, when suddenly the assignment of pink to boys plummets dramatically! An even faster selective sweep!

Del Giudice finds this significant.

The discrepancy between the two searches raises an intriguing historical puzzle. While the PBR account remains unsupported, quotes from newspapers and magazines suggest a pattern of variable and/or conflicting conventions in the late nineteenth and early twentieth centuries (see the Appendix section). However, the marked inconsistency observed in newspapers is virtually absent from the books published in the same period; instead, the pattern found in books overwhelmingly conforms to the standard convention of pink for girls and blue for boys.

I repeat: this “PBR” thing is a strawman made of bullshit. What has been pointed out repeatedly is exactly what he says here: “variable and/or conflicting conventions”. His work confirms what EP critics have been saying all along.

He also thinks the difference between books and magazines is a mystery. No, it’s not. He’s talking about a period when color printing was becoming increasingly common.

“When color began to be added to the products themselves,” Banta writes, “advances in color printing and reproduction followed. Starting in the 1920s, American consumers went from a commercial world of white towels and black Model Ts to a range of products with a fantastic palette of hues from which to choose.”

Right. So it wasn’t genes. It was a shift that occurred as the media began to impose color conventions on the public. It’s exactly what we’d expect if sociocultural influences were fixing arbitrary preferences on us.

Thanks to Matt Lodder for bringing this crap to my attention and getting my morning off to a pissed-off start.

Del Giudice, M (2017) Pink, Blue, and Gender: An Update. Arch Sex Behav (2017). doi:10.1007/s10508-017-1024-3


  1. says

    Some good news to make your morning better:
    The german Bundestag, in a surprising twist for both pro and con, just made SSM lawful. In a week the Bundesrat will ratify this bill and SSM will be a reality for 80 million more people.
    Last Sunday I was on a LGBTI+ protest and no one believed to see SSM in the next four years. You would have been laughed out of the protest if you suggested that it would happen in less than two weeks.
    And than on Monday a surprising domino chain was set in motion that let to the Bundestag voting about a law change that the government blocked for four! years at the last meeting of the Bundestag before the elections in September.

  2. cartomancer says

    What gets me most about all this is how narrowly parochial these people’s appreciation of cultural history is. They go back to 1881 for goodness’ sake, and then extrapolate from that directly to ancestral genetics. No attempt to find out whether these colour preferences they talk about existed 500, 1000, even 2000 years ago. Or, for that matter, whether they exist in societies outside the English-speaking world today.

    Indeed, they seem to take it as a given that all human societies actually have a notion that children should be assigned colours by gender. Most societies have never even thought in those terms – the colour of the things their children have was entirely irrelevant to them. Indeed, as well as the advent of colour printing and cheap coloured dyes we’re also looking at the advent of consumer mass-production and changing social ideas about the place of children in society during the Victorian era.

    More widely, historical ideas about the use of colour have been divided on many lines. In the ancient world the big one was class – Rome had strongly codified sumptuary laws about who was allowed to wear purple and how much. Fashions in dress and opulence also tend to undergo gender flips over the centuries for all sorts of cultural reasons. Slashed clothing, for instance, where the outer layer of fabric is cut in rows so you can see lower layers beneath, was very popular among European noble ladies in the 13th-14th centuries because it allowed them to parade how they can afford large quantities of expensive coloured cloth. By the 15th century it had become fashionable among hard-drinking German Landesknecht mercenaries looking to show off their wealth in as gaudy a style as possible – from a very feminine style of dress it had become a very masculine one. Henry VIII is often shown in this sort of clothing to emphasise his masculine qualities and importance.

    Moving away from history and colour to other cultural tropes, the Japanese have the idea that it is feminine to enjoy sweet foods with lots of sugar in them, so manly people actually hate sweet foods. I’m sure their lazy evo-psych goons are even now trying to explain this with reference to neolithic dietary differences, even though nobody outside Japan has even heard of this notion.

  3. says


    Moving away from history and colour to other cultural tropes, the Japanese have the idea that it is feminine to enjoy sweet foods with lots of sugar in them, so manly people actually hate sweet foods. I’m sure their lazy evo-psych goons are even now trying to explain this with reference to neolithic dietary differences, even though nobody outside Japan has even heard of this notion.

    Oh, I think you’re very wrong there. The trope of feminine = loves sweets is an incredibly common one, found in all manner of cultures.

  4. cartomancer says

    Caine, #4

    I’ve come across that half of it in lots of places, yes. But the corollary – that masculine people actively despise sweet foods – I’ve seen nowhere else. I could well be wrong of course.

  5. says

    I remember back in 1997, I sat through a bunch of presentations from a marketing team, in which they explained how various logo colors and backgrounds on websites could manipulate customers’ perceptions because of the evolved-in priming to color signals.

    It’s easy and tempting to dismiss that kind of crap as “pop psychology” but it’s not: it’s “mainstream psychology” and you can see what happens, when people try to scientize that sort of stupid idea in order to prop it up: you get tons of motivated reasoning and manipulated conjectures.

    What about the USAian political Red-Blue colour reversal?

    The marketing people told me that red is the color of aggression and energy, and blue is solidity and honor. In other words: it doesn’t matter because there is an “analysis” that all colors are good – except for when you reach for an analysis that a particular color is bad.

    This is just horoscopes, except it’s only 2 values and it’s gendered.

    What is midway between pink and blue?

  6. themadtapper says

    I continue to be baffled by the unrelenting desire Evo Psychs have for skewering this particular windmill. I mean putting aside all the obvious objections involving cultural differences, surely they must realize that not all berries and fruits are pink, right? And that for some of them pink is an under-ripe color? A genetic predisposition to want to gather and eat specifically pink stuff would be very limiting. One would think that the best adaptation for picking out the best foodstuffs to gather would be better cognition, yet curiously you don’t hear many Evo Psychs suggesting that women evolved to be smarter than men because of the need to identify ripe berries.

  7. says

    Also, the “gatherer” side of the hunter-gatherer equation often seems to be about tubers and grains and other starchy foods, not necessarily brightly colored fruit.

    Shouldn’t gatherers be fine-tuned for greens?

  8. Ed Seedhouse says

    Pink has been the “girl colour” for as long as I’ve been alive. Therefore it has *always* been the girl colour.

  9. cartomancer says

    I’m only fighting if the contest is either:

    a. Crossbows at dawn.
    b. An utterly cringe-worthy white, middle-class rap battle
    c. Guess the Weight of the Cake.

    Those are the only three forms of combat that my insurance will sanction at the moment. Though given the circumstances either paintball or a who-can-gather-the-most-berries competition would be most appropriate.

  10. blf says

    The mildly deranged penguin points out cheeses come in all colours — even pea green and horse puke — and also in all colors, therefore, there is a bias towards certain smells, or in some cases, aromas, of cheese, depending on presupposed biases. Obviously.

  11. chigau (違う) says

    I’ll pass on the duel, then.
    We’re going to have a ton of berries this year.
    raspberries, currants, saskatoons, cherries

  12. profpedant says

    I suspect that the pink/blue stereotyping would not exist if we did not have color vision and the ability to assume that a perceived difference signifies an important difference. So, in the very narrowest of technical definitions the pink/blue stereotype has a ‘biological basis’. But one needs to be very careful about generalizing, this technical definition is so thin that you could cut yourself pretty badly if you overgeneralize.

  13. says

    If pink was commonly seen in the 21st Century as a male colour I’m sure we’d hear some evo-psych explanation about how this is because certain parts of the female anatomy are pink.

  14. says

    chigau and cartomancer:
    Lavender or mauve? That’s tough! I was going to say feldgrau but that’s not pink enough.
    I was kind of hoping I could come up with a nice “just so story” for why feldgrau was really a ‘girly’ color.

  15. DrewN says

    IIRC, throughout most of (European) history, boys were dressed in frilly pink dresses bedecked in ribbons & lace until they were around 7-8.

  16. robro says

    PZ @ #11 — Wouldn’t that be greens and ochers? All the tubers I eat have a reddish-brown color. Note that the “gathers,” regardless of their gender, would also be gather firewood. In fact, they probably gathered more firewood than anything, particularly berries.

    DerwN @ #21 — Boys of means, perhaps. Poor boys (the vast majority) probably wore whatever came to hand, as did the girls. Color didn’t even register on the concern scale.

  17. Richard Smith says

    Strawmen made of bullshit are a leftover from just after our hunter-gatherer phase. They were used to cover the wattle in early “permanent” structures.

    “Daubmen” just never really caught on.

  18. Pierce R. Butler says

    cartomancer @ # 6: … masculine people actively despise sweet foods – I’ve seen nowhere else.

    What does the prototypical True Scotsman™ never do?

    Chromotography debaters above: Purple comes from combining red and blue. Lavender comes from combining – ta dah! – red, white, and blue.

  19. What a Maroon, living up to the 'nym says

    Speaking of PBR, what is the evolutionary reason that most USAians drink crappy beer?

  20. mgb003 says

    I’ve long thought that this blue for boys convention came from the popularity of Gainsborough’s “Blue Boy” as an art print to decorate middle class living rooms. I mean, the words are right there next to each other: blue and boy. How could you not make the association?

  21. anbheal says

    The Mazda Miata came out in Japan a few years before it was introduced to the American market, and it was targeted at young women. The main colors were canary yellow and baby blue, and the ads invariably showed early-twenty-somethings in perky summer outfits jumping and hugging, with the slogan: “So-ro-ree-tee GAH-ru!” (Literally, Sorority Girl). Men in Japan would never have driven a Miata. But when I got back to the States two years later, it was the Poor Man’s BMW. So I’m curious about the evolutionary triggers of canary yellow to gunmetal silver in low-end sportscars as their eco-system changed from the Pacific Rim to the Western Hemisphere, and the sexual selection underlying that rapid evolution.

  22. KG says

    Or, for that matter, whether they exist in societies outside the English-speaking world today. – cartomancer@3

    Come now; English speakers always assign colours in the way evolution intended. Unfortunately, non-English speakers obscure this scientific truth with their filthy foreign cultural traits.

  23. The Very Reverend Battleaxe of Knowledge says

    anbheal @ 29:

    Oh, there were plenty of gomers here ready to call the Miata a “chick car”. I guess because it’s not a pickup?

    Fortunately, a lot of people embraced it as a “Steel (Lotus) Élan”. Never heard the “Poor man’s BMW thing….

  24. chigau (違う) says

    you can tie two birds together
    they have four wings
    but they cannot fly

  25. blf says

    you can tie two birds together
    they have four wings
    but they cannot fly

    how doth the coconut migrate

  26. Nerull says

    American auto enthusiasts revere the Miata has a perfect starter car – cheap, easy to repair, more aftermarket parts than you know what to do with, great handling, and not fast enough to instantly get you killed.

    Outside of that subgroup, it is often considered a “girly” or even “gay” car, so the perception from Japan isn’t entirely nonexistent.

  27. jopaoletti says

    Thank you a million times. Make that a billon. I tried very, very hard to explain my research to that SOB and he just would not listen.