28 June 2009

Were Prehistoric European Cave Artists Female? (Updated)

I want to get my two cents in ... before this brouhaha becomes writ in stone.

Or before I'm accused of ignoring (rather than cheering on) the newly discovered 'Uppity Prehistoric Female Artists' who were supposedly painting cave walls 30,000 years ago.

What's the scoop?

The underlying idea is very clever, really.

Upper Palaeolithic people loved to stencil, paint, and press their hands into the soft clay of cave walls -- as if to leave unique, indelible signs of their presence. Some cave walls are literally covered with hand prints, but hands also appear around more complicated painted images, such as those surrounding the spotted horses from the Pech Merle cave in south-western France (below).

So whose painterly hands are these?

It has long been known that men's ring fingers (the fourth digit) are usually longer than their index fingers (the second digit). Women's second and fourth fingers are generally equal in length.

This is the famous finger-length ratio, known in fingering circles as 2D:4D -- that is, the relationship between the length of the second digit, 2D, and the fourth, 4D. Men generally have a low digit ratio, calculated by dividing the length of 2D by the length of 4D. Women tend, in the case of the ratio at least, to be equal.

I bet you're looking at your own hands right now. But hang on a moment, and listen to the argument.

Until recently, according to Pennsylvania State University archaeologist, Dean Snow, most scientists assumed these prehistoric hand prints were male. But, he says, "even a superficial examination of published photos suggested ... that there were lots of female hands there."

Looking at some stencilled hands, Snow could see that "The very long ring finger on the left is a dead give-away for male hands. The one on the right has a long index finger and a short pinky -- thus very feminine."

To assess prehistoric hand-prints from European caves, Snow used modern hands for comparison. "I had access to lots of people of European descent who were willing to let me scan their hands as reference data," said Snow.

By carefully measuring and analysing the Pech Merle hand stencils, Snow found that many were indeed female -- as, for example, those in the 'spotted horse' picture (above). And so he concludes, "We don't know what the roles of artists were in Upper Paleolithic society generally. But it's a step forward to be able to say that a strong majority of them were women."

I hate to be a party pooper but...

this begs three questions.

First, and to my mind most serious, is:

How do we know that today's 2D:4D finger ratio was the same for the early modern humans who painted the caves?

We have very little surviving skeletal material from the Upper Palaeolithic (and what we do have usually looks like those bits and pieces on the left): fragments from a total of 31 adults, mostly their teeth. Still, palaeo-anthropologists can do remarkable things when they get their hands on a bit of skull or a jawbone -- but what they can't easily do is sex the bits. Out of 31 adults, 3 are probably males and 1 is female; that's it: the sex of the others cannot be determined.* So we have no real sample of either sex. And finger bones are rarely found: there are only 4 phalanges (from hands) on Palaeolithic Aurignacian sites.

That's a big methodological problem.

But not the only sticking point. I have another quibble with Prof. Snow's method:

It assumes that the modern Europeans who were measured for this study are, in some way, descendants of this Upper Palaeolithic painterly population.

I can't see any historical reason to accept this. But even if I did accept it, what makes us think that the 2D:4D is a universal constant? In the last ten years or so, about 150 digit-ratio studies have been published. There is now evidence that the digit ratio varies among ethnic groups: Caucasians tend to have high 2D:4D, while black and East Asian people tend to have low 2D:4D. And, worse, 2:4D differences may have genetic roots (rather than being caused by testosterone in utero, as previously thought),** so -- if it's going to work at all -- getting to the right ethnic group is of prime importance.

Why extrapolate back from Caucasians ... to a Palaeolithic group that arrived in Europe not so long before, coming out of Africa?

And, finally, even if we accept the research at face value, it still only tells us that there were women in the caves at about the time when the paintings were made -- and not who made the paintings.

Women in the caves?

That is not news.

"Most scientists" have long given up the idea that only male adults were active inside the caves. Kevin Sharpe, for example, worked for years on the so-called 'finger flutings' -- lines drawn with fingers in the soft clay -- in Gargas Cave in the French Pyrenees and at Rouffignac in the Dordogne.

Line flutings are not as dramatic as paintings -- to our eyes. But they are well worth studying in great detail because, in some caves, the same people made both the painted images and the line flutings. So, when we know everything we can about flutings, we will also have learnt something about the cave artists. And perhaps about the content of their 'art', too -- because the flutings, drawings, and paintings may have arisen from the same social context and impulses.

So, who did the fluting? This is what Sharpe thinks:

three people fluted much of Gargas, a child (perhaps a boy), a woman, and a man. About eight or nine people fluted (and probably decorated) Rouffignac, including one baby, two young children, and 5 or 6 others including both males and females, at least one of whom was an older juvenile or adolescent.
In other words, everyone.

Maybe we should start thinking of the 'art' as a family affair rather than impose our role divisions on the past. Otherwise, we may just be reflecting our own predispositions and not the reality in the caves at all.

That's why it always seems so convincing.

Update: 01 May 2017 


A new project has the Spanish archaeologist Hipolito Collado and his team going from cave to cave, taking scans and high-resolution photos of all of Europe’s prehistoric painted hands. 

They then post them in detailed, 3D format in a free-to-use online database, as part of an EU-funded project called Handpas.

Read more on Seeker

* Dominique Gambier, "Modern Humans at the Beginning of the Upper Palaeolithic in France", in Conceptual Issues In Modern Human Origins Research (1997), Chapter 8.

** The evidence reviewed in Finger Forecasts by Jennifer Huget in The Washington Post. More background, especially on variability and 2D:4D diagnostics, from Finger It Out at FlatRock.org; see also Science News: Finger Length Ratio May Predict Women's Sporting Prowess.

My thanks to David Meadows at Explorator for signalling this breaking news.


Top left: Handprint from Chauvet Cave.

Top centre: from Finger Forecasts by Jennifer Huget,
The Washington Post.

Middle centre: "Spotted Horses" mural from Pech Merle cave. Credit: Dean Snow, via National Geographic News.

Middle left: Photographs by Roberto Ontanon Peredo, courtesy Dean Snow, via National Geographic News.

Lower left: A human jawbone from a Romanian bear hibernation cave, dated to between 34,000 and 36,000 years ago, along with a facial skeleton (center) and a temporal bone
. Credit Erik Trinkaus.

Lowest left: A finger-fluted 'tectiform' from Rouffignac Cave, from K. Sharpe & L. Van Gelder, Human uniqueness and upper paleolithic ‘art’, Fig. 1.


  1. How do they know that "in some caves, the same people made both the painted images and the line flutings"?
    Another question: how did they deal with their fingernails in those very old times? I suppose they kept them down with a rough stone. But it's hard to imagine the brutish hunters regularly minding their nails. Still, if they didn't, they would become a real problem. I remember how Benvenuto Cellini complained about the pain his long nails gave him when he was in that dungeon. Perhaps they just rubbed them in the sand or on the wall every morning, like a bear. (One of the caves in Atapuerca has prehistoric bear-scratches on the wall.) Could finger flutings have something to do with this bit of hygiene?

  2. Measure and count the number of different hands doing the fluting versus number and size of footprints, I imagine -- and then follow the footprints to each painting, etc. I did mention, I think, that Sharpe (and his wife L. Van Gelder) were terrific on detail.

    Don't know about fingernails. I used to bite my nails. I suppose they could too (why didn't Cellini? Had he lost his teeth?). In any case, the finger marks and hand prints that I've seen don't show long nails.

  3. Just read your post after seeing it at Four Stone Hearth. Great review and critique. I've linked to it here.

  4. In fact, Cellini complained of loose teeth in that dungeon.
    And toe-nails? Did you use to bite them too?

  5. No, I didn't, 100 swallows; but don't you think walking around barefoot over rough ground all your life might do the trick? Anyway, I have not seen many images of footprints so can't really say if they had long toenails.

    I told you Cellini probably had a dental problem!

  6. I'm not entirely sure why it matters or why the news reported it as a surprise that women might have done some of the handprints really.

  7. www.costasmaritsas.com - english flag. the answer - the unfittest man maked the art...

  8. If we don't have enough finger length evidence for the population that made the hand prints, how far back do we have finger length evidence for those locales? I understand that, with populations moving around, same locale but different era does not mean one population is descended from another. However, wouldn't it be cool if our technology and record-keeping get to the point that we can genetically link a subsequent population with known finger lengths to the previous population associated with the hand prints? And if a graph over time shows that the finger ratios stayed somewhat the same in that population... Oh, I suppose it wouldn't guarantee the previous generations with no finger length data would have the same ratios. Shoot. Thought I was onto something, there.

    Anyway, thanks for the article. Believe it or not, I found you with surfing starting at engrish.com. Heh!

  9. great post, any alternative to the canon is much admired.

  10. Ben, thanks for writing.

    I am not opposed to this or any other canon but I am always sceptical. It helps to keep testing the models before they harden into accepted fact -- when they're nothing of the sort.


Blog Archive