National Geographic

Scientific Punch-Up Over Idea That Fighting Shaped Our Hands

If you learn martial arts, one of the first things you get taught is how to make a fist properly. Classic mistakes include sticking the thumb inside the other fingers or curling it around the side. The right way is to wrap the thumb over the index and middle fingers. In this shape, the fingertips are cushioned against the pads of the palm, and the first two fingers are cushioned against the thumb pad and the thumb itself.

By getting martial artists to hit a punchbag, and measuring the forces acting on their fists, David Carrier and Michael Morgan from the University of Utah confirmed that this shape allows the various parts of the hand to buttress each other, turning a flat hand into a stiff, compact club. This channels the force of a punch into the palm, wrist, and forearm, and protects the delicate fingers.

But more controversially, Carrier and Morgan also suggest that this might explain why the proportions of our hands evolved in the first place—for stability during combat, rather than dexterity during tool use.

Compared with the hands of other apes, our palms and fingers are shorter, and our thumbs are longer, stronger, and more mobile. That makes for a stronger fist, and would have allowed “competing males to strike with greater force and power while greatly reducing the risk of injury to the hand,” they write.

It’s an idea that has divided opinion. I contacted four scientists about the study. Two expressed their respect for Carrier’s wider work but were unconvinced by his new idea (although neither wanted to comment on the record). A third—Brigitte Demes, who studies the limbs of primates at Stony Brook State University—said that the actual experiments were sound, but “the interpretation is far-fetched”.

The problem is that, as Carrier and Morgan write, the goal of their study was “to test the hypothesis that the proportions of the human hand make it an effective weapon”. They did that. But the hypothesis that fist-fights shaped the evolution of our hand is different. Going from one to the other is like showing that computers are good at surfing the Web, and suggesting that this is what they were invented to do.  It involves speculative just-so stories—stories that may well be correct, but that need evidence.

But John Hawks, an anthropologist from the University of Wisconsin-Madison, found the paper very reasonable. “We do have a problem understanding why human hands have shortened fingers and longer thumbs compared to apes,” he says. “The change in hand proportions came long before the first appearance of stone tool manufacture, and chimpanzees make and use wooden tools without having our hand proportions.”

From Morgan and Carrier, 2012

In which the idea steps into the ring

Carrier first got his idea about fist evolution after working on sperm whales. He suggested that the males use a big waxy organ in their heads as battering rams during fights. He was discussing the idea with colleague Frank Fish, who didn’t buy it. Fish raised his fist and said, “I can hit you in the face with this, but that is not what it evolved for.”

Carrier started wondering about the forces acting upon a punching fist. First, he and Morgan asked boxers and martial artists to lash out at a punching bag with either open palms or closed fists, while recording the strikes with an accelerometer. To their surprise, both techniques delivered the same amount of force.

Next, they asked the volunteers to press their fists against a force-measuring instrument, with their hands rolled into different positions (see above). They showed that the main knuckle joint of the index finger is twice as stiff if the fingers are properly tucked into the palm (B vs. C), and twice as stiff again if the thumb wraps around the fingers (A vs. B). This classic fist prevents the fingers from moving too much during a punch, and the knuckle from overextending. It also transmits twice as much force into the wrist than the less supported shapes, again saving the finger bones from undue stresses.

So, the story goes like this: Our male ancestors fought each other for mating rights. Their hand proportions evolved from those of a typical ape, to those that allowed them to whack each other without breaking their hands. With their non-self-destructing punches, our fore-fore-fore-forefathers got more sex, and gradually their hands attained the proportions of a modern human’s.

The ability to more dextrously wield tools might have also played a role, but Carrier and Morgan argue that there are many ways that an ape-like hand could have changed into one that was good for manipulating tools. For example, the thumb could have grown longer, or the fingers could have shortened without also shortening the palm. Both would have given a precise grip. As it happens, the route that we took is the only one that also provides a strong fist.

Ape hand and foot, by James Swoboda

In which the idea takes some hits but doesn’t fall down.

But Demes spots several problems with this idea. “For a start, the human hand is not particularly well designed for fist-fighting,” she says. “That’s why boxers wear gloves, and why there is even a fracture known as Boxer’s fracture.”

“Our closest relatives, the chimpanzees, also don’t fight with their fists,” says Demes. “They fight primarily with their teeth, using their hands to grab you and bite.” When they do use their hands, they deliver powerful open-handed slaps. You can see both techniques at work in this video. Pity the martial artist that pits their closed fists against the open hands and teeth of a chimp.

But Hawks adds that “it is clear that early hominins de-accentuated fighting with large canine teeth.” Some people think this was due to less aggression between males. “Carrier gives an alterative point of view: that fighting was still important, at least for some early hominins, but took a different form.”

This is a critical point. Our hands can make strong fists, but that would have given our ancestors an advantage only if they fought a lot with their fists. If they used other equally effective striking techniques, it is hard to see how the advantages that Carrier and Morgan describe would have provided enough evolutionary pressure to mould the shape of our hands. As Fish originally said to Carrier, “I can hit you in the face with [a fist], but that is not what it evolved for.” So how did they fight? We don’t know.

“There are more plausible evolutionary scenarios for all of the features that Carrier bases his fighting story on,” says Demes. These include the idea that our hands allowed us to more precisely manipulate objects or that they changed as a side effect. Hands and feet are governed by similar blueprints, and as our feet adapted for two-legged walking, our hands may have been carried along for the evolutionary ride.

And there is one more possibility: There might be nothing to explain about our hands. Carrier and Morgan’s story assumes that our ancestors had hands with the proportions of other apes. But Hawks points out that many monkeys have hands with proportions like ours. “If our common ancestors with other apes were not ape-like in their hand proportions, there may be nothing about human hand proportions that needs explaining,” says Hawks.

There’s a silver lining, though. “I did learn from the paper that next time I get into a fist fight, I’d better make sure that I curl my fingers up properly,” says Demes.

Update: T. Ryan Gregory points out another flaw that riffs off my opening.

As anyone who has studied martial arts knows, this way of making a fist is NOT intuitive and takes practice to become comfortable. The authors of the paper had trained martial artists strike heavy bags, and compared this to open hand strikes. I don’t doubt that the fist position they use is more effective. But it’s not representative of what most people do today when punching unless they have been trained — let alone what early hominin ancestors were doing. And I REALLY doubt this created a strong enough selective pressure to create a trade-off with tool use or other more obvious functions of the human hand.

Reference: Morgan and Carrier. 2012. Protective buttressing of the human fist and the evolution of hominin hands. J Exp Biol. http://dx.doi.org/10.1242/jeb.075713

More on hand evolution:

There are 19 Comments. Add Yours.

  1. Jumbybird
    December 19, 2012

    Why didn’t they just grab a rock or branch and hit you over the head… that makes more sense than punching someone.

  2. Dave
    December 19, 2012

    Hi Ed. As a long-time lurker on your Discovery Blog I had to have a look at your new home. I wish you well. The comment space seems to be a bit restricted though.

    On the subject, as a one-time karate enthusiast I confirm the hand position but would like to point out that the martial art punch is very different than the boxing punch. The former uses the first two knuckles of the the hand which causes the hand to stiffen and form a strait-line relationship with the wrist, thus decreasing the striking area and applying more force and penetration to the target, In contrast the boxing punch is delivered with the plane aspect of the fist, which is what causes the phalanges to break (particularly on a multifaceted target like a face) and causes the boxer’s injury.

    You could never break a brick with a boxing punch without severe damage to yourself.

    Having said that if I were an early hominid, I think that being able to wield a rock or strong branch with dexterity would be more effective than either.

    Thanks Dave. Re: comments, we can pass any comments about design to our tech people. I agree with you that the field is narrow, but it’s vertically expandible. And it sure beats the Discover system where you have to register. – Ed

  3. Thomas
    December 20, 2012

    Karate punches were designed to break through wooden armor, not strike the face. The biomechanics of a karate punch are actually *weaker* than a boxing punch when striking a face, with particular risk to your wrist. A boxer risks breakage of the metacarpals but has a more stable wrist alignment.

  4. SP
    December 20, 2012

    Yes, you would have thought that picking up a rock would be the dominant strategy. Lovely write up though.

  5. Mike
    December 20, 2012

    Just to play devil’s advocate – computers weren’t invented to surf the web but they turned out to be good for doing so and have since evolved so that it has become one of their primary purposes. More than one element affects physical evolution and I think it’s totally plausible that early humans discovering this extra, helpful use for their fists had an impact on further evolution of these ‘functional’ hand shapes.

  6. Ed Yong
    December 20, 2012

    Mike, yes, it’s totally plausible. But the point is that this paper, while showing that fists are mechanically strong, doesn’t actually provide evidence to confirm or deny the hypothesis.

  7. James Sweet
    December 20, 2012

    This research shows it is conceivable that fistfighting was *a* factor in shaping the human hand (it doesn’t prove it was a factor, and I am very skeptical myself, but it does at least show that it’s conceivable). However, we can be pretty much certain that it wasn’t the *dominant* factor: There are countless possible forms for the human hand that would be better punching tools but be lousier at using tools. Clearly, the dexterity and tool use was a much stronger selective pressure.

  8. Lou Jost
    December 20, 2012

    Off topic, but I want to congratulate you on your move. That Discovery site after the changeover was ugly and the comment system was barbaric. Why do business types so often spoil nice things? I am also glad to see the return of your personalized space on the right. Best of luck here!

    One small complaint: on my high-res screen, the textbox for writing comments is so small that I cannot see even one sentence of what I have written, and the scrollbar on its left is very hard to use since the space between the “scroll up” and “scroll down” arrows is so narrow that the scrollbar completely fills it.

  9. Jorge Silva
    December 20, 2012

    @Lou Jost, not making excuses for bad design (it would have been very easy to make the input element fit the width of the container, whatever that would be on any given screen) but your situation can be ameliorated by clicking and dragging down the bottom right of the comments box. Make that as big as you like and then start typing.

  10. badfrog
    December 20, 2012

    Just no. I owned a chain of kung fu and tai chi schools for 25 years and have spent over 45 years training. The fist is not the most effective hand shape, and the punch is deliberately designed for pro boxing to limit the damage and deaths. The most powerful movement above the waist is called brachiation, the movement monkeys and apes use to swing through the trees. Used properly, it will kill another man, and break your fist. The hands are tools, not weapons. The power strikes are with hammer fists and chops, using the pisciform bone of the wrist, which is very solidly attached. Still better is a stone or a stick. If you want to see what the real weapons are, study the larger monkeys. They slap and pinch, headbutt, grab and throw, grab and break. And bite, which I also have practiced extensively. I have NEVER seen a monkey make a fist, and I have looked extensively, especially with the black macaques, which are the largest monkey in the world at over 110 pounds. One to one, they can kill any man. This idea is simply incorrect. We are toolmakers, not fighters. Fighters have claws or other obvious weapons. We have brains. In a group, with weapons, our organization has allowed us to overcome all opposition.

  11. Ed Yong
    December 20, 2012

    Lou and Jorge, as I mentioned up above, we’ve passed the comments about the comment boxes to our tech folks. This is something we raised at the start, but there were other more important design elements to straighten out.

  12. William Jungers
    December 21, 2012

    The hominin thumb remains a relatively mobile and gracile organ until quite late in human evolution, millions of years after the origins of bipedalism. And the modern ape thumb (e.g., chimpanzee digital proportions) is almost certainly not primitive for the LCA — and earlier Miocene “apes” possess relatively long thumbs along with relatively short hands (see Sergio Almejica’s pubs on this front). .

  13. Tom Swiss
    December 21, 2012

    According to a karate proverb, it takes three years to learn to make a fist. (And three more to learn proper stance, and three more to put it all together and have a proper punch.) Kids very often put their thumb on the inside, and even boxers (who are used to those big gloves and so often don’t know how to make a good fist) often strike with the small knuckles of the third and fourth fingers, rather than the proper first and second. So it’s not very instinctive.

    The idea that karate punches were “designed” to punch through wooden armor is based on a mythological history of karate.

    A hammer-fist or knife-hand (“karate chop”) blow can certainly be effective, but its circular motion is more easily blocked or dodged than a linear thrust. Elbow blows can also be highly effective, but have less range. So the punch has an important place in a fighter’s toolbox.

    Even a dilettante like me who sticks with karate training long enough will develop a punch capable of breaking a stack of four 1-inch pine boards. I’ve seen real experts break stacks of concrete blocks. And there’s not a tremendous difference between a karate “reverse punch” and a boxer’s “straight right” (except for the fist training). So the idea that boxing punches are somehow designed to limit injury strikes me as another myth. (Martial arts are full of them, sadly.)

  14. Bas
    December 21, 2012

    Left hook palm strike to the liver!

    Seriously though, palm strikes seem much less risky, as far as breaking bones in the hand, and according to this article, they deliver the same amount of force as a closed fist, though reducing reach slightly.

  15. Dave
    December 21, 2012

    I agree with Tom Swift – I could (when younger) break boards with the karate reverse punch. I doubt if it would be possible with the boxing punch.

    The reason is that the karate punch is delivered to the target with the maximum possible speed (straight line) with the most energy transfer (because the striking area is small) allowing the energy (half-m-v-squared) to transfer quickly to the target and doing the most local damage.

    The boxing punch, by contrast relies on momentum transfer over a wider area and does more concussive damage in a larger volume.

    When martial artists wear gloves they can’t develop the fist so favour the boxing style when using hand-blows.

    Anyway, more to the point, the article is making a completely non-testable hypothesis.

    It is like claiming that earlobes developed so that we could wear earrings. Or that our fingers are long so we can pick our noses.

    And I’d still hit someone with a spanner (oops wrench) given the choice.

  16. Alex Fairchild
    December 27, 2012

    Alright, new digs at natgeo!

    Another option: good ol neutral drift. Primate hand proportions are all over the map, let’s not hunt for selection on every feature.

  17. Jason Dowd
    December 28, 2012

    “Far from the truth lay the antique assumption that man had fathered the weapon. The weapon, instead, had fathered man.”

    Robert Ardrey

  18. Shi Pasang
    June 9, 2014

    Why are men and women’s hands similar if only males were fist-fighting for mates?

    This sounds to me like utter drivel. About on a level with pop-anthropology and psychology. You know what I mean – taking common stereotypes from contemporary culture and suggesting these stereotypes reflect some deep, inescapable necessity of ‘biology’ or ‘evolution’. The arguments are usually trolled out to promote a sexist agenda rather than racism but they are on the same intellectual level as claims that diffetent races have destinies determined by biology. I hope no serious scholar or scientist gives this ‘theory’ anything but contempt.

Add Your Comments

All fields required.

Related Posts