The genealogy of morals

by Chris Bertram on September 23, 2003

There’s been much blogospherical and press comment about the recent report that capuchin monkeys have a built-in sense of fairness. In case anyone missed the story here’s Adam Cohen’s summary in the New York Times :

Give a capuchin monkey a cucumber slice, and she will eagerly trade a small pebble for it. But when a second monkey, in an adjoining cage, receives a more-desirable grape for the same pebble, it changes everything. The first monkey will then reject her cucumber, and sometimes throw it out of the cage. Monkeys rarely refuse food, but in this case they appear to be pursuing an even higher value than eating: fairness.
The capuchin monkey study, published last week in Nature, has generated a lot of interest for a scant three-page report buried in the journal’s letters section. There is, certainly, a risk of reading too much into the feeding habits of 10 research monkeys. But in a week when fairness was so evidently on the ropes — from the World Trade Organization meeting in Cancún, which poor nations walked out of in frustration, to the latest issue of Forbes, reporting that the richest 400 Americans are worth $955 billion — the capuchin monkeys offered a glimmer of hope from the primate gene pool.

Interesting, suggesting at least that monkeys on the receiving end of unfairness will would prefer to have nothing than be part of an unjust arrangement. It is a result that is consonant with lots of behavioural experiments involving humans, who will often walk away from a deal rather than maximizing their return. (See lots of places, but Skyrms’s Evolution of the Social Contract has some discussion.) But as Radley Balko points out , we’re a bit short of a true commitment to fairness here. If the monkeys were really into fairness, wouldn’t the one offered the grape spurn it rather than be party to such inequity?

Which brings me in touch with some of Henry’s Hobbesian speculations below, or at least to the subject of weapons. The poor monkey at the sharp end of unfairness in the wild is probably too weak to do much about it except feel grouchy and depressed. Not so, the human hunter gatherer, who, if he (and I’m afraid it is just he at this stage) feels aggrieved, can use language to form coalitions and spears to get his own back. Which provides a pretty good incentive to those with meat or other resources to share (or else). At least that’s the speculation contained in Christopher Boehm’s marvellous Hierarchy in the Forest: The Evolution of Egalitarian Behavior, which I’ve blogged about before. Principles of justice emerge as the weak use their language skills and weapons to conspire against the strong – a very Nietzschean thought.

It is an odd business, though, how so many people look to primates and hunter gatherers for a vindication of morality: that NYT piece even had the headline “What the Monkeys Can Teach Humans About Making America Fairer.” (Ernest Gellner has a funny, though otherwise misguided chapter lampooning such justificatory attempts in his Plough, Sword, and Book, entitled “Which way will the Stone Age vote swing?”) For whatever our ancestors or evolutionary relatives did or do, their behaviour can’t provide any kind of justification for principles. Still, I suppose there’s comfort to be had in the thought that we might be hard-wired to react against injustice. That fact—if it is a fact—may not justify principles of justice but it does give those of us who believe in them a tiny grain of confidence in the unjust getting their comeuppance … eventually.



Cosma 09.23.03 at 9:38 pm

It’s been a while since I read Gellner’s book, but I thought the point of that chapter was that there’s no justification to be had from the practices of our ancestors. Where then is that chapter misguided?


Chris 09.23.03 at 9:52 pm

Sorry, Cosma, that’s what I was trying to convey. I’ll re-edit slightly to eliminate the unclarity.


jimbo 09.23.03 at 11:00 pm

My favorite example of this kind of thinking is every feminist’s favorite primate, the Bonobo. The usual line is that, see, these guys are freely sexual, let the women run everything, etc. so why can’t we? Of course, the fact that we are, in fact not Bonobo chimps, that we have a different evolutionary history, and we are set up to have different kinds of behavior, never seems to cross these people’s minds…


Zizka 09.23.03 at 11:36 pm

A friend of mine has two mules, one of whom likes carrots while the other doesn’t. Once when he happened to have a carrot he gave it to the first. The second one bit him. He didn’t exactly want a carrot, he just wanted fairness.

(However, a freemarketer would understand this as the kind of envious egalitarianism which had to come to an end in order for there to be progress.)


back40 09.24.03 at 1:59 am

From the Emory U. press release.

The researchers recorded a 95 percent completed exchange rate with the subjects during the equity test, in which both subject and partner received cucumber as the reward for the same amount of work. The completed exchange rate fell to 60 percent during the inequity test, in which subjects observed their partners receiving grapes for completing the same amount of work. A further decrease to 20 percent of completed exchanges occurred in the effort-control test, when partners received the higher-value reward for less work.

Some monkeys, less than half, seemed to resent the good fortune of others. More, over half, resented lesser compensation even though they worked harder. When we look closer at the empirical results we can more easily degunk the situation of Cohen’s spin: “What the Monkeys Can Teach Humans About Making America Fairer.” YMMV. As with most other things there are a range of responses in natural systems rather than the hard edges of principles.

The myths about egalitarian H/G societies fail to account for the ways that such people measured wealth. When living a nomadic life possessions are a burden, you have to cart them about. Food is both bulky and perishable and the best way to carry it is in the stomachs of the tribe. Eat for a season when the opportunity arises. Food sharing is not egalitarian, it is pragmatic. Food sharing is not even equal. Hunters often consumed the best bits, such as the liver, on the spot of the kill. The honor went to those primarily responsible for the kill. In times of hardship the hunters got more than others so that they could continue to hunt and so be able to support the tribe.

Status in such societies is measured and rewarded not by possessions but by alliances. The most worthy were more sought after for friendship, mutual protection and sex. Complex methods to signal such worthiness, such as bringing food to the tribe and heroic defiance of danger, are evolutionarily stable ways of expressing true worth since poseurs fail, they have to be truly worthy to pull it off. The theory of costly signaling – demonstrations of worth that can’t be faked, that truly cost the signaller and so can only be done by the worthy – are as pervasive in nature as cooperation, mutual aid and sharing. Consider the peacocks’s tail – he has to be pretty studly to grow that thing, haul it around and not starve or be captured. Peahens notice.

Our principles cannot conflict with universal norms and preferences and succeed, but they aren’t determined by our nature. Is and ought etc. Our options are not infinite but we have choice.


Tom T. 09.24.03 at 2:30 am

I don’t understand why the monkey behavior is interpreted as having something to do with fairness. It seems just as reasonable to ascribe the monkey’s actions to simple envy and greed.


bigring55t 09.24.03 at 4:00 am

tom t.-
Fairness/Unfairness describes the situation to which the monkeys respond. The response of the monkeys is described as an acceptance or refusal of a treat they had been conditioned to accept. The response of the monkeys is not interpreted by the researchers but described statistically. Envy/Greed is an interpretation of what you think the monkeys are feeling and not scientifically verifiable. Note, this does not mean monkeys do not have emotions, just that the point of the study was how the monkeys reacted to a given situation, not how the monkeys felt about the situation.


Stentor 09.24.03 at 4:31 am

In times of hardship the hunters got more than others so that they could continue to hunt and so be able to support the tribe.

I would think that depends on the tribe. In more tropical climates, the proportion of food obtained by hunting is fairly small, so the most important people to support would be the gatherers (usually women, though men often gathered for personal consumption).


back40 09.24.03 at 5:09 am

True. Circumstances and behaviors vary.


Tom T. 09.24.03 at 5:23 am


Thank you for your response, but let me try to push my thought a bit further. Certainly, envy/greed is unscientific and anthropomorphic; what I was trying to suggest is that “fairness” is just as unscientific. I agree with you that the only valid point of the study is how the monkeys reacted, but we don’t have any way of knowing whether the monkeys reacted the way they did because they perceive the situation as unfair (or if they can even conceive of fairness). If I might rephrase your first sentence, I would suggest that it is more accurate to say that “Fairness/Unfairness is one way in which human beings might describe the situation to which the monkeys respond.” Alternatively, they might describe it as envy or greed.

I have not read the actual research paper, and so perhaps in the context of that paper it is accurate that “the response of the monkeys is not interpreted by the researchers but described statistically.” In de Waal’s comments quoted in the Times article, however, it certainly sounds like he is trying to derive moral lessons beyond mere statistics, when he says “a lot of the notions we use in our moral systems are much older than our species.” That may be true, but there is nothing in the NYT article that suggests a scientifically verifiable basis for de Waal’s statement (or that allows one to exclude contrary speculations that are equally anthropomorphic and unscientific).


dsquared 09.24.03 at 6:42 am

The theory of costly signaling – demonstrations of worth that can’t be faked, that truly cost the signaller and so can only be done by the worthy – are as pervasive in nature as cooperation, mutual aid and sharing. Consider the peacocks’s tail – he has to be pretty studly to grow that thing, haul it around and not starve or be captured. Peahens notice.

Please note that this is by no means the only theory of peacocks’ tails and not even currently the most popular one …


Keith M Ellis 09.24.03 at 6:45 am

“For whatever our ancestors or evolutionary relatives did or do, their behaviour can’t provide any kind of justification for principles.”

What an interesting thing to say. While I agree with it wholeheartedly (or -mindedly?), I can’t help but wonder at this assertion given that so many people — some of them even philosophers — have so often attempted to do exactly what you’re saying can’t be done.

This study is being discussed on just about every blog, and Macleod makes the point that he wonders if the conservatives who’ve embraced sociobiology as a rationalization for the justice in the status quo will take notice. But, of course, as jimbo’s quip above illustrates, it certainly hasn’t only been the political right that’s used naturist arguments when convenient. Leftists embrace such arguments, too. For example, in the homosexuality debate. That debate is close to my heart and I disagree with my fellow gay activists in their embrace of the naturist argument for its social acceptance. To my mind, whether it’s nature or nuture is beside the point given that there are so many other behaviors that are arguably “natural” that we don’t accept (nor, of course, are we obliged to, I think). In particular, you have the perverse result of this point of view to be that it argues for the acceptability of homosexual acts performed by the genetically homosexual, but implicitly concedes the argument against homosexual acts performed by those not genetically homosexual. I’m about as straight as most people get (not ideologically, but by preference), though I’ve tried sex with men on three difference occasions out of curiosity. So, does that mean that it’s okay to legally prosecute my behavior? Of course not. Anyway, my point is that arguments from nature are almost irresistable to everyone, and yet the are unpersuasive rationally, and, in practice, they’re also unpersuasive. People still think mostly in terms of what they imagine are reasonable evaluations of the “correctness” of a behavior, not on the basis of what nature may have determined. I have no difficulty believing that certain kinds of criminal sociopaths are “defective” biologically and are in those determined by nature to commit their crimes. No one I know of, however, is arguing for a social acceptance of that behavior by those so inclined. Thus, a naturist argument for *everyone* is ultimately a sort of intellectual crutch used to either avoid a deeper examination, or to short-circuit a rational discussion with someone who has a differing viewpoint.

Having said that, it in no way diminishes the obvious attraction that a naturist argument has for almost everyone.


dsquared 09.24.03 at 6:58 am

Is it not incredibly bad science to assume that just because monkeys do something, they must have been programmed to do it by their genes?


Jed Harris 09.24.03 at 7:37 am

William Harms (a student of Skyrms) has an interesting argument (see that moral truth can be tied to “proper functions” (roughly, the effects that cause specific traits to become selectively stabilized). This builds on Ruth Millikan’s work in teleosemantics which involves similar arguments about propositional truth. Note that any such normative arguments are relative to the selective context, which in humans is cultural as well as biological.


Chirag Kasbekar 09.24.03 at 8:24 am

The fairness argument is intuitively plausible. But I wonder if it is simply signalling of (strong) preference in the absence of spoken language — especially given a general security of food sources or some mixture of both those explanations.

BTW, I think Zizka is right about what free marketers have been saying. Hayek, for one, I think argued that the quest for ‘distributive justice’ is an atavistic psychological response from the egalitarian hunter gatherer times and isn’t in tune with a complex civilisation.

Not a totally convincing argument, of course.


dsquared 09.24.03 at 8:57 am

Strikes me that the monkey might just have learned that a pebble is more valuable than he/she thought, and that if it exchanges for a grape, it makes no sense to exchange it for a cucumber.


Keith M Ellis 09.24.03 at 9:39 am

“Is it not incredibly bad science to assume that just because monkeys do something, they must have been programmed to do it by their genes?”

That’s a good point, but I think you may be overstating it. I answered your same question on DeLong’s site, but I’ll answer it again here.

I haven’t read the study, I should say at the outset. Did they use a control group of monkeys that had had no contact within several generations, with these that displayed that behavior? Better yet, use as a control group some moneys that have been inhumanly treated (as there are, sadly, not a few of these) that have had no contact with any other monkeys at all in their lives. Of course, a properly controlled experiment of this nature might be difficult to perform.

But, since I think that non-human primates are of course capable of learning and are influenced by their own cultural variables (and this isn’t a radical concept, there’s lots of good science that demonstrates that animals exhibit lots of learned behavior, and much that is culturally normed) then you’re right — simply assuming that a monkey’s behavior is purely genetic is a very bad assumption.

On the other hand, I think that A) it’s indisputable that more non-human primate behavior is non-learned than is human behavior; and B) that a behavior that *is common* to both non-human primates and humans is more likely to be genetic simply by virtue of our shared genetic heritage *in contrast* to our *unshared* cultural heritage. So, in this context, no, it isn’t far-fetched to assume this behavior is at least partly genetically influenced.

A good study would rid itself of all sorts of preconceptions about human and non-human primate behavior in terms of the nature/nurture debate, and try to control for those variables in order to produce a reliable finding.


Dave 09.24.03 at 9:42 am

“Food sharing is not egalitarian, it is pragmatic. Food sharing is not even equal… In times of hardship the hunters got more than others so that they could continue to hunt…”
Maybe not equal, but sounds a lot like “From each according to his abilities; to each according to his needs” to me. And that’s one of my criteria for equality.
“we might be hard-wired to react against injustice” Well, we might be. But every example I’ve seen points to perceiving and reacting to injustice against ourselves, not against others or the body politic in general. If there seems to be a rule to the monkeys’ and mules’ behaviour it’s “get at least as much as the other guy” That’s a long way from equity.


Chris Young 09.24.03 at 10:57 am

It’s an incredibly bad assumption if it’s simply stated as though it were established, but it may be a reasonable hypothesis to base further investigation.

It’s probably worth bearing in mind that capuchins have been evolving separately from humans for about 19 million years, which is a fair old time even to a geologist. What this seems to me to be saying is that behaviour that looks like sulking because you haven’t got your way has now been observed in two species (capuchins and people) whereas it was previously assumed to be unique to us. This might, or might not, eventually lead to interesting generalisations about consumption preferences in mammals under conditions of abundance (I bet the little buggers would scoff the cucumbers quick enough if they were hungry), but it doesn’t seem to say much about concepts of fairness, because neither the animal with the grape or the one with the cucumber ever tries to do anything about it.

The point about the human concept of equity is that it is actively sought, with spears perhaps in a H/G group, with politics or litigation in a more complex society. If you don’t think that something should be done about the fact the the bloke (usually) at the next desk is being paid more than you to do the same job, you’re not afflicted with injustice, you’re just disappointed.

I’m reminded of the meme about mallards and rape. The blessed Dawkins (somewhere, don’t make me read it again) argues that it is reasonable to use the term rape to describe male ducks constraining females during mating because it’s obviously analogous to rape by humans. But it isn’t. The blessed Pinker, in The Blank Slate, proudly presents a list of attributes of human society which are apparently universal. One of these, sadly for us all, is rape. Another is the prohibition of rape.

And therein lies a tautology. Rape is rape because we condemn it as rape. There are societies which endorse behaviour by men which would certainly be regarded as rape in contemporary Europe and North America. Such societies still condemn rape, though, and fiercely, by their own definition. Ducks don’t. What ducks do is simply duck mating behaviour.

So I’d be cagey about reading too much into the tale of the poor deprived monkeys. What they do is monkey pissed off behaviour. It may or may not be analogous to some form of human behaviour, but if it is I suggest that it’s more likely to be the simian version of throwing a hissy fit than rebelling against injustice. And those things are a long way apart – ask any two year old.


Charles Stewart 09.24.03 at 12:59 pm

Relativists about crime: while I agree there is some variation in what various societies regard as serious crimes, I think the similarities are far more striking. In particular with rape, the only important deviations I know of is whether rape in marriage exists or not, and whether consensual sex with minors is called rape or not. In the former, the concern is that married women might get uppitty, in the latter, this is surely a misdescription: shouldn’t it be called child abuse? In short, I don’t think attitudes to rape are interestingly diverse.

A better source of diversity is with regard to warfare: have a look at this brief but fascinating page


Chris 09.24.03 at 2:13 pm

D^2: while my original post was intended somewhat light-heartedly, the issues between you and Keith lead me to press you on the locus of your scepticism. Not that I don’t think that scepticism is a fair response to the capuchin monkey case, but I’m curious about how deep you want to go.

There’s a pretty common form of triangulating argument in evolutionary biology (as I understand it) that if many now distantly related species share some characteristic X then it is a good bet that their closest common ancestor had X too. Even if some of the group have lost X, we’ll want to guess that their ancester had X. So even though ostriches can’t fly, we’ll want to say that the nearest common ancestor of ostriches and pigeons (and sparrows and eagles etc etc) could.

Are you sceptical about that form of argument in principle?

Or are you sceptical about that form of argument as applied to behaviour?

Or to psychological characteristics?

Or to humans and their nearest kin?

Or do you just think the evidence adduced in this (and like cases) isn’t sufficiently strong to serve in this kind of argument (though the argument would be ok with better data)?

Just asking.


khr 09.24.03 at 2:58 pm

A related point is the game-theoretical approach based on teh “Prisoner’s dilemma”. On very general principles and from basic computer simulations, it appears that core points of social ethics can be developed from very basic principles. Essentially, one of the best strategies to success in a society (in a very broad definition of “society”) is to cooperate with others as long as they cooperate, too, but to retaliate in a measured way against those who harm you (the so-called “Tit for Tat” strategy).

Googling on “Prisoner’s dilemma” and “Tit for Tat” gives a long list of relevant sites, e.g.

Karl Heinz

Hamburg, Germany


zizka 09.24.03 at 3:33 pm

The range of sexual behavior among our nearest relatives is pretty wide. To cherry-pick the species that most fits our theory is invalid. But is there a generalization possible? In any case, there are a large number of behaviors (language, heavy industry, MTV) unique to our own species, so that even if there is a primate norm, we may well be outside it.

It reminds me of the Catholic dogma my friends got. Promiscuous sex is dog-like. But it is also wrong for a pregnant women to have sex, because pregnant cows won’t. Any stick is good enough to win the argument.

The variation in standards of rape is quite large. In many societies rape is an offense against the family, not the woman, and is equally punished whether the woman is willing or not. Rape is sometimes regarded the woman’s fault, and she can be punished when the rapist isn’t. And in societies without law, rape can only be avenged by the family of the victim; unprotected women are routinely enslaved and raped. (In societies of this type, even murder is condoned if the victim has no protector. The prohibition of murder is by no means as universal as people think, except insofar as you define murder as prohbited homicide).


PG 09.24.03 at 7:05 pm

But much of the time, conservatives don’t perceive their preferred systems as unfair. In fact, they see a highly progressive income tax and similar redistributions of wealth as unfair.

If A is making more money than B, A in effect has produced extra pebbles and therefore is justified in getting a better treat.

So how do you work around that?


Keith M Ellis 09.24.03 at 7:36 pm

I’m a little confused by the obfuscation of what is, in *this* context, a relatively simple concept. “Rape” is sex forced on an unwilling partner by physical coersion. That you’ll find throughout nature. I recognize that this unadorned and unnuanced variety of rape is almost certainly the minority form among humans. I mean, let’s not get into a PC-inflected argument about what “rape” is, because I have to tell you I almost certainly know more and have more experience with the subject than you do. I’ve been a hospital intervention advocate for female (and male) rape survivors for a large regional crisis center. I *know* that the social dimensions cultural relativity of rape are immense. I *know* that a very large portion of rape in human society is almost certainly not a manifestation of a reproductive strategy. Nevertheless, it *is* a reproductive strategy among many species, some primates being most relevant, and it would be absurd to assume, as so many people are wont to do, that we don’t share a great number of evolutionary adaptative behaviors with our closest relatives. In this example, I think it is absurd to assume that *no* aspect or variety of human rape is related to similar reproductive strategies found in other primates.

I recognize that the historical context of evolutionary psychology ne’ sociobiology and its misusers, inside the larger and longer sorry historical context of “social darwinism” and Nazism all conspire to badly taint this field and any ideas related to it. But I am getting impatient with it because among many this rejection is now completely unthinking and reflexive. You unfairly seem to ridicule Dawkins and Pinker (with the — unless I am misreading you — facetious “blessed”) even though neither of them are politically motivated and both of them are highly respected scientists. It’s not like they’re Herrnstein and Murray, for crying out loud.

You wrote: “Ducks don’t. What ducks do is simply duck mating behaviour”, which although as a practical matter I am willing to agree with the contrast you’re making; as a matter of principle, I am highly suspicious of it. You seem to be assuming (and bear with me on this since I know nothing of duck mating behavior so I’m just riffing on your example) that there is no sense in which a duck that is forced to engage in sexual intercourse by another duck is in any sense legitimately aggrieved. Okay, sure, to say otherwise seems quite nutty. But, let’s move closer to humans, and change the behavior under examination a bit. How about “chimps” and “murder”? You have one chimp that gets pissed-off at another chimp and kills him. I ask: do you think the victim chimp is less (or not at all!) aggrieved to be killed because, by definition, and similar to your duck argument, any chimp behavior is “natural” and therefore completely indepedent of any value judgments (which is your argumentive basis for saying that “rape” is a concept which simply cannot apply to animals other than humans)? My point is that your argument would seem to require that all human behaviors upon which we make value and moral judgments — and that’s practically all of them excepting autonomous functioning, isn’t it? — would be similarly assumed to be qualitatively distinct from any superficially similar behavior exhibited by animals.


Chris Young 09.25.03 at 8:21 am

Keith – Damn, I never meant to start a tangent about rape, I wanted to make a point aboout capuchin monkeys. That idiocy about mallards has been current for at least 30 years to my certain knowledge, so I thought it would stand as an example of bad pop EP that didn’t need introduction.

But on your substantive point, I would have thought that if you observe a simple behaviour in species A which looks a bit like a simple behaviour in species B, then unless you can support the analogy by showing that the wider behavioural context in which it occurs is also similar, any claim that they are the same thing looks a bit weak. Specifically, I don’t suppose any animal much likes being killed, including the pig that died to make my bacon buttie this morning, but Goodall reported several instances of common chimps killing other members of their troop, and the troop did not act collectively either to retaliate or to prevent repetition. To me this suggests that the Gombe population at least don’t share a common set of responses to intraspecific killing with humans, and that makes it a big linguistic stretch to talk about “murder”.

Back to the capuchins, if you read what the people who did this work wrote in Nature, you find two points which were a bit under-reported. Firstly, the “resentment” behaviour was only observed in females, whereas humans of both sexes are equally capable of reacting to unfairness. This seems to me to reduce the likelihood that the two behaviours are a common adaptation, since if it were, the genes responsible would be expressed in a strikingly different way. But let somebody do the work, and I’m happy to be convinced.

The second point, which was where I was trying to draw my doomed analogy with the ducks, is that the animals which responded so graphically to being given cucumbers did not subsequently change their social behaviour towards either the experimental team (who had perpetrated the injustice) or their conspecifics which got the grapes (and benefitted by it).

If there is one thing that can be said with certainty about the human sense of fairness (see games theory passim and your own excperience) it’s that when it is violated people change their behaviour towards those who have treated them unfairly. So, on balance, I feel the claims this team is making – and they do talk about “fairness” in their paper – are more than a bit overstated for the strength of their story.

BTW, I have immmense admiration for Dawkins and Pinker as scientists, but I find them dreadful as advocates for their views to the laity. I know this is a personal foible, but there you go.


John Rosenberg 09.26.03 at 2:56 am

Terrific post, and comments. For anyone not sick of this Monkey Business, I have additional discussion here:


Bruce 09.28.03 at 4:02 am

“The poor monkey at the sharp end of unfairness in the wild is probably too weak to do much about it except feel grouchy and depressed.” Actually, there is chimp research showing that the number 2 and 3 chimps occasionally form alliances and depose the alpha chimp if he’s not careful. See de Waal, Good Natured, pp. 172-73.

Comments on this entry are closed.