I once attended a ceremony in which a man from the Liguru tribe of eastern Tanzania was made a member of the Vindunda tribe from slightly farther west. It was a great and rare event and I was lucky to be on hand for its demonstration of how astonishingly fluid human identities can be. Of course, they should be fluid since the identity itself is invented. Linguists often dispute the extent to which language shapes human thought, but one area where it appears decisive is in matters of identity. It is hard to imagine how a person could understand himself to be Lliguru or Vidunda without language to name that identity. Apes may understand themselves to be members of a particular group, but without a name for the group there seems no way that two apes who are strangers can recognize that they share an identity. And since identities are crucial to so much of a person's behavior and thinking, this appears to be one way that language changes the way people think. Science magazine last Friday had a paper and an editorial that compares social networks among hunter-gatherer societies with ape networks and provides much food for this blog to chew.
I was struck by the way the paper, "Co-Residence Patterns in Hunter-Gatherer Socieities Show Unique Human Social Structure," takes human cooperation for granted. It is one way the papers have changed dramatically since I've begun this blog, and it allows scholars to get beyond the first implications of cooperation, which is that cooperation allows people to exchange their knowledge of a topic. Speech then allows groups to become self-aware so that members might call themselves Americans, or Waliguru, or Arabs, or whatever.
The central finding of the Science paper is that "most individuals in residential groups [of hunter-gatherers] are not genetically related." Thus, their cooperativeness cannot be explained using the popular concept of genetic altruism; i.e., when I help you I'm helping my own genes prevail. The team of authors then go on to say "inclusive fitness cannot explain extensive cooperation in hunter-gatherer bands." That radical claim holds if "inclusive fitness" is taken to work only on the gene level, but if you take a multi-level view of the matter you don't have to overthrow Darwin to accept these new genetic findings. It does seem to be a serious blow to attempts to explain human society in terms of "selfish" genes.
Creationists are going to get wind of this argument and try to make the most of it. So those of us who believe in multilevel selection should be primed to push the other side, pointing out that there is a full Darwinian explanation for human cooperation between genetic strangers.
The paper's abstract opens, "Contemporary humans exhibit spectacular biological successes derived from cumulative culture and cooperation." I would add intelligence to the issue and, when we are talking about "contemporary humans," we should also note that language has probably become more powerful than it was in the earlier days of Homo erectus.
Since the paper is more interested in "cumulative culture" than in language this blog post is more of a riff than a report. The authors' interest and Babel's Dawn's are parallel but not identical. Cumulative culture is a distinguishing feature of all human societies. If you define culture as group behavior that is passed on socially rather than genetically, then many species show culture. Some chimpanzees make a particular tool; some don't. Human culture is much more complex because it builds on other cultural details. A ritual to make a person a member of a tribe, for example, is not just one way of doing things. It requires the prior existence of other cultural things—the tribe itself being the most obvious.
Why don't chimpanzees have a cumulative culture? The authors say, "large social networks may help to explain why humans evolved capacities for social learning that resulted in cumulative culture." How might that work? The obvious advantage lies in the larger number of ideas and more diverse experiences available to draw on, but can that be the whole story? But do you really need a large network to produce innovations? If a network of any size can lead to a new idea, over time chimpanzees should be able to form a cumulative culture.
This question about why chimpanzees lack cumulative culture is very similar to one that comes up often on this blog: why don't apes have language? I have repeatedly rejected the suggestion that apes aren't smart enough for language. They are smart enough, as proven by the ability of trainers to teach them sign language. Apes could surely prosper more if they regularly communicated on the level of two-year-old. They do not talk because they are not communally minded. Is lack of community also what keeps ape culture from becoming cumulative? I don't think the answer is as firm.
You can learn from others by watching, even if the other does not cooperate. Thus, if one ape builds on an existing cultural artifact, others should be able to see that and do the same. Are apes smart enough to do that? A definitive answer require more data, but I doubt it. Do two-year-olds ever add to the existing culture? I doubt it, but don't have any data on the matter. So perhaps this points to an area where more than community, large or small, is required. Perhaps here we do see a point where apes just are not smart enough to match humans.
Australopithecus presumably had an ape-level culture—tools, for example, that smashed things. They may have been learned as chimpanzess do, by observing. The Oldowan tools of Homo habilis appear to be the first cumulative cultural artifact. The Oldowan tool kit includes cutting flakes and stones that make the cutting flakes. That italicized part is cumulative: a tool that makes a tool. I have heard that crows have such things, but crows are real smart. Apes don't seem to be that smart. Are two-year-old human toddlers?
Homo culture was apparently cumulative from the beginning, but for the first two million years it did not do much accumulating. The Oldowan tools were lightly cumulative. After a million years the Oldowan tools were joined by a more elaborate tool, the Acheulean axes that were worked on both sides. A million years later tools start to change a little more rapidly. And during all this period the fossil evidence makes clear that their brains were getting bigger. So there seems to be a case to be made that the cumulative culture which has put all humans in a different category from our ape cousins does owe a great deal to our improved understanding.
The authors say, "The origins of [cumulative culture and cooperation] may be related to our ancestral group structure." They then show that this point is not true, at least in the sense of kinship structures. Cooperation, they found, cannot be explained in terms of kin ties. The New York Times quoted Michael Tomasello, a regular hero on this blog, as saying if kin selection was much weaker than thought, "then other factors like reciprocity and safeguarding one's reputation have to be stronger to make cooperation work." All that sort of thing—reputation, favors, trade, moral codes, group identity—depends on language. Speech made us who we are.