Folk Physics for Apes: The Chimpanzee's Theory of How The World Works

  • Daniel J. Povinelli
Oxford University Press: 2000. 407 pp. £49.50, $85
Hang on: we may have overestimated chimpanzees' ability to comprehend how things work. Credit: TONY STONE

As discoveries surfaced that chimpanzees can make tools and show self-recognition along with social politicking and cultural variation, the mental gap between us and them appeared to shrink so far that serious proposals were made to extend human-rights legislation to great apes. In some countries special protection is already in place.

Given this Zeitgeist, the results presented by Daniel Povinelli and his colleagues in Folk Physics appear quite shocking. They describe 27 meticulously conducted and previously unpublished experiments designed to assess what chimpanzees really understand about the way their physical world works. Seven young chimpanzees were tested on a dozen different kinds of tool use between the ages of five and ten years. Again and again, these juvenile chimpanzees apparently failed to take into account basic aspects of causality, such as that food will not fall 'up' into a container swung from its usual position so as to lie above the food, or that a hook needs to do more than merely touch its target to be a useful tool.

Povinelli's central conclusion is that there is a major, qualititative difference between the everyday 'folk physics' of humans, who can mentally represent unobservable causal factors such as gravity and force, and the chimpanzee's folk physics, which is limited to perceptually available information. Given the significance of this claim, and that Povinelli's group also studies children, it is surprising that we are not reassured by data showing that young children respond differently to the tasks the chimpanzees were set.

Be that as it may, the experiments follow a sustained logic that is fully and clearly explained. If the conclusions are correct, they have far-reaching implications for both chimpanzees and humans. How seriously should we take these results? There is space here to highlight just two main concerns. One is that the juvenile chimpanzee subjects were separated from their mothers in infancy (usually at birth) and reared as a peer group. How would cognitive performance be affected in human subjects reared in the same way? Were inputs that are crucial for a developing chimpanzee absent? Povinelli's answer is that the juveniles' experiences with a variety of tool-based tasks were in all likelihood richer than those of wild chimpanzees. This seems a compelling point. Caution remains, however, if primatologist Tetsuro Matsuzawa of Kyoto University is correct in inferring, from his analysis of how skills develop in the wild, that there is a critical period in infancy that is required for the proper, hierarchical development of competent tool use.

A lengthy opening chapter criticizes the 'argument by analogy' — the (mistaken) assumption that if members of two different species behave in similar ways, the same underlying cognitive processes must be at work. I was not persuaded that such criticism offers a radically new insight. Comparative psychologists already know that any behavioural similarity, on further probing, may turn out to be achieved by different cognitive processes. I am equally unconvinced that a fundamental logical flaw exists here, for it would imply that even the deepest probing could never establish both the differences and the similarities in cognition that comparative psychologists seek to delineate.

Perhaps Povinelli has himself fallen victim to the argument by analogy. He illustrates his critique by quoting his earlier studies, which gave a resounding negative verdict to the question of whether chimpanzees understand 'seeing'. This conclusion appears to be based on the analogy that a child failing such tests (understanding the implications of someone covering their eyes) would be unlikely to understand seeing in general. Recently, more natural experiments (involving competition with other chimpanzees over food) were carried out by Brian Hare and his colleagues showing that chimpanzees can and do discriminate important aspects of seeing (and perhaps even the 'knowing' following on from this). Therefore, we now have to consider that such an analogy does not extend to chimpanzees.

And so to the second major concern. If Povinelli's gigantic prior analysis of chimpanzees' folk psychology can be overturned by an elegant experiment more intuitive for chimpanzees, what of the prospects for the current, equally voluminous onslaught on folk physics? Time will tell. Whatever the answer, this book presents a rigorously documented set of internally consistent results that offer a stalwart challenge for anyone harbouring ambitions to chart the true mentality of chimpanzees.