In our final week we take up issues concerning the evolutionary history of the capacity to attribute intentions and intentionality. Is this capacity a primitive phenomenon arising early in our evolutionary history and now common to many species? Or is it a much later arrival, perhaps common only to a handful of primates, or perhaps unique to humans? Evolutionary questions of this kind are notoriously difficult to answer, in part because, when it comes to psychological traits, the fossil record is less than helpful. The favored approach, then, is to follow so-called "cladistic" methodology and make inferences about the capacities of a common evolutionary ancestor from the characteristics of extant species that are presumed to be descendants of that ancestor. Of course, a given capacity might have evolved independently in the modern day members of a particular clade (a family of species including all descendants of a common ancestor) but, on grounds of parsimony, the assumption is usually made that, if the descendants have it, the common ancestor probably had it too.

Where to look? There are plenty of intelligent species around who conceivably might have a theory of mind (e.g., dolphins and other cetaceans) and there are plenty of highly social species around for whom an appreciation of intentionality might have proved useful (e.g., elephants, wolves, etc.). But our methodological and conceptual headaches will be relieved to some extent if we choose close relatives of our own species with whom we are known to share many relevant psychological capacities: i.e., non-human primates. Which primates? An influential working hypothesis in this area (formulated by Gordon Gallup) is that members of the great ape-human clade might be the only species capable of appreciating intentionality (and mental states in general). The argument is essentially simulationist and it runs as follows. The great apes are the only species known to have the capacity for self recognition. There is no convincing evidence that any other species, including lesser apes (e.g., baboons) and other primates (e.g., various monkey species), has this capacity. A capacity for self recognition implies some form of self concept, possibly one that includes the ability to reflect on one's own mental states. Species with such a concept might therefore be in a position to use the self to model the mental states of others. Hence, we should expect to see theory of mind abilities only among the great apes.

Some evidence concerning the outcome of that expectation is presented in this week's readings. The Premack & Woodruff target article is a seminal piece that has been extremely influential in shaping more recent thinking on these issues in the comparative and developmental literatures. Coming at a time when many comparative psychologists were still uncomfortable attributing minds to non-human species, let alone theories of minds, their ideas were clearly quite radical. Unfortunately, few were convinced by the authors' rich interpretation of their findings but the article nevertheless stimulated a lot of thinking about how one could convincingly test for the presence of a theory of mind. In their commentaries, three philosophers (Dennett, Bennett, and Harman) independently suggested that an appreciation of false belief represents a kind of litmus test for theory of mind, and this insight was subsequently pursued vigorously by Wimmer & Perner in empirical studies with children. Incidentally, prior to Wimmer & Perner's studies there had been two earlier, independent, demonstrations of the false belief phenomenon. The relevant papers (one by Gordon & Flavell, the other by Johnson & Maratsos ) both appeared in 1977 but the findings languished because the theoretical context that Premack & Woodruff helped to shape was not yet in place. This, I think, is a particularly compelling illustration of the old adage, "If ya wanna get ahead, ya gotta have a theory."

The two more recent pieces on apes' appreciation of intentionality by Povinelli et al. and by Call & Tomasello represent conceptually and methodologically stronger approaches to this topic, although they leave open many questions in these respects. Some of the questions we might consider in relation to these papers, and the earlier Premack & Woodruff findings, are the following. Suppose the apes succeeded on these tasks, would we then have convincing evidence of an appreciation of intention or intentionality? Would plausible, lower-level interpretations of the findings be possible (either because of flaws in the methodology or conceptual underdetermination)? Conversely, suppose they failed, would that be convincing evidence of the absence of the relevant capacities? Or are there features of the methodologies that impose unfair hardships on these species? (If looking at these issues across different cultures is fraught with difficulty, then that difficulty has to be an order of magnitude greater across species). On first blush, the findings of Povinelli et al. and Call & Tomasello appear to be quite different, but are they really that different? How strong are the Call & Tomasello data? How much influence do the data from a single orangutan, Chantek, have on Call & Tomasello's interpretation? (Incidentally, some of you may have noticed Chantek on CNN last week -- see <a href = "http://cnn.com/EARTH/9711/28/signing.orangutan.ap/index.html">here</a>. What does Chantek's performance suggest about the role of experience, of culture, in all this? From a certain perspective, convincing data from even a single animal may be enough. It would certainly suggest that the capacity was there in some (latent) form in the species as a whole. The common ancestor might then have had the capacity which, for idiosyncratic reasons, went dormant in one species and rampant in another. Clearly, there are many interesting possibilities here.

Finally, the Povinelli & Godfrey paper is a more speculative piece raising many intriguing questions that we will want to discuss concerning the relation between the evolution of theory of mind and the evolution of altruism and ethical systems. I will let their argument speak for itself while providing just one addendum. Their interpretation of the attributional capacities of chimpanzees now seems to have been somewhat overoptimistic (Povinelli would certainly now agree). To what extent, if any, does this damage the argument that theory of mind capacities play a central role in the emergence of certain forms of altruism?