Abstract

One of the problems that arises in attempts to attribute thoughts to non-human animals is that is difficult to find sentences of human language that accurately express the content of those thoughts. I argue that semantic theories that do not depend on an understanding of propositional content as being linguistic in nature can provide a helpful perspective from which to approach questions about the content of the beliefs of non-language-using animals. In particular, I consider Robert Stalnaker's possible worlds semantics and the situation semantics of Jon Barwise and John Perry. I argue that both of these theories provide ways of understanding the content of beliefs on which it is far more plausible that non-human animals have beliefs than theories that take propositions to be linguistically structured. I further argue, however, that Barwise and Perry's account is ultimately more helpful than Stalnaker's and avoids some significant problems faced by possible worlds theories. In my approach to these topics, I consider the plausibility, on various influential theories of mind, of the idea that animals have beliefs and I provide an exposition of Jonathan Bennett's account of what types of animal behavior justify the attribution of beliefs. I conclude by arguing that the combined work of Bennett, Stalnaker, and Barwise and Perry can be used to address a line of reasoning I refer to as the specificity argument. According to the specificity argument, if an animal cannot express its beliefs in language, it is impossible to attribute content to any belief one might attribute to it with enough precision to make sense of claims that the animal has one belief rather than another. It is then concluded that non-language-using animals do not have beliefs. I use the work discussed above to show that the specificity argument does not, by itself, show that non-human animals do not have beliefs. As well, the approach to understanding content that I consider will be helpful in considerations of the content of non-language users in general, and is a promising first step in the assessment of many arguments against the claim that animals have beliefs