Cognitio 2011

Abstract: On “Animal Concepts: Content and Discontent”, Charter and Heyes (1994) claim that the research project on attributing concepts to non linguistic animals has no sense, since it cannot be articulated with any of the (psychological) available theories of concepts (definitional, prototypes and exemplars theories). Particularly, Charter and Heyes argue that the idea of concepts has not been yet successfully detached from the idea of natural language. Thus, we don’t have a coherent explanation of animal concepts. They recognize that prototype theory can be formulated in non linguistic terms; nevertheless, as they claim, animal prototypes cannot be distinguished from stimulus generalization, so we cannot test if a non linguistic animal has (prototype) concepts.

Over the last years, a philosophical debate on animal concept has arisen. Nevertheless, no significant answers to Charter and Heyes’ strong criticism have been put forward. This debate is strongly linked to empirical research and ideas, both coming from cognitive ethology. And also, this debate has been centered on the possibility that, even without language, animals can satisfy certain required conditions for concept possession such as generality, identification, and inferential capabilities (Camp 2009, Glock 2009, 2000, Newen and Bartels 2007, Millikan 2000, Bermúdez 1998). Although it seems rare, Davidson, Brandom, Evans and Peacocke –all of them defenders of a linguistic perspective of concepts– are among the philosophical precursors of these debates.

We are in front of two parallels debates: On one side, a debate about the psychological nature of concepts; on the other, a debate about concepts attributions to animals. According to philosophical and psychological literature, definitional, prototypes and exemplar theories are theories about the psychological structure of concepts (Machery 2009, Laurence and Margolis 1999, Charter y Heyes 1994). Besides, all of them presuppose that concepts should be understood in terms of particular representational entities. In this sense, they would be theories about the way mental representations involved in higher order cognitive processes (such as categorization and reasoning) are structured.

But, as I claim, the problem of animal concept attribution must be previously elucidated by a theory of concepts possession, and it should not depend on having an appropriate account inside a previous psychological theory of (human language dependent) concept’s structure. In other words, we must focus, first, on the question of animal concept attribution and, only later, on what kind of structure those concepts have. In this sense, some methodological consequence on psychology of concepts should follow from the debate on animal concepts.

Also, as I argue, the actual state of the discussions on animal concepts supports certain philosophical labor neglected by some recent forms of reductionism (Machery 2009, Laurence y Margolis 1999, Charter y Heyes 1994); that is, the elucidation of the criteria for concepts possession; which is being performed not just on analytical ground but in articulation with some recent theorization on cognitive ethology, compared psychology, and evolutionary anthropology.

To support my proposal, I adopt a pluralist notion of concepts (Weiskopf 2009). Pluralism relies on the existence of higher order generalizations besides fine-grained psychological account of concepts; this implies that concepts should be understood in terms of a super-functional-kind, which includes theories, prototypes, exemplars, and so on. My thesis is that this notion of concepts is closely related to the philosophical one, and can it be employed to identify and characterize the abilities required to possess concepts (with independence of psychological notion of concepts) and, at the same time, to differentiate and articulate –at least heuristically– the problem of concept structure and the problem of concepts possession.