Concept Possession Isn’t Good Enough

Typically, nonconceptualism is introduced in terms of concept possession. Take for instance, the first claim from the recently updated SEP entry on nonconceptual content:

The central idea behind the theory of nonconceptual mental content is that some mental states can represent the world even though the bearer of those mental states need not possess the concepts required to specify their content.(Bermúdez and Cahen 2015)

Similarly, Byrne (2005):

Mental state M has non-conceptual content p iff it is possible to be in M without possessing all the concepts that characterize p.

For a final example, take Toribio (2008) on state nonconceptualism:

For the state nonconceptualist, it thus seems consistent to hold that both perceptual experiences and beliefs share the same (conceptual) content, but that for a subject to undergo a perceptual experience, the subject need not possess the concepts involved in a correct characterization of such content. (All emphasis mine.)

It is initially plausible to define nonconceptual content or states in terms of concept possession. My infant daughter is apparently able to see the blue ball (she crawls towards it and grasps it in her hands); but she arguably isn’t cognitively sophisticated enough to possess the concepts blue or ball. So, her visual experience of the blue ball is a mental state that can be had without possessing these concepts, and its content is – at least in that sense – nonconceptual.

However, I believe that this understanding of nonconceptualism is too weak. For whether a mental state can be had only if a certain set of pertinent concepts is possessed doesn’t seem to tell us a whole lot about the nature of the state, let alone its content. Instead, we should define nonconceptual states in terms of concept exercise or employment.

Here’s an analogy: To have a German bank account, you have to have an address. But this doesn’t mean that possessing an address fixes anything about the nature of German bank accounts. Nor does it determine the nature of the contents of a German bank account. If we want to learn anything about bank accounts with the help of the addresses of their owners, we have to look at the way in which addresses contribute to someone’s having a bank account.

Similarly, facts about concept possession aren’t guaranteed to give us any more than background conditions for having certain mental states, or tell us much of anything about the nature of mental states or their contents. It’s much more promising to look at how concepts contribute (or don’t contribute) to someone’s having a mental state, at how concepts need to be employed (or not) in someone’s undergoing this mental state.

A more direct way to attack the claim that nonconceptual states or contents should be defined in terms of concept possession is to look at Michael Tye’s PANIC theory of phenomenal character (Tye 1995, 2000, 2003a, 2003b). According to his view, the contents of perceptual experience are nonconceptual. At the same time, representations that have these nonconceptual contents have to be poised to have an impact on the central cognitive system. This means that the subject has to possess concepts corresponding to the contents of her perceptual experiences, which she could employ to form beliefs that take up the content of her experiences. That is to say, the subject has to possess concepts that characterize the content of her conscious experiences, which are nonetheless nonconceptual. My point here is not that Tye’s view is correct (also, I wish to be cautious on whether this is exactly Tye’s view). Rather, I think it is coherent and prima facie not implausible to say that the subject has to possess the concepts needed to specify the content of a certain perceptual experience, but that the latter is nonetheless nonconceptual.

A related motivation for rejecting the possession state view is that it is forbiddingly difficult to argue that a mental state has to have a certain kind of content – conceptual or nonconceptual content – because it presupposes (or doesn’t presuppose) the possession of certain concepts. By contrast, it is possible to argue from the fact that a subject has to exercise the pertinent concepts (or doesn’t) in undergoing a mental state with a certain content, to the claim that the mental state has to be of a particular kind. I will present an argument to this effect in the next post.


Bermúdez, J. & Cahen, A. (2015), ‘Nonconceptual Mental Content’, in E. N. Zalta, ed, The Stanford Encyclopedia of Philosophy (Fall 2015 Edition), URL = <>.

Byrne, A. (2005), ‘Perception and Conceptual Content’, in M. Steup & E. Sosa, eds, Contemporary Debates in Epistemology, Blackwell, Malden, 231–250.

Toribio, J. (2008), ‘State Versus Content: The Unfair Trial of Perceptual Nonconceptualism’, Erkenntnis 69, 351–361.

Tye, M. (1995), Ten Problems of Consciousness, MIT Press, Cambridge, MA.

Tye, M. (2000), Consciousness, Color, and Content, MIT Press, Cambridge, MA.

Tye, M. (2003a), ‘On the Virtue of Being Poised: Reply to Seager’. Philosophical Studies 113, 275–280.

Tye, M. (2003b), ‘The PANIC Theory: Reply to Byrne’. Philosophical Studies 113, 287–290.