Yes, We Can: Get from the State View to the Content View

In my previous post, I referred several times to the state view/content view distinction. As has been argued by authors such as Byrne (2005) or Crowther (2006), the distinction is problematic for nonconceptualists to the extent that they want to make a claim about perceptual content. For central pro-nonconceptualist arguments such as the argument from fineness of grain or the argument from animal and infant perception show, at most, that there are mental states that a subject can be in even if she doesn’t possess the pertinent concepts. They have no bearing on the kind of content that these mental states have.

In my view, we should think of nonconceptual states as states that one can undergo without exercising all of the concepts needed to specify their contents. By contrast, conceptual states are states that one can undergo only if, in undergoing them, she exercises all of the pertinent concepts. Given this exercise version of the state view, conceptual content-bearing mental states have conceptual contents (viz. Fregean propositions) and nonconceptual content-bearing mental states have nonconceptual contents (viz. scenario contents). – Or so I’ll argue now. I’m trying to distill a rather complex argument from the book here, so if anything remains unclear, feel free to ask about it in the comments!

I have to start with the theoretical purposes that motivate conceptualists and nonconceptualists: Why do they care what kind of content perceptual experience has? Here are three interrelated things they try to achieve. They want to give an account of how our thinking can be constrained by the world. This account has two aspects. One, how can our beliefs about the world be justified? Answer: By rational constraint on belief from the world. This constraint has to go through perception, so we need rational constraint on belief from perceptual experience. For this, experience itself has to have content that is rationally constrained by the world.

Two, how can our thought about the world be about the world in the first place? Answer: By our being in touch, perceptually, with the world. Conscious perceptual experience that presents the world to us is a necessary condition for content-bearing empirical thought. Both aspects of the account presuppose that we are dealing with conscious perceptual content, with how the world strikes the perceiver.

This relates to the third aim of conceptualists and nonconceptualists: They want to give an account of perceptual content that does justice to the phenomenal character of experience, to how the world is perceptually given to the subject.

Let me here focus on the epistemological aim and show how this gets us from conceptual states to conceptual contents (and from nonconceptual states to nonconceptual contents). I’ll start with conceptual states. Often, in justifying a belief, a subject will perform inferences which move her from one occurrent belief to another; these inferences require her to exercise the concepts needed to specify the contents of the relevant beliefs. For instance, when Dagmar moves from the belief that this cat is orange and the belief that cats are animals to the belief that some animals are orange, she thereby exercises, in different combinations, conceptual abilities to think of this cat, to attribute the property of being orange to things, to think of cats, and to think of animals. The particular conceptual abilities that Dagmar exercises, in the particular combinations that she exercises them in, determine both the contents of her beliefs and whether her transition from the first two beliefs to the third is able to transmit justification to the third belief.

So, when we decide on which contents to ascribe to her beliefs, we have to respect which conceptual capacities she exercises, and how she combines them. If we do this, we will end up with propositions that are just as finely grained as her conceptual capacities. Also, the inferential transition she performs will be matched by the logical relations between the propositional contents of her beliefs. We will end up with Fregean propositions, the paradigmatic conceptual contents. Fregean propositions for conceptual mental states further ensure that we don’t make mistakes about which of her beliefs are justified for her – by either putting too much or too little detail or structure into her belief contents. (For instance, we might end up with this mistaken claim: Mary Jane is unaware that Peter Parker is Spiderman; yet her inference from her beliefs that her neighbor is Peter Parker and that Spiderman is a superhero to the belief that her neighbor is a superhero transfers justification to the latter belief.) Generally then, when we ascribe belief contents, we should respect the subject’s conceptually shaped take on the world.

But what about perceptual experiences? Shouldn’t we ascribe Fregean propositional content to them in virtue of similar epistemologically-minded reasoning? After all, they justify beliefs, and the contents we ascribe to them should respect the way in which we support our beliefs by what we perceive.

In my view, the answer has to depend on whether one is a conceptualist or a nonconceptualist. Conceptualists, who hold that conceptual abilities are employed in experience just like they are in belief, will happily answer ‘yes’. By the same reasoning I presented above, we have to ascribe Fregean propositions as perceptual contents, or so they will say.

According to nonconceptualism, on the other hand, perceptual experience doesn’t require its subject to exercise (all of) the pertinent concepts. So, one motivation for ascribing Fregean propositions as perceptual contents drops out of the picture. To do justice to the perceiver’s perceptual take on the world when ascribing perceptual content, there’s no need to look to the particular combination of conceptual abilities exercised in undergoing an experience. Rather, we should go by the phenomenology of perceptual experience when fixing its content: It should reflect the subject’s experience of being located right in the middle of the perceived world. This can be best achieved by ascribing scenario content to perceptual experience, I believe. (Cf. Peacocke 1992)

Still, this leaves the worry about perceptual justification. Here, I can only hint at a response: Constraining our content ascriptions by the subject’s perceptual perspective actually helps us to avoid mistaken claims about which beliefs are justified for her. To give a cheap example, if an object is hidden from my sight, I’m not in a position to have visually justified beliefs about it. So, to ascribe a content to my visual experience involving the object would be a mistake. In general, perception’s contribution to the justification of belief is limited by how the world actually strikes perceivers in conscious experience.

Obviously, there are further important worries about perceptual justification for the nonconceptualist (which I address in the book). The take-home message for this post is that the debate between conceptualists and nonconceptualists is shaped by the theoretical aims presented above, viz. to give an account of justified belief, empirical content, and perceptual phenomenology. In the light of these aims, whether the subject has to exercise the pertinent concepts in undergoing a mental state is the decisive factor for whether conceptual or nonconceptual content ought to be ascribed to it. For only in this way can we ensure that we end up with a correct picture both of the subject’s justified beliefs and of her perspective on the world.


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

Crowther, T. (2006), ‘Two Conceptions of Conceptualism and Nonconceptualism’, Erkenntnis 65, 245–276.

Peacocke, C. (1992), A Study of Concepts, MIT Press, Cambridge, MA.

Back to Top