Wilfrid Sellars famously distinguishes between the scientific and the manifest image of human (1962). In the manifest image, human subjects have mental states such as beliefs and desires that are intentional in the sense of being about things. In the scientific image, humans are explained under causal laws, no different from the rest of the natural world. Given this distinction, naturalizing intentionality can be characterized as the attempt of finding intentional states and their properties in the causal structure of the natural world. As we take a quick look at the promising ways of naturalizing intentionality, we see that they take the form of realist or anti-realist attempts. A realist, Jerry Fodor, for example, argues that naturalizing intentionality requires finding explicit representations in the brain (1975, 1987). These representations are syntactic linguistic structures that gain semantic properties based on how they are combined. In this account, for each belief or thought someone has, there is a sentence-like neural token in one's brain. By contrast, an anti-realist, such as Paul Churchland, argues against the Fodorian ideal based on the fact that our best neuroscientific models (i.e. connectionism) do not posit such explicit representations in the brain. Therefore, Churchland thinks that we should give up on intentional properties that the manifest image of human appeals to (1989).

These two views share a common assumption:

For an intentional state S (whether conscious or unconscious) to be real, there must be a “content-bearing” physical (neural) particular N corresponding to S.

I argue that the common assumption is false. Everyday intentional states that the manifest image posits and neural representations posited by the scientific image are independent of each other. Let me demonstrate this claim briefly.

For the common assumption to be true, conceptions of mental representation in cognitive science and everyday intentional explanations must have the same conceptual origin. This is indeed the case for Fodor. Everyday cases of intentionality are the reason why he posits semantic properties in the brain (1987).

It is not the case for the post-Fodorian cognitive science, however, that positing “neural” representations is based on accommodating intentionality in the everyday sense. A good way to see this is to look at the distinction between personal and sub-personal level representations. While personal-level representations are conscious, occurrent first-person experiences with intentional properties, sub-personal level representations are unconscious, causally/functionally individuated “information-bearing” neural particulars. Fodor explicitly states that his posits of sub-personal representations constitute the personal level representations that correspond to propositional attitudes (1987, p. 24). On the other hand, in contemporary cognitive science sub-personal level representations are posited entirely because they are explanatorily useful in causal explanations of functionally analyzed information-processing (Clark, 1997). This shows that, while personal representations are posited exhaustively pre-scientifically in the manifest image, sub-personal level representations are posited exhaustively scientifically in the scientific image.

The upshot is that representations in the level of neural particulars posited by cognitive science are conceptually independent of representations in the level of everyday intentional phenomena. Therefore, the common assumption is false. What makes an intentional state real is not a content-bearing neural particular as the physical counterpart.

If it is true, my critique entails two things.

1) One can be a realist about intentional properties in the everyday sense without commitment to content-bearing physical particulars as causally potent counterparts of each intentional state.

2) The relationship between intentional explanations (and their posits) and scientific/causal explanations (and their posits) of human cognition is not straightforwardly clear.

One important lesson we should draw from this story is that the common assumption is shaped by what one takes intentionality to be. If one takes intentionality to be representational in the Fodorian sense, then it is straightforward to look for information-bearing physical particular counterparts in the brain (by inference to the best explanation). In that sense, what one takes intentionality to be determines where one looks for it in the natural world. To reflect on the second outcome, therefore, I suggest a reconsideration of concepts of intentionality and mental representation; two concepts that are often taken to be equal in the recent literature of philosophy of mind.

First, intentionality. As opposed to the popular view that intentional states are mental representations, we shall introduce a notion of intentionality which has always been there from the start: intentionality as directedness. Intentionality as directedness does not insinuate that intentional properties must be found in the brain as properties of representational states. It reveals the possibility that the things that conscious agents are directed at need not be mysterious abstracta-or mental content with mysterious properties. Directedness in this sense can be construed as a directedness to things themselves - devoid of representations that invite physical counterparts as their explanation.

Second, mental representation. If cognitive science, distinctively of our everyday sense of intentionality, somehow proves or strongly implies that information-bearing physical particulars are required for information processing, we can accept them not as representations with intentional/semantic properties but as what we might call enabling representations that merely helps the agent to achieve her final intentional state with semantic properties. Enabling representations bear information naturalistically and non-propositionally. One promising candidate for such enabling representations is Peter Godfrey-Smith's sender-receiver model (2014).

In the final part of my presentation, I suggest a particular action-based picture of “intentionality as directedness”, mainly following Rowlands (2010), as a fruitful approach to the relationship between the manifest and the scientific image of human. 

References

Churchland, Paul, 1989, A Neurocomputational Perspective: The Nature of Mind and the Structure of Science. MIT Press.

Clark, Andy, 1997, Being There: Putting Mind, Body, and World Together Again, Cambridge, MA: MIT Press.

Fodor, Jerry, 1987, Psychosemantics, Cambridge, MA: MIT Press.

–––, 1975. The Language of Thought, Harvard University Press.

Godfrey-Smith, Peter, 2014. Sender-Receiver Systems within and between Organisms. Philosophy of Science 81 (5):866-878.

Rowlands, Mark, 2010, The New Science of the Mind: From Extended Mind to Embodied Phenomenology, MIT Press

Sellars, Wilfrid, 1962, Philosophy and the scientific image of man. In Robert Colodny (ed.), Science, Perception, and Reality. Humanities Press/Ridgeview. pp. 35-78.