What the Phenomenal Concept Strategy Is Not David

What the Phenomenal Concept Strategy Is Not David Papineau Rio-2013 Conference on Phenomenal Concepts

Plan 1. Phenomenal Concepts and Mary 2. The Unimportance of the Experience Thesis 3. The Explanatory Gap 4. Intuitive Dualism 5. Explanations of Intuitive Dualism

Phenomenal Concepts and Mary Here’s a simple view. Phenomenal concepts are concepts of conscious properties that are a priori distinct from any material concepts of those properties. (‘Material’ here = concepts that Lonely Mary could have. ) This is all we need to respond to the original Knowledge Argument. Mary comes to know something new at the level of concepts, not reference.

Phenomenal Concepts and Mary This doesn’t require us to hold the ‘Experience Thesis’— phenomenal concepts can only (normally) be possessed by those who have had the relevant experiences) nor any kind of ‘Use-Mention Thesis’— exercises of phenomenal concepts (normally) involve (versions of) the relevant experiences). These extra theses might be needed to deal with Kripkean/conceivability/explanatory gap arguments—I’ll come back to that—but not the Knowledge Argument itself.

Phenomenal Concepts and Mary (Can we really detach the Knowledge Argument from these further arguments? But that’s my memory of the history. First the knowledge argument made us bring in distinct phenomenal concepts. Then later further arguments appealed to features of these concepts to argue that there is something problematic about a posteriori identities involving phenomenal concepts. . . If this is right, then from our perspective thirty years later, the Knowledge Argument doesn’t look like much of an argument. But think back to a time when it was taken as given that all concepts of mental states were logical behaviourist/analytic functionalist concepts. Given this, Mary should have known everything. Here’s what we used to say to Kripke. Sure, if you focus on specific brain states, then it will seem possible that you could have c-fibres without pains (because the c-fibres might be wired up differently). But if you specify that all the relevant physical states are the same, including the behaviour, then it’s not even conceivable, let alone apparently possible, that a physical duplicate could not be in pain. Mary knocked out this response. )

The Unimportance of the Experience Thesis A focus on the Experience Thesis puts the wrong emphasis on the phenomenal concept strategy. For a start, it can make it seem that the crucial step in explaining Mary’s ignorance is that she hasn’t yet had the experience that yields the new concept. And then it can seem that it is bad for the phenomenal concept explanation that Mariana/Experienced Mary— who has had the experience and does have the new concept—is still ignorant. But in truth Mariana drives home the phenomenal concept strategy (against the ability hypothesis). It’s not the lack of the concept, but its a priori distinctness, that the phenomenal concept explanation hinges on.

The Unimportance of the Experience Thesis A focus on the Experience Thesis can also make it seem that the phenomenal concept strategy, along with the Knowledge Argument that prompts it, is undermined by Burgean considerations. But even if we grant the strong Burgean claim that no concept can satisfy the Experience Thesis, and so that Mary herself won’t be ignorant of anything, it is easy enough to resurrect variants of the Mary argument that are as effective as the original and again call for the phenomenal concept (in my sense) strategy. Michael in fact gave us one yesterday. Suppose Mariana passes her concept around, but nobody is yet in a position to know which (materially conceived) experience it refers to. How so, if everybody already had all material knowledge? Well, their knowledge is new at the level of concepts, in virtue of a concept that is a priori distinct from all their material concepts. This mightn’t be a phenomenal concept according to the Experience Thesis, but it is in my sense. (Another such example would be a community some of whom who were born with a phenomenal concept, but hadn’t yet figured out which material concept it co-referred with. )

The Explanatory Gap Joe said yesterday (I wrote this as he was saying it) that the phenomenal concept strategy is a response to the conceivability/explananatory gap argument. I certainly agree that we have ‘explanatory gap moments’. But I (as a phenomenal concept strategist) don’t explain these as due to the special features of phenomenal concepts. I don’t think that the right response to the conceivability/explananatory gap argument is anything much to do with the features of phenomenal concepts. (In fact I don’t really understand how anybody thinks that the special features of phenomenal concepts account for the ‘explanatory gap’. OK—I know that Nagel’s famous footnote and Hill and Mc. Laughlin offered to account for the ‘appearance of contingency’ in terms of the special features of phenomenal concepts. But what are they really doing except just repeating that M = B identities are a posteriori? )

The Explanatory Gap Let me go more slowly and explain how I think of these matters. There is the puzzlement contrast (something seems unexplained specifically in the mind-brain case). Then there is the derivability contrast. I certainly agree that M = B identities aren’t a priori derivable from the physical facts. I don’t of course (along with other Type Bs) think that this discredits physicalism. But I diverge from most other Type Bs (though not Joe anymore) in denying that the puzzlement contrast is just the derivability contrast. There’s the familiar argument that the derivability lack covers a lot of cases where we aren’t puzzled.

The Explanatory Gap Here’s a different line of argument. Explanations of particular and general facts standardly cite causes, which is not at all the same as offering a priori derivations of anything. Come to think of it, what exactly is it that is supposed to get explained when we can do the a priori derivation, but not otherwise? (a) That M = B? But identities don’t need explaining. (b) Why we should believe M = B? Well maybe we do often explain why we should believe identities by showing how B fills the same causal role as M. But (contra Jackson) this doesn’t require that our knowledge of M’s causal role is a priori. (c) Why M has the causal role it does? Well M = B can help explain this, but it will do that just as well even if it is not a priori derivable.

Intuitive Dualism I have a straightforward view. The puzzlement contrast is just that we can’t intuitively believe M = B identities. (And so of course we think that there is something to be explained there. ) Evidence (1). Think of how even physicalists are prone to talk about such ‘identities’. They say that brain processes ‘generate’, or ‘yield’, or ‘cause’, or ‘give rise to’ conscious states. . . Evidence (2). Why does it strike us physicalists as so much as possible that zombies can lack pains? Compare the reaction of someone who believes Cicero = Tully to the possibility of ‘their’ being distinct . The best diagnosis of the ‘appearance of contingency’ is that it is an intuition of actual falsity.

Possible Explanations for Intuitive Dualism a. Culture. Maybe the intuitive dualism is an upshot of our specific religiousmetaphysical heritage. (Are all times and cultures dualist? I suspect yes. . ) b. Modules Paul Bloom (Descartes’ Baby) and many cognitive anthropologists hold that intuitive dualism stems for the incommensurability of our Understanding Mind-Reading and Folk Physics modules. This explanation arguably overgeneralizes c. The antipathetic fallacy I used to argue, from the use-mention feature of phenomenal concepts, that it is natural to think of the ‘B’ side of M =B identities ‘leave out’ the feeling itself. Pär Sundström argues that this overgeneralizes too. d. Aspirations to transparency Maybe people feels that phenomenal thinking is so close to its referents that it must reveal all their essential properties, but it doesn’t reveal them to be brain states. . . e. Resistance to merging Andrew Melnyk suggests that normally when we embrace an identity, we ‘merge the files’, but that something stops this happening in the M = B cases.

Possible Explanations for Intuitive Dualism I’m undecided between these explanations. Note that there is no reason why more than one—indeed all of them—shouldn’t apply. Note also that there is no reason to assume anything about phenomenal concepts, apart from their a priori distinctness from material concepts, when we embark on the search for such explanations. Moreover none of the explanations suggested above (except c and perhaps d in appealing to the use-mention feature) does appeal to any of the standard extra features of phenomenal concepts attributed to them by philosophers.

Conclusion Phenomenal concepts are nothing more than concepts that are a priori distinct from material concepts. THE END