Friday 14 October 2016

Distributed Verification: Semantic Deference Needn't Bottom Out in a Non-Deferential User

Woodward in 'Reference and Deference' writes:
Linguistic deference and conceptual deference are widespread phenomena. However, some philosophers say that deferring has to stop at some point, for if everyone were deferential with respect to a given word or concept no one would ever succeed in attaching a definite content to it. That thesis is the stimulus for this paper.

Fodor certainly holds the thesis. In his latest book Concepts, Fodor (1998, p. 154) says, ‘Adherence to conventions of deference couldn’t be a precondition of conceptual content in general, if only because deference has to stop somewhere; if my ELM concept is deferential, that’s because the botanist’s isn’t’. (cf. Fodor 1994, p. 33).
That thesis, or almost that thesis, is the stimulus for this blog post too. But my target is different from Woodfield's. His target - that which he argues against - is literally the thesis that deferring has to stop at some point, in some sense. My target is rather the thesis that deferential concept use can only terminate successfully when someone has the concept and uses it non-deferentially.

So again, while Woodfield's paper may seem to be upholding the same sort of view as I am here, I don't think it is the same. His point seems to be that we need not think that deference must stop, in that a bunch of experts who are good at different things may go on being disposed to defer to each other indefinitely. But my point here is, not that it may sometimes never be finally settled whether a has P, but that it is possible for such a question to be settled without it being the case that any single person's use of 'has P' is non-deferential. 

Imagine three chiefs in a tribe, each one good at certain verification tasks. Everyone in the tribe agrees that for an animal to count as having property P, it needs to appear Xish to Chief A, Yish to Chief B and Zish to Chief C. 

This is, in a way, especially clear if 'has property P' just means 'appears Xish to Chief A, Yish to Chief B and Zish to Chief C'. But that this sort of situation could arise without such a disjunctive meaning having to do with three different people's phenomenology is worth realizing as well, since it may help us see to what extent this sort of distributed verification may happen with more normal, real-world language and concepts.

When someone in this community wants to know if an animal has P, they may put it to the chiefs. In a case where the verdict ends up being that the animal does have P, we might imagine the process going as follows. The questioner asks the three chiefs, who approach the animal together, perhaps from different angles. Chief A says 'It has P if Chiefs B and C have no objection', Chief B says 'It has P if Chief C has no objection', and then Chief C just nods and says 'OK, I guess it has P then!' and everyone is happy.

This is just a quick, initial attempt to show that this kind of 'distributed verification' is possible, and that therefore semantic deference needn't bottom out in a non-deferential language user.


Fodor, J. (1994). The Elm and the Expert. Cambridge, MA: MIT Press.
Fodor, J. (1998). Concepts. Oxford: Clarendon Press.
Woodfield, Andrew (2000). Reference and deference. Mind and Language 15 (4):433–451.

No comments:

Post a Comment