Re: [quickphilosophy] Fodor on Concepts III: Other Arguments Against BCP

  • From: wittrsl@xxxxxxxxxxxxx
  • To: quickphilosophy@xxxxxxxxxxxxxxx
  • Date: Tue, 24 Aug 2010 23:58:24 -0700 (PDT)

Hello Walto & quick-fill group:
For his second obliteration of Bare-Bones Concept Pragmatism (BCP) in his 
article "Having Concepts" Fodor argues that BCP's epistemic account of 
possessing concepts fundamentally conflicts with the compositionality of 
thought and language. This, Fodor has argued, is essential for explaining the 
productivity and systematicity of thought and language.
BCP, in Fodor's characterization, comprises both sortability and inference, 
and, really, I think, nothing else; these are necessary and sufficient criteria 
as Fodor relates them. Again, as I think Neil pointed out, it's not for Fodor 
to state their conditions, but for the conceptual pragmatists to do the same. 
That's one thing that is glaringly missing from this argument. Where are the 
citations that somebody actually puts forward BCP? I see none.
But,OK, BCP asserts (so Fodor) a sorting capability within concept possession. 
[rla aside: Does Fodor not accept this? Is it possible that the concept of a 
dog as such might not be sortable from a cat as such, and yet the two concepts 
should nonetheless be distinct?] Fodor argues that if BCP is correct, then to 
have a concept C is to be able to sort objects in its extension.
Here is Fodor's argument:
1. OK, what about dogs?
2. You can't pick out dogs that are relativistically outside our perceptual 
scope. There might be dogs that are so far away that the finiteness of the 
speed of light precludes our seeing them and discriminating them from non-dogs. 
[rla: What an unbelievably specious argument. I'm appalled.]
3. [But no one in BCP claims that we have to sort physically unknowable 
objects. Was the possible dog outside our lightcone whelped by a dog inside our 
cone? If not, why even suggest that it's a dog? Why even suggest that it 
exists, let alone make this imaginary thing a crucial entity for deciding our 
intellectual dilemma? This is really a quite specious line of argumentation. 
What does BCP say? It can only be fresher.]
4. The BCP requires us to identify dogs that died a long time ago and for which 
there are no traces. So, for instance, we'd have to be able to say that the 
"dogs" that darted around the Achaean camp in the Iliad were really dogs, not 
jackals, in order for BCP to be correct. These dogs/jackals are gone. We can't 
know anything about them. Therefore, the epistemic account of concept 
possession is false.
5. We're not done yet. If it is required [rla: who said it was? Fodor has 
quoted nobody on these goofball points]  that dog owners distinguish dogs from 
anything else, then only God has this ability, and if one is an atheist, then 
indeed no one can distinguish dogs from anything else [rla: because you might 
make a mistake with a jackal or a dingo or a fox or even a wolf]; thus, you 
can't sort dogs. Thus, you have no concept of dogs.
6. So, maybe you need good instances of dogs to be able to identify them? or 
maybe you need favorable conditions? But, so Fodor argues, neither the quality 
of the conditions for identifying dogs nor the conditions which might be 
favorable for identifying them are compatible with compositionality; BCP (which 
is not compositional) cannot be a good candidate for explaining the possession 
of concepts (which is compositional).

(My errant thought/language: Does it not seem odd that thought and language are 
taken to be isomorphic in this argument? Are they really? Navajo vs. Sanskrit?)

--- On Tue, 8/24/10, Ron Allen <wavelets@xxxxxxxxxxx> wrote:

From: Ron Allen <wavelets@xxxxxxxxxxx>
Subject: Re: [quickphilosophy] Fodor on Concepts III: Other Arguments Against 
To: quickphilosophy@xxxxxxxxxxxxxxx
Date: Tuesday, August 24, 2010, 10:21 AM



      Hi Walter & quick-phil group:
Fodor's second refutation of Bare-Bones Concept Pragmatism (BCP) begins in 
section 3.2 of his article "Having Concepts." He introduces the ideas of 
productivity and systematicity. Human language and thought are both productive 
and systematic.
That language and thought are productive means that "there are approximately 
infinitely many thoughts that one can think." There's a footnote to this odd 
formulation, where Fodor rejects as a red herring the possible counterpoint of 
"performance constraints". My guess is that someone might make the obviously 
correct reply that a human being cannot think infinitely many thoughts. A 
finite brain and a finite lifetime in a finite universe ought to be constraint 
enough, but Fodor just blithely ignores the obvious here. Probably, Fodor is 
just making the assertion that given a few thoughts, one can combine them or 
extend them in simple ways, and this results in a new thought. So, OK, we can 
probably live with that.
That language and thought are systematic means that "there are certain 
symmetries of expressive power in human conceptual systems/languages." 
Fodor does not provide a list of the symmetries...perhaps this has been done by 
someone else, say Chomsky. What he does is to give a rather poor example of 
such a symmetry in language & thought: "John loves Mary." If a mind can 
entertain the thought that John loves Mary, then, because of systematicity, it 
can entertain the thought that Mary loves John. Well, OK, we can probably live 
with this example, but almost any other subject-transitive verb-direct object 
sentence in a natural language is going to fail the symmetry relation that 
supports systematicity. For example, what if I think "John loves the color 
blue." Do I get to think that "The color blue loves John"? I would say, yes, 
sure, you can formulate such a thing, but it doesn't make much sense. But, 
maybe that's what Fodor wants, just
 some vague set of rules for flipping around language elements.
I think one could argue that this symmetry is exceptional, applying to some 
language elements, but not others. Indeed, different natural language families 
have quite different rules for combining concepts. It seems natural to think of 
verbs as have past, present, and future tense. But, in point of fact, if you 
talk about tense like that, it marks you as someone who speaks one of the 
languages in the Indo-European family. Other languages don't have this. For 
instance, in Navajo, you express your thoughts as something that you learned 
from someone else, something that you were a direct witness to yourself, or 
something that is a timeless truth. We could go on and on. But, I guess the way 
to get anywhere is just let Fodor have his proof by example, not quibble about 
how actual languages work and how they don't work the same way, and see where 
we grind to a halt.
The next idea Fodor introduces is compositionality. Language and thought are 
systematic and productive because they are compositional. "There is, for each 
natural language, a finite set of 'lexical primitives' (words, more or 
less) and a finite system of constructive principles that the primitives fall 
under. The latter [rla: he means the constructive principles] apply recursively 
and may iterate without bound."  Fodor says that compositionality is required 
to explain productivity and systematicity, and that therefore compositionality 
is non-negotiable. If someone proposes an explanation of concept possession and 
that explanation is incompatible with compositionality, then that theory of 
concept possession [rla: you can see that Fodor is going to blow away BCP 
because of this] must be rejected.
I'll look at the argument in more detail in my next post.

--- On Tue, 8/24/10, walto <calhorn@xxxxxxx> wrote:

From: walto <calhorn@xxxxxxx>
Subject: [quickphilosophy] Fodor on Concepts III: Other Arguments Against BCP
To: quickphilosophy@xxxxxxxxxxxxxxx
Date: Tuesday, August 24, 2010, 7:13 AM







Other related posts: