[Wittrs] Re: Ontologically Basic Ambiguity: Mode of Existence

  • From: "SWM" <SWMirsky@xxxxxxx>
  • To: wittrsamr@xxxxxxxxxxxxx
  • Date: Mon, 22 Mar 2010 02:11:06 -0000

--- In Wittrs@xxxxxxxxxxxxxxx, Gordon Swobe <wittrsamr@...> wrote:
> --- On Sun, 3/21/10, SWM <wittrsamr@...> wrote:
> > His CRA depends on a dualistic notion of consciousness.
> You may think as much, but Dennett certainly does not give us reason to 
> believe it.

He did in that text I transcribed onto the list. You just have to go through 
his argument there. I have done it in a more streamlined way. See below.

Anyway, Dennett's argument is there. You can say it's wrong but then you have 
to show why. What you can't say is that Dennett "does not give us a reason to 
believe it" because that is just factually incorrect.

> In the quote that you provided, Dennett *insinuates* but does not actually 
> claim or prove (because he cannot) that Searle's view equates to Cartesian 
> dualism.

He asks who would believe the CRA's claim that more of the same wouldn't make a 
difference and responds rhetorically that a Cartesian dualist would.

Since the CRA is Searle's argument and no one else's, and Searle defends it and 
obviously finds it convincing, it's pretty clear that Dennett is saying that 
Searle shares the Cartesian dualist conception of mind because that's what it 
takes to swallow the argument.

The context is pretty explicit, even if Dennett doesn't say, in so many words, 
what I have said, that "Searle must be a Cartesian dualist because he thinks 
the CRA demonstrates that what the constituents of the CR can't do in the CR, 
they can't do in any other configuration either."

I have made this argument several times now in a number of pretty clear steps 
so all you have to do is show me where my argument goes awry and that there is 
a way to hold the conclusion of the CRA, BASED ON THE PREMISES AND THE 
DEMONSTRATION OF THE CR ALONE, without thinking consciousness cannot be reduced 
to constituents that aren't themselves, conscious.

(Note that you have to show this based on the CRA, not some other argument, 
since my beef is with the CRA. As it happens I find Hawkins' argument against 
the possibility of using computational programming to produce a human like 
intelligence a strong one, if not, itself, entirely convincing.)

> Dennett's argument takes this form:

So Dennett DOES give us reasons to "believe it"! If you agree that he does and 
are prepared to deal with his argument below, then why would you say he doesn't 
give us reasons above?

> 1) Searle believes the CRA.
> 2) A Cartesian dualist would believe the CRA for such-and-such reason.
> Even if we take 2) as true, it does not follow that Searle believes the CRA 
> on account of his holding to a dualistic world-view.

What Dennett fails to do in that text is make a detailed argument about the 
Searlean position which is probably because it is only marginally relevant to 
his claim. His main point there is that the CRA doesn't work and he offers a 
very detailed explication of why, which boils down to the argument that the CR 
is underspecked based on a conception of consciousness that has no room for the 
possibility that features like understanding may be more properly functions of 
systems than of the constituent processes of such systems. Searle's underlying 
mistake is more of a sideshow in Dennett's explication. On this list, however, 
it seems to be front and center with numerous Searle adherents fighting to 
defend his non-dualistic honor.

Dennett is more interested in developing an account of consciousness and, 
indeed, he really doesn't deal with the CRA until the very end of the book and 
then it's almost as an afterthought.

> Frankly I lost a lot of respect for Dennett based on that quote. It strikes 
> me as a case of intellectual dishonesty.

I think you just don't get the distinction between system and process 
properties that is relevant here. It's about whether a feature like 
understanding is a complex function of certain kinds of systems (many processes 
running together in an orchestral type arrangement) or a kind of mysterious 
property which some brains cause in some unidentified way. Searle drops the 
ball when he gets to brains and that's because he is stuck in a very deep 
contradiction with his own claims via the CRA.

> The CRA stands on its own, distinct from any considerations about dualism or 
> non-dualism.

The problem with the CRA isn't that it requires a dualist presumption as a 
suppressed premise to the argument. If dualism is true then so what?

The problem, rather, is that the CRA is WRONG because the conclusion doesn't 
work without that suppressed premise which, once exposed, is not defensible on 
Searle's own grounds, causing Searle, himself, to fall into self-contradiction 
vis a vis brains vs. computers. Indeed, no one who agrees with the CRA 
acknowledges that suppressed premise is even there and so they ignore it, 
relying on Searle's own explicit denials that he is a dualist as justification 
for doing so. But if it's implicit in his argument, his denials are of no value 
in determining whether it's there or not.

> We can and should consider it a simple logical proof. One needn't have any 
> understanding of philosophy to follow the simple logic of the CRA.

And yet it must finally hinge on the truth of its premises. Searle tells us, at 
one point, that the truth of the third premise is self-evident or trivial or 
some such. As Dennett puts it later on, it seems "obvious". But being obvious 
doesn't make it true as Dennett proceeds to explain.

The best exposition of the third premise goes something like this:

"Syntax by itself does not constitute and is not sufficient for semantics."

This certainly looks obvious on its face. After all, syntax is not semantics 
and when we have an instance of syntax that is never enough to also say we have 
an instance of semantics. The two terms designate distinctly different things 
(syntax = rules being implemented; semantics = things being understood).

So the premise certainly looks true if understood as a denial of identity 
because, in fact, syntax is not the same as semantics and when we have an 
instance of syntax it is never sufficient to say aha, we have an instance of 
semantics! An instance of syntax does not constitute an instance of semantics 
and is not sufficient for us to say we have an instance of semantics.

But the CRA is drawing conclusions about what causes understanding (as in 
'brains cause consciousness' - Searle), not what IS (as in "is the same as") 

So that critical premise actually contains an equivocation, i.e., "does not 
constitute and is not sufficient for" is used in two ways.

As a denial of identity it is obviously true, trivially true as Searle puts it. 
No one would deny it. But for the conclusion to work it must be taken as true 
in a causal sense, i.e., that syntax cannot cause semantics.

Searle relies on our realization of the non-identity of syntax and semantics to 
support a claim of non-causality and that just doesn't fly because non-identity 
has no implications for non-causality.

The problem arises in how we conceive of consciousness (or, in this case, 

If understanding is a system property rather than a process property, then a 
system made up of syntactical processes may, indeed cause understanding, even 
if no individual process in the system understands, has understanding or 
whatever. Thus syntax may in fact cause semantics even while it is trivially 
true that no instance of syntax in isolation is an instance of semantics. The 
CRA relies on the power of this equivocation of its terms in the third premise 
for the seeming obviousness of its conclusion. but the third premise, read 
correctly, is not trivially true at all!

Searle's CR is underspecked because he gives us a system that lacks the 
requisite functionalities (operations accomplishing certain tasks) that would 
actually constitute a system with understanding and he does this because his 
conception of consciousness relies on the notion of its nonreducibility. If it 
isn't reducible to things unlike itself, then it cannot be a system property 
and is ontologically basic, something that, somehow, belongs to this or that 
constituent process. But this idea of its irreducibillity is , of course, a 
dualist claim.

Dennett sees this in his commentary, even if he doesn't articulate it precisely 
as I have just done. However, he says enough to make it clear that he is taking 
the same position.

> Dennett needs to disprove Searle's simple logic. Specifically, he needs to 
> SHOW that syntax by itself is somehow constitutive of or sufficient for 
> semantics. He cannot offer any such argument, so he casts aspersions.
> -gts

I have just shown how, i.e., the system reply as Dennett makes it (namely that 
the CR is underspecked).

Nor did Dennett "cast aspersions" in that text I quoted. In fact, his decision 
NOT to specifically name Searle a Cartesian dualist, which you and others here 
seem to take as warrant to say Dennett isn't really making that particular 
claim, is evidence he avoids personalizing the issue and focuses, instead, on 
the problem with the CR and its argument alone.


Need Something? Check here: http://ludwig.squarespace.com/wittrslinks/

Other related posts: