[Wittrs] Re: Who beat Kasparov?

  • From: "SWM" <SWMirsky@xxxxxxx>
  • To: wittrsamr@xxxxxxxxxxxxx
  • Date: Fri, 19 Mar 2010 22:19:26 -0000

--- In Wittrs@xxxxxxxxxxxxxxx, Gordon Swobe <wittrsamr@...> wrote:
> --- On Fri, 3/19/10, SWM <wittrsamr@...> wrote:
> >> There exists no understanding in the CR, Stuart,
> >> because *syntax by itself does not give semantics*. This is
> >> a fact in organic brains, including yours, just as it is a
> >> fact in computers.
> >

> > Then your conclusion is already built into the argument and
> > it's circular in which case it's not logically valid. You do
> > have the intuition you are expressing here, no doubt about
> > it. But if Searle were just making an argument on intuition
> > he would have no need to lay out logical steps as he does.

> Nothing circular going on here. The CR thought experiment merely 
> *illustrates* a simple axiom which should seem obvious to anyone who thinks 
> about it. That axiom, combined with the first two, lead logically to the 
> conclusion that programs don't cause minds.

But it isn't obvious to everyone though it is certainly an intuition we have. 
Even I have it at times, when I'm not paying close attention to the 

And if it's not obvious to any but those to whom it is already obvious then it 
really demonstrates nothing! It just expresses a point of view.

But recall that the point of the CRA is to generalize from the CR. The only 
basis for such a generalization lies in having that intuition that you claim 
"should seem obvious to anyone who thinks about it." Yet it seems less obvious 
to me the more I think about it.

As I have said from the outset, this comes down, in the end to competing 
conceptions of consciousness. If one is stuck in one view, one will just not 
see the other. But the CRA cannot make its case based on the fact that the case 
is already made before the argument is presented because then there would be no 
point for Searle to have made such an argument in the first place as he did.

> If I wanted to write a book titled _Syntax By Itself Is Neither Constitutive 
> Of Nor Sufficient for Semantics for Dummies_ then I would want to include in 
> that book a story with pictures about a man in a Chinese room who cannot, no 
> matter how hard he tries, understand Chinese symbols given that he has only 
> information about how to manipulate the symbols according to their forms.
> It's not a complicated concept, really!
> -gts

No it's rather simple, in fact. It's the first thing that hits us when we start 
to think about this. After all, how could a dumb program merely going through 
certain rote steps understand anything? And then the question is, if such a 
program can't, then how could lots of programs (more of the same) do so? Zero 
plus zero is still zero, right?

But the logical implication of that is that nothing "dumb", nothing that is 
merely some physical behavior, could ever constitute understanding. But this 
comes down to a failure to see that understanding can be a system rather than a 
process feature and that systems consisting of many "dumb" processes can, in 
fact, do things that the process cannot. The way to overcome this failure is to 
think about what in our experience understanding really consists of.

Think of that cartoon in that article you routed us to, the man observing the 
Chinese character on the wall and thinking about a horse (in the form of 
visualizing a horse image). Such thinking can be achieved by a complex system 
of otherwise mindless processes and, if so, then the system has the 
understanding which its constituent parts lack.


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

Other related posts: