[Wittrs] Re: Meaning, Intent and Reference (Parsing Fodor?)

  • From: "SWM" <SWMirsky@xxxxxxx>
  • To: wittrsamr@xxxxxxxxxxxxx
  • Date: Thu, 11 Feb 2010 18:52:27 -0000

--- In Wittrs@xxxxxxxxxxxxxxx, "jrstern" <jrstern@...> wrote:

<snip>


> Fodor requires a bit more philosophical background,
> Dennett writes at least half for the layman at all
> times.  Dennett is an outstanding writer, Fodor is
> merely very good.  Fodor, I think tries, for better
> or worse, to grapple with details, Dennett is more
> about frameworks.
>

Good philosophy, that is clear thinking about difficult concepts, ought to be 
doable in ordinary language. While there's no denying that sometimes something 
is gained by switching to formal logic or to the technical language of 
specialists, more often than not, in my experience anyway, such a switch 
(better: such a dependence!) mostly manifests unclarity, i.e., an inability to 
say directly what one has in mind. While I am not prepared to say that of Fodor 
yet, because I don't know enough of or about his work, I do think that may be 
one legitimate explanation for his writing being so inaccessible. You yourself 
say you've been reading him for 30 years and still don't feel you fully follow 
him. THAT says something!

It's not about laymen vs. specialists but about understanding by being clear 
vs. confusion and obfuscation. (Of course, even here there are degrees of 
occurrence. Searle, for instance, is often very clear and explicit, but then he 
almost inexplicably lapses into what looks like confusion -- one wants to say 
that's not possible, he's too well known, too credible a professional 
philosopher for that, but then that would be to confuse the merits of his case 
with the authority he presumably carries while making it.)

>
> > But Fodor just gets my head spinning.
> > I wonder if he is more like Edelman here than Hawkins?
>
> I don't think either of those are in the game.
>
>

Neither are philosophers or explicitly engaging in philosophy although both 
stick toes in the water. Edelman is a biologist and Hawkins is a computer 
engineer. But because both are concerned with the issues of thinking and 
thought, minds and understanding, they naturally cover similar ground as people 
like Fodor, Dennett, Searle, et al. In the end, though, it can't be about the 
credentials of any given writer but about what they have to say. Both Edelman 
and Hawkins present very detailed cases for their ideas about how the brain 
works (part of the brain only, in Hawkins' case), though Edelman, it seems to 
me, fumbles badly and gets lost in the complexities he identifies (though I 
think he brings some interesting insights to the table). As to the 
philosophical game, at least at this stage, I think Dennett is the clear winner 
of this particular race. But my understanding of Fodor is still too shaky to 
really judge him.


> > > Certainly he's a physicalist in that physical brain state
> > > corresponds with meaning, but just what that means (!) is not
> > > necessarily clear, in regards to balancing "methodological solipsism" and 
> > > correspondence with distal objects.
> >
> > Yes, not at all clear. The mental language is
> > presumably some corresponding set of processes
>
> Wow there cowboy, slow it down.
>
> Is a language a process?  Is English a process?
>

I was asking what kind of language could happen in brains and, if one wanted to 
say that there were such languages, would we have the same thing in mind by 
"language" as when we speak about English or Chinese or Esperanto? As I noted 
in that same post, Ramachandran likens the communication of information within 
brains between neurons and neuronal clusters to a language, too. But his use 
clearly enlarges on the usual idea of language though perhaps not 
inappropriately. I was asking whether Fodor's idea of a language of thought in 
brains is equivalent to the language of brain cells described by Ramachandran?

>
> > that underlie each and every distinct thought
> > we have
>
> Does the print on a page "underlie" the
> content of the book?
>

If I say something in English and Searle's Chinese Room translates it into 
Chinese, then what I said, what I had in mind, is found in my ideas as I 
expressed them, namely in the English words and sentences I used. Thus my 
statement(s) in English underlie the translation offered for them in Chinese 
and, indeed, if the Chinese version diverged too radically, it would be a bad 
translation, the original meaning not being captured, conveyed, etc.

On the Fodorian view, as you have described it at least, there is a mental 
language in which our thoughts happen and then a translation process (and 
processor?) that turns them into English. I am asking if that is the picture 
Fodor wants us to have?

>
> > and which somehow get translated into the
> > language(s) we actually speak to one another
> > and think in.
>
> Speak, yes.
>
> But the idea of an LOT is that it *is* the language
> we think in.  *I*, but NOT Fodor, think that the LOT
> is not only the language we think in, it IS the
> thought itself.  Yes, "your thoughts are written in
> the brain like writing on the page", that is the
> sentence that everyone tries to disclaim before
> launching into endless rantings about "thoughts"
> and "concepts". Except me.  I embrace the demon.
>

I recall a fellow from the Popperian CR list insisting that thoughts are 
wordless and only after being thought do they get translated into words. I find 
that a little strange. When I think about my own thoughts I do it in language 
(talking to myself) though I agree that sometimes I become aware of inexpressed 
thoughts, suggesting that thinking may not be exclusively linguistic. But the 
only way a thought becomes accessible to me, the only way it registers you 
might say in my awareness is when I verbalize it (even if only in my head). 
Then it seems to become part of my overall understanding, etc.

So is Fodor's language of thought the underlying occurrences before we 
verbalize or fully verbalize them?

>
> > What kind of "language" must such
> > a mental language be?
>
> It is quite clear that the advent of the digital
> electronic computer was the motivation for modern
> theories like Chomsky's and Fodor's.  Any language
> should be physically realizable by a TM, and any
> language that can be realized by a TM can be
> emulated by a UTM, of which you are reading this
> on one now.


To be "physically realizable" opens the usual questions. Yes, we can form the 
words mechanically with a mindless machine (Searle's CR?) but where is the 
understanding that makes the sounds (or symbols) words? That doesn't seem to be 
physically realizable by forming the words in the right syntactical order in 
the right context for the meaning at all. Still, people like me will argue that 
we can achieve even that with physical processes and I think you would agree. 
But clearly just manipulating zeroes and ones in a computer via an algorithm 
isn't understanding and doesn't supply understanding alone. Something else 
seems to be needed.

With Dennett I would argue that what's needed is a sufficiently complex 
process-based system operating in a certain way (the way this is physically 
realized). But what you have described doesn't seem to be that since you are 
suggesting we equate language in its full sense with whatever a computer does. 
I would say that is a more limited sense of language, akin perhaps to 
Ramachandran's suggestion that the fundamentally mindless informational 
exchanges between brain cells is also "language". Yes we can say that but only 
in a special sense while the question before us is what would it take produce 
language in the usual sense, i.e., a system of sounds or symbols that carry 
meaning between conscious beings?


> So, it doesn't really matter exactly
> what it is like, it will be highly intertranslatable.
> It will be interesting to find out, after all,
> exactly what it is like, but only for practical
> reasons, there is nothing theoretical riding on it
> at all.
>

I am inclined to think that translation tends to be imperfect and that this 
militates against information being "highly intertranslatable" between 
languages (the more so when the languages are significantly different in form). 
Of course we do have translatability and sometimes information is "highly 
intertranslatable" but I am inclined to think that is a function of conditions, 
contexts and so forth.

>
> > How do we discover it, recognize it when we see it,
> > distinguish it from other brain processes, etc.?
>
> Compsci 101, intro to programming
> Compsci 201, turing machines and automata
> compsci 202, neural networks
> Compsci 301, compilers
> Compsci 401, operating systems
> Compsci 501, computational linguistics
>
> Knowing it when you see it, is indeed the question.
>

Short of a full blown field of study aimed at identifying Fodor's supposed 
language of thought, how does he say we would recognize it or describe it?


<snip>

>
>
> > I'm skeptical of this approach.
>
> There is no other.
>
> Fodor has always said that,
> and I have always agreed.
>
> Josh
> =========================================

I meant "skeptical" in more than the usual, standard way of always keeping an 
open mind and refusing to be convinced so long as there is anything still 
pending to be resolved, etc. I mean I am inclined to think, at least at this 
stage, that Fodor is wrong, unless his "language of thought" turns out to be a 
fairly innocuous usage, i.e., something like Ramachandran's point. But I can't 
believe that would be his position as he seems to want to hang so much more on 
this than Ramachandran does.

SWM

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

Other related posts:

  1. Home
  2. wittrsamr
  3. Archive
  4. 02-2010