Thinking without language?

[kenneth Collins]

=nothing= 'requires words' except communication.

when there are 'verbal symbols', they do impact upon the nervous system's on-going
information-processing dynamics, but that doesn't 'require' that verbal-symbolic stuff be
in-there if 'thought' is to occur.

=nothing= 'requires' verbal symbolism except communication.

'language' is 'just' an evolutionary add-on, albeit, an extremely-useful one that
multiplies the information-processing power inherent in individual nervous systems by
enabling individual nervous systems to be 'chained' together.

in this thread, you 'AI' and 'language' folks are addressing such 'chaining',
=exclusively=. nervous systems are extremely-very-much more that that which you're
addressing.

there's immense worth in addressing the stuff of 'communication', but it must be, first,
recognized that that is what's being exclusively discussed (communicated with respect
to).

ken (K. P. Collins)

Lee Sau Dan ~{@nJX6X~} wrote:

> >>>>> "John" == John Turnbull <john at turnbull.org> writes:
>
>     John> The second one I would agree with, if you are not conscious,
>     John> you are not thinking.  To me thinking is the answering of
> .......................................^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>     John> questions, which requires explicit communication between
> ..........^^^^^^^^^
>     John> different parts of the brain, and some language is needed
>     John> for that.  For me words are the language.  Pictures and
>     John> diagrams are occasionally useful to help clarify things or
>     John> gain a different insight, but the question and answer are
>     John> still in words.
>
> So, thinking MUST be verbal, you YOUR definition.  Period.
>
>     John> Have you ever hummed a tune without realizing it?
>
> I can hum a tune in my mind.
>
>     John> Just
>     John> knowing what you are doing and feeling maybe consciousness,
>     John> but I don't think that it is thinking.  Asking the question
>     John> "what am I doing?" tends to get a verbal answer.
>
> I can "ask" that "question"  without thinking in words.  I often think
> in  a language-less manner.   I can  easily form  the question  in any
> language I'm fluent at.  So,  when I think, I use language-independent
> concepts.  Only  when I have  to speak them  out do I  transform these
> ideas into words, and order them  by the grammar rules of the language
> desired.  So, in  a Cantonese context, I would  say [NO13 tsou22 k at n35
> mE53  a33]?   In  a Mandarin  context,  I  would  say <wo3  zai4  gan4
> shen4me>?  In an English context, I would say "what am I doing"?  In a
> French context, I would say "qu'est-ce que je faire maintenant?", etc.
> When I don't need to speak  it out, the idea can stay language-less in
> my mind.
>
>     John> what am I doing?  why am I doing it?  is there a better way?
>     John> all without words?  Maybe so, but it seems to me a lot
>     John> better to use words.
>
> Yes,  except  for the  "why"  question.   Answering  a "why"  question
> requires reasoning,  and reasoning  often needs verbal  thinking.  But
> answering "what" and  existence questions, I can do  it without words.
> Even  when   answering  "why"  questions,   I  can  sometimes   do  it
> language-lessly.  So, that's again  without words.  (When I don't need
> to speak  out the  answers, why  do I have  to render  the ideas  in a
> language?)
>
>     John> OK, so you draw pictures with no words.
>
> I often  avoid words in the  pictures.  When I need  to represent some
> ideas (esp. abstract ones) with  words, I usually use the first letter
> (in case  of English)  to represent the  whole idea.  So,  that serves
> only as  a symbol (somehow arbitrary)  or mark to remind  me that that
> particular letter stands for that particular idea.
>
>     John> My diagrams tend to
>     John> be more of arrangements of words on the page.  Words are so
>     John> much more efficient than wasting all that time with images.
>
> When  I do  geometry or  solve geometric  problems, I  would  draw the
> figures  wordlessly.   Yes, I  would  *label*  some features  (points,
> edges,  angles) with letters,  but they  are only  labels.  If  I have
> enough pens of different colours, I  could do the labeling by means of
> colours  instead of letters.   That doesn't  prevent me  from thinking
> about the geometric problems.
>
> --
> Lee Sau Dan                     $(0,X)wAV(B(Big5)                    ~{@nJX6X~}(HZ)
> .----------------------------------------------------------------------------.
> | http://www.cs.hku.hk/~sdlee                      e-mail: sdlee at csis.hku.hk |
> `----------------------------------------------------------------------------'