[FOM] Shapiro on natural and formal languages

[hendrik@pooq.com]

On Wed, Nov 24, 2004 at 04:57:26PM +0000, Roger B Jones wrote:

> However it is not uncommon for philosophers (in my limited
> experience) to write as if formal notations could have
> no other purpose, and as if they can only be evaluated
> in terms of how well they relate to natural languages.

It is not uncommon for writers, who spend a good deal of time using
language,  to think that language is essential to thought.  And the
philosophers that publish, at least, are writers.  Pinker spends a
good deal of space in one of his books debunking this idea (although
I forget whether this was in "How the Mind Works" or "The Language
Instinct").  Formal notations do appear to encode some aspects of
thinking, though it is definitely not any kind of direct translation.
If someone were to believe that thinking is impossible without language,
he might very well come to conclude that formal notations model
natural language.

I will observe, though, that I do not think that natural languages
model or mimic thinking either, although language they do encode some
of our thoughts for cummunication.  I refuse to acept that human
thinking involves declensions, conjugations, irregular verbs, and
irregular nouns.  I am willing to believe that our thinking uses
mechanisms for handling rules and exceptions unconsciously, and that
we use these unconscious mechanisms to process language.  We probably
use some of these mechanisms to process formalisms,  but most of the
time we avoid processing formalism altogether -- they are remote
from many of the ideas we want to express, even when we are doing
mathematics.  That is why mathematics textbooks contain a lot of
natural language, after all.

-- hendrik