Re: Syntax Vs. Semantics

From: HARNAD Stevan (
Date: Fri Jun 07 1996 - 20:17:03 BST

> From: "Jenkins, Nick" <>
> Date: Mon, 27 May 1996 12:51:28 +0100 (BST)
> Symbols form meanings in the head of the person reading them, not in
> themselves, so how can we comprehend symbol manipulation when it seems
> to lack meaning?

This point is not clear at all: What are you asking? How we can know
what symbols mean? No problem: Ask someone to define or translate them
for you if you don't know already. And what does manipulation have to do
with it?

> Symbols are manipulated by rules based on arbitrary
> shape, and must therefore be interpreted in someone's mind.
> The rules
> based on this arbitrary shape manipulation are known as syntax.
> Semantics, however, are the meanings of the symbols themselves, and are
> interpreted depending on the person reading them.

Not necessarily; everyone might interpret them exactly the same way. The
point is only that the semantics are not IN the symbol system, and that
the manipulation of the symbols is not based on the semantics.

Semantics is the meaning of symbols, syntax is rules for manipulating
their (arbitrary) shapes. It is not clear why you are mixing in bits of
the symbol grounding problem with the definition of syntax and

> Whilst a word either does or does not exist,


> or necessarily mean a
> specific thing (semantics),


> syntax has to be learnt.

Not clear what point you are making; the only syntax learning we really
talked about was Universal Grammar, and that isn't learnt. But you seem
to be mixing together a lot of things.

> One way we do
> this, Skinner suggests, is by speaking and either being rewarded when
> we achieve correct syntax, and not being rewarded when we do not.

Yes, and when it comes to syntax, Skinner is wrong: We DON'T learn it.

> Eventually our minds are shaped grammatically by trial and error.


> Chomsky suggested that children Learn syntax by listening to others and
> then trying it out for themselves - again trial and error.

Chomsky said the opposite: that you can't learn Universal Grammar by
imitation and trian and error.

> He called
> the knowledge required for the production and comprehension of
> linguistic sentences `linguistic competence'.

Yes, but that's irrelevant here.

> So syntax and semantics
> are the two main levels of language comprehension. Both affect the
> meaning of speech, but in different ways.

Semantics IS meaning. I strongly suggest you read the material again and
sort it out.

This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:46 GMT