Newsgroups: comp.ai.philosophy
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!udel!ssnet.com!stingray.ssnet.com!pagesat.net!internet.spss.com!markrose
From: markrose@spss.com (Mark Rosenfelder)
Subject: Re: What's innate? (Was Re: Artificial Neural Networks and Cognition
Message-ID: <D3yqy8.L6G@spss.com>
Sender: news@spss.com
Organization: SPSS Inc
References: <3gtu3i$rf3@mp.cs.niu.edu> <3hf0fa$7p@agate.berkeley.edu> <D3t179.633@spss.com> <3hgscc$su6@agate.berkeley.edu>
Date: Tue, 14 Feb 1995 00:06:08 GMT
Lines: 34

In article <3hgscc$su6@agate.berkeley.edu>,
 <jerrybro@uclink2.berkeley.edu> wrote:
>markrose@spss.com (Mark Rosenfelder) wrote:
>>  Just asserting
>> 
>> >The first one makes no sense, which is a fact about its semantics.
>> 
>> is not an explanation; it's a restatement of the problem: people don't
>> say that, and have trouble understanding it.  But *why* do people
>> have trouble making sense of it?
>
>But isn't the syntactic description of the role of "and" itself
>just a way of describing the same problem?  Does it get us anywhere?

I dunno.  We'd probably have to see what other facts are explained
by a syntactic vs. a semantic account.

>I'll stab at it: "and", as far as I can see, always has to actually
>join two like terms, fore and aft, and strongly resists any other
>application.  Moreover, it seems able to join just about anything,
>nouns, verbs, phrases, sentences, and whole paragraphs.  Possibly
>this wide applicability (in this one dimension) stands
>in the way of further applications.  I think we can expect that
>there's a limit to how widely we can use something.  For example,
>a word can only gain so many meanings before we will want to
>create a new word.

This seems similar to ideas also expressed by Jim Balter and Marvin Minsky;
as I said elsewhere, I think there's something to it.  I'd be inclined
to call it a syntactic rather than a semantic constraint: the problem
seems to be the number of *types of syntactic construction* (see your own
list above) that conjunctions can conjoin.  That is, they lead to a lot of
*syntactic* complexity, and restraints on what can be done with conjunctions
may reduce the amount of syntactic processing (parsing) we have to do.
