[Prev][Next][Index][Thread]

Re: C# is not Dylan (was: Re: C# : The new language from M$)

Not to get into a huge discussion but, having also ready both Pinker and
Chomsky (even before he moved into using linguistic semantics as a method
for analyzing politics), I would be surprised if either of them would
automatically give a pass to applying their work to artificial, (and
especially) primarily non-verbal languages.  If that were so, we would not
be designing languages that had routine calls deeper than about three or
four levels (the deepest recursion normally occurs in languages); we'd be
limiting structures to no more than about six or seven elements (else, we'd
not be able to remember them all), etc.  In short, the argument about
artificial languages even mapping onto cognitive structures the same way as
natural languages do seems quite specious to me.  You, of course, are
entitled to your own viewpoint.

faa

Rob Myers <robmyers@cwcom.net> wrote in message
B584F881.3CDF%robmyers@cwcom.net">news:B584F881.3CDF%robmyers@cwcom.net...
> > From: "Frank A. Adrian" <fadrian@uswest.net>
> > Date: Sat, 1 Jul 2000 22:30:03 -0400 (EDT)
> > To: info-dylan@ai.mit.edu
> > Subject: Re: C# is not Dylan (was: Re: C# : The new language from M$)
> >
> > Yes, but it makes about as much sense as citing Pinker's and Chomsky's
works
> > as a defense of saying that syntactic choices don't matter when applied
to
> > artificial languages.
>
> Having read Pinker (if anyone hasn't then do, The Language Instinct is
> inspirational), I'd say that his accessible explanation of Chomsky's
syntax
> trees, deep and surface structure, and the process of creolization are
very
> relevent both to the creation of artificial syntax and its socialized use.
A
> language's "artificiality" has nothing to do with how applicable this is,
if
> anything, artfificial languages are easier examples (apart from Fortran
:-).
>
> BNF and Chomsky trees come from the same period and are mirror images of
> each other to a degree. The contrast between deep & surface structure is
> important in deflating a debate (however nonsensical) about infix vs.
prefix
> syntax for languages with fundamentally similar underlying systems.
>
> When considering a language whose readability is entirely up to the user,
> the structures that programmers use to achieve this must be regarded as
> external to, and a socialized appropriation of, the syntax of the original
> language and therefore a form of creolization. Machine-generated Lisp
looks
> very different from the elegance of good human-generated Lisp.
>
> I don't see how immature attempts at demonization are more relevant than
> this. In my 3 years on this newsgroup this is the closest I've seen to a
> flamewar, and comp.lang.dylan isn't moderated.
>
> - Rob.
>
References: