[ExI] Meaningless Symbols

Spencer Campbell lacertilian at gmail.com
Sat Jan 16 22:46:59 UTC 2010


2010-01-16 Aware <aware at awareresearch.com>:
> This discussion shares much in common with PHIL101-type bantering
> common in college dorms--less the wine, beer and marijuana.
>
> If you want to gain some traction, might I suggest the following?
> I've given it to others with useful effect and I think I've posted it
> here before.  If the purpose of this discussion is to increase
> understanding, rather than just to be right (John, are you listening?)
> then you should at least be familiar with the thinking presented by
> John Pollack in this paper.

How do you know? Everyone on the Internet is perpetually trapped in a
quantum superposition of drunk, high and sober.

Now that I think about it, that explains a lot.

Twenty-nine pages later, I feel confident in saying something with a
little more substance. But only a little.

It took me a while to relate it to the current subject, but then I
came up with this: "Does the symbol 'I' have meaning?". Looking at it
that way, I'm not so sure the concept of meaning is meaningful. Two
definitions are competing for sovereignty in my head:

"A symbol is meaningful if it has a referent."
or
"A symbol is meaningful if it is part of a consistent function."

Interpreting the first definition is easy. The second definition
hearkens back to an argument made here yesterday.

2010-01-15 Eric Messick <eric at m056832107.syzygy.com>:
> The fact that [a computer] comes up with the correct answers for the following:
>
>  0 + 0 = 0
>  1 + 1 = 2
>
> indicates that it understands the fundamental difference between the numbers zero and one.

In this case, '0', '1', and '+' are the symbols. Pay no mind to '='
and '2', for they are there only for our convenience.

It seems to be inherent in the meaning of these symbols that they can
be put together in one-dimensional sequence, and that any sequence of
the symbols is itself a symbol. Some of the resulting compounds are
meaningful, and some aren't. For example:

'0 + 1' is meaningful.

'0 1 +' is not. Unless we're using reverse-polish notation, which we
should, but aren't.

It's also noteworthy that there are more meaningful symbols than there
are meanings. '0 + 1' is a different symbol from '1' by itself, but,
obviously, they share precisely the same meaning.

The only question is, are the meanings in the computer, or in our heads?

I'm playing catch-up, here, so please excuse the redundancy. It seems
to me that the question is, itself, meaningless. If I take John
Pollack's paper as an axiomatic starting point, it seems readily
apparent that 'I' has meaning by the same definition as '1', that is,
it plays a significant role in forming meaningful compound symbols. 'I
am human' means something very different from 'You are human', where
'human', 'am' and 'are' are symbols too.

'am' has the same meaning as 'are', though, again, it is obviously a
different symbol. The grammatically astute will have a few more fly
off the top of their heads right now.

So, back to the present. 'I' has meaning to us. Therefore, if a
machine can be built which uses 'I' in exactly the same way, both in
conversation and within its own c-thoughts, 'I' will have meaning to
that machine.

Anyone have a soldering iron handy?



More information about the extropy-chat mailing list