Tuukka's Comment taken from Good Math, Bad Math

49 views
Skip to first unread message

cdipoce

unread,
Jan 9, 2012, 2:58:50 PM1/9/12
to CTMU Discussion, ma...@tuukkavirtaperko.net
Taken from http://scientopia.org/blogs/goodmath/2011/12/01/the-annoying-ctmu-thread/
:

"I think Langan has real merit in trying to create, basically, an
ontology that is a context-free language, as opposed to a finite-state
machine. He is quite methodical about that. I think that's what
everyone should have been doing ever since someone asked, whether
P=NP. Every metaphysicist, that is. Wheeler-style reality theories are
the key and Langan is the first one I perceive to have actually
created something like that. " - Tuukka

I'd really appreciate it if you would further unpack your remarks.

Message has been deleted

Tuukka Virtaperko

unread,
Jan 11, 2012, 6:06:13 PM1/11/12
to CTMU Discussion
cdipoce,
A context-free language is recursively defined by itself, as opposed
to a finite-state machine, which operates according to rules it cannot
change. Now, since Langan is trying to construct a theory of
everything, he would fail it his theory were based on rules that are
somehow "outside the theory itself", because then his theory wouldn't
be a theory of everything.

Langan tries to have the CTMU recursively define itself. This is quite
different from having a circular argument. Circular arguments do not
"define" anything. But on the other hand, recursively definitions are
completely reasonable. For example, the Fibonacci sequence is
recursively defined.

He is not the first to do this. In the Good Math Bad Math ("GMBM") he
was ridiculed for doing so, but his attempts are much less ridiculous
than those of Rescher in constructing a recursive definition in such a
way that Rationality and Virtuousness are equivalent. Admittedly, I'm
not sure whether that exactly qualifies as a recursive definition. It
seems too simple and arbitrary. Anyway, this Rescher, who wrote the
book "Rationality" (1989), was the youngest person to obtain a PhD of
philosophy from I recall Chicago University, and he's written about a
hundred books and maybe four hundred articles. So if this guy can make
definitions which are seemingly recursive or at least circular (he
insists that it's not a "vicious" circle :D maybe he's never heard of
the word "recursion" because he's not a computer scientist), why can't
Langan?

Langan's theory is relativistic in the sense Buddhists speak of
relativism (NOT in the Western sense). For more info, see "Two truths
doctrine" from Wikipedia. The problem is that Langan tries to reach
for what Buddhists call "ultimate truth" without taking into account
certain relevant limitations, which imo are better covered by Robert
Pirsig in his Metaphysics of Quality. But Pirsig does not do much to
stress the recursive nature of the Metaphysics of Quality, while for
Langan, the entire theory is about that.

Langan expects very much from the abilities of language to convey
meaning. He's missing something young Wittgenstein also missed. But
he's right in that reality must be recursively defined, lest there
would have to be a meta-reality, of which reality is a subset. To do
so would beg the question of what is reality like. I am trying to
publish work on these limitations asap.

-Tuukka

P.S. I tried to send this two days ago, but apparently it didn't come
throught.
P.P.S. Deleted and re-sent to get line length right.

On 9 tammi, 21:58, cdipoce <cdip...@gmail.com> wrote:
> Taken fromhttp://scientopia.org/blogs/goodmath/2011/12/01/the-annoying-ctmu-thr...
Reply all
Reply to author
Forward
0 new messages