Re: Not enough info for your brain (was Re: Dawkins video)

Greg Billock (billgr@cco.caltech.edu)
Wed, 17 Jun 1998 00:20:55 -0700 (PDT)

Glenn Morton:

[refs]

> Can I forward the above, not the below, to Gillian Brown? She is the one
> who originally contacted Wieland and got Carl's comments which included the
> claim of publications. I want to know if Carl can back up his claim.

Sure.

[DNA and brains]

> I don't think we are disagreeing here in spite of what you may think. I am
> trying to get someone on the design side of the creationist argument to
> tell me where the information comes from. The design group always says
> that design requires information, yet there is demonstrably not enough info
> to pre-determine our brains let alone at the same time to predetermine our
> organs.
> If the anti-evolutionists are so confident that information is a
> pre-requisite for design then they should be able to tell where exactly
> the specifications reside.

OK, I see what you mean. For what its worth, it seems that a lot of
the 'information' lies in the external world, and the reliance on things
like real-world light patterns, sound patterns, and so forth, laws of
inertia, air friction, and whatnot being what they are in order to
tune the brain in to its 'design.'

Of course, we're not all that sure how many ways there are to make a
world, either, so information theory isn't straightforward to apply
there, either.

> As to evolutionary mechanisms, Terrence Deacon in the Symbolic Species
> talks about how during development, different parts of the brain compete
> for connections. The larger the part of the brain, the more connections it
> generally makes. With man, the prefrontal portions of the brain have
> enlarged dramatically over that in the chimp and thus their competion for
> connections creates many connections to our frontal lobes, making us human.
> The brain is underdetermined informationally speaking but we are all alike
> because the competition for connections plays by the same rules in all of
> us, generally.

Yes, I've seen the stuff on that, and it is fairly convincing (the
neuronal competition). I'm not sure I'm willing to commit on humanization
factors, though. :-)

[more to life than DNA]

> I agree, but it flies in the face of the argument presented by Intelligent
> design folks as proof that information is necessary for evolution to proceed.

In the simplest case, for sure. Taking into account environmental
'information' seems more of a hard job than DNA information, even, so
it's hard to see what would come of such an attempt. Perhaps it would
be fruitful, though.

[...]
> much like a message in a known language, with chemicals acting as letters
> and combining in defined sequences to form words, phrases, and sentences.
> The 'message' is decoded by the cell much the same way the dots and dashes
> of messages in Morese Code cann be decoded by anyone who knows
> it."~Percival Davis and Dean H. Kenyon, Of Pandas and People, (Dallas:
> Haughton Publishing Co., 1993), p. 6

As we've been saying, this is a hyper-simplistic view. (Even
in 1993.) Cells are tuned for the DNA they want to 'decode' and the
DNA builds up some of the decoding mechanism itself (meaning there is
a kind of blurriness about code and decoder).

> I would ask the question, "If the origin of the brain doesn't require all
> this information, why does the origin of life require the specified
> information they claim?

One fairly common explanation is that most of the information gap is
made up by noticing that its us that are alive. That is, the information
required to specify that there are living systems is mostly (according
to some ideas) spent in specifying which ones result, being as there
are so many ways to do it. My view is that information isn't going to
be of much use in solving the puzzle, but I could be wrong.

> >Just as a final pedantic note in broadcast: information theory is
> >concerned with ensembles of events/messages/arrangements, and not with
> >less-well-defined ideas of complexity. If there is only one possibility
> >for a message, it doesn't matter whether the message is 'BOO' or the
> >entire Encyclopedia Brittanica; the information content is the same:
> >zero. That doesn't make information theory outmoded and inadequate,
> >it just means that if you are going to use it, you have to make an
> >effort to specify the context of the transmission (source and sink),
> >the statistics of the messages, and so forth.
>
> That was Wieland who was saying info theory was outmoded.

Yes. I didn't want to imply it was your view, so I addressed the
comment in 'broadcast' mode. :-)

-Greg