Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site denelcor.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxl!ihnp4!zehntel!hplabs!hao!denelcor!lmc
From: lmc@denelcor.UUCP (Lyle McElhaney)
Newsgroups: net.origins
Subject: Perhaps probability
Message-ID: <539@denelcor.UUCP>
Date: Tue, 21-Aug-84 23:00:30 EDT
Article-I.D.: denelcor.539
Posted: Tue Aug 21 23:00:30 1984
Date-Received: Thu, 23-Aug-84 04:49:06 EDT
Organization: Denelcor, Aurora, CO
Lines: 38

> Creationists brandishing probability arguments against evolution have
> been told that probability cannot meaningfully be applied to the question
> "how did what's here happen to become as it is".  If this is true, then
> it would appear a reasonable reply to request that evolutionists refrain
> from such arguments as well. ...But if probabilistic considerations are
> inadmissible as evidence, then such arguments are as fatuous as those of
> the creationists are alleged to be.  Right?

Now, wait a minute. I've not seen anyone say that probability arguments are
inadmissable. The point is that probability, like any tool, is subject to
gigo - the results obtained are only as valid as the input supplied. The
argument against the the "probability arguments against evolution" applies
to the assumptions made, not to the statistics derived therefrom. The
assumptions of the argument are:

 - that all possible combinations of any two (atoms, amino-acids) are
   equally probable;
 - that there is only one possible right combination which will fulfill
   any given protein's function;

The "fatuousness" of the argument is in the assumptions, not the statistics.
The assumptions are demolished; the argument is invalid. Any similar argument
not based on these assumptions is not affected.

> "Undoubtedly.  Probably.  Interia kept it around.  Improved its
> efficiency.  The one that was handy."

Yes, I guess I could be accused of looseness in my language, but that's the
way I like to write (and read) net news. If you want turgid thesis-style
wording and bullet-proof logic, I'll have to bow out of this one -- I have
neither the time nor the inclination to make my wording unassailable.
Net.origins is hard enough to read and follow without driving away the
audience with the kind of iron-clad verbage available in the referenced
books. I prefer to keep the wording light, and strike the points of the
arguments, rather than belabor them.
-- 
		Lyle McElhaney
		(hao,brl-bmd,nbires,csu-cs,scgvaxd)!denelcor!lmc