Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site cybvax0.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!gamma!epsilon!zeta!sabre!bellcore!decvax!genrad!mit-eddie!cybvax0!mrh
From: mrh@cybvax0.UUCP (Mike Huybensz)
Newsgroups: net.origins
Subject: Re: Non-prediction, Falsification
Message-ID: <547@cybvax0.UUCP>
Date: Fri, 24-May-85 15:15:40 EDT
Article-I.D.: cybvax0.547
Posted: Fri May 24 15:15:40 1985
Date-Received: Sun, 26-May-85 00:54:55 EDT
References: <1129@uwmacc.UUCP>
Reply-To: mrh@cybvax0.UUCP (Mike Huybensz)
Distribution: net
Organization: Cybermation, Inc., Cambridge, MA
Lines: 76
Summary: 

In article <1129@uwmacc.UUCP> dubois@uwmacc.UUCP (A Ray Miller) writes:
> > [Michael Ward]
> > [rhetorical]  How can you make predictions about how random changes will
> > occur? 
> 
> You can't.  That's exactly my point.  People should stop talking about
> what they "expect" - or at least specify that they expect *something*
> but that they don't know *what*.  (Of course, such an admission might be
> somewhat damaging.)
> 
> In my mind, there is a rather striking similarity between
> 
> 	(a) God made it that way because he wanted to.
> and
> 	(b) this particular event happened that way because it happened
> 	    to.
> 
> Big deal.  Either one could be true, in point of fact.  Neither falls
> within the realm of testability.

While it may be impossible to make predictions about individual random
changes, it may be quite reasonable to make predictions about aggregates
of random changes.  That's what thermodynamics, quantum mechanics, and
evolution are about.  Just as you can use a bell curve to predict
particle veloceties in a gas, you can use evolution to predict a hierarchy
of similarities among organisms.  You could just say "God made that
distribution": but that's not testable.  The specific curve is testable
against the particle veloceties.  Evolution is testable against the
similarities among organisms.

> Thus making it *even more difficult* to falsify.  If they don't change,
> that's ok.  If they change rapidly, that's ok.  If they change
> gradually, well, we never ruled out gradualism, you know...  But the
> more the time for speciation gets whittled down, the more reasonable it
> becomes to ask for direct observation.  The more the time span of
> change of Bauplan is shortened, the greater the susceptibility to the
> requirement for demonstration of direct evidence becomes.

So, let's see: gawd created the animals in a day, but I didn't see him
create any yesterday, so creationism should be rejected.  :-)

As you point out above, observation becomes simpler if punctation is the
tempo of evolution.  Thus evolution becomes easier to falsify by lack
of direct evidence.

Even so, it would take a long time for (say) speciation of a new mammal
to occur.

> > Criticizing the theory of evolution for not predicting what it says
> > cannot be predicted makes as much sense as throwing out the principle of
> > uncertainty because it doesn't predict the location and velocity of a
> > photon.
> 
> You might toss it if it didn't predict either one specifically.

Uncertainty doesn't predict either.  You want to toss it?

> > It offers a general mechanism by which one
> > species can change into another and has pointed the way to fruitful
> > research.  It is true that at the present time we do not know the
> > mechanism by which species change or remain the same, but I expect that
> > this is a situation that will change in our lifetime.
> 
> How can it offer a general mechanism at the same time that we don't
> know the mechanism?
> 
> There is a contradiction in this paragraph.

I know that a bus is a mechanism for getting from here to there.  I don't
know what kind of engine an individual bus has: it may be electric, gas,
or something I haven't learned of yet.  I believe in electric and gas
motors because I've seen them demonstrated in the lab.  Thus I'm willing
to believe that buses can exist and move.
-- 

Mike Huybensz		...decvax!genrad!mit-eddie!cybvax0!mrh