[FRIAM] random v stochastic v indeterminate

[Nick Thompson]

Steve, 

 

Thanks for staying with me on this.  

 

To be honest, I have never encountered anybody who believed that natural
selection alone is capable of producing evolution, unless it was somebody
who includes some variation-generating mechanism within the notion of
natural selection.  I have encountered people who think that natural
selection is not NECESSARY to evolution, attributing most change to random
walks of various sorts.   I have never understood those folks, but they have
had their day.  

 

The heresy I am trying to expunge is that in which evolution is understood
as "a delta-q in the Hardy-Weinberg equilibrium", which amounts to saying,
natural selects whatever nature selects and whatever nature selects is
evolution.  Darwin would have been baffled by such a formulation. 

 

Nick 

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:friam-bounces at redfish.com] On Behalf Of Steven A Smith
Sent: Friday, August 11, 2017 1:56 PM
To: The Friday Morning Applied Complexity Coffee Group <friam at redfish.com>
Subject: Re: [FRIAM] random v stochastic v indeterminate

 

Nick -



I am very glad to note that you are recovering and your scrappiness is
properly returning!

[NST==>The best cardio rehab is for you-guys to keep annoying me.  Thanks
for that. <==nst] 

You might check with your cardiologist on this one, I'm not sure a rise in
BP is the same as exercise-stimulated increased heart rate, but in any case,
I'm glad we can be of service!




What's powerful about it?  

Nothing more than it is such a succinct statement negating the popular
fallacious apprehension of the mechanism of evolution, suggesting that there
is a causal link between "selection" and "innovation"...   the innovation
step is in the mutation, but as the quote states clearly, said *innovation*
is *preserved* (selected for) by the natural selection mechanism.  

[NST==>Wait a minute!  What is the misapprehension of which you speak?   Can
you put it explicitly.  

The misapprehension of which I speak is that natural selection *alone* gives
rise to innovation.  Without mutation, all that is achieved by natural
selection is a reduction of diversity in the genotype/phenotype toward some
"optimum" for the selection criteria, or more likely a "wandering" around
geno/pheno space as the selection pressures "wander".   I believe that this
is the mechanism behind what is known as "island dwarfism".   There is no
*innovation*, merely selection for a feature within the phenotypic
distribution (body size) already in the population.

I was NOT suggesting that YOU hold this misapprehension, just chiming in on
the point made by Jenny with her original quote.



And, when you say that mutations are "random", what precisely do you mean.

I don't know that *I* have said that mutations are "random".    I agree that
"random" is notional.  But I think of a signal as being "random" if the
receiver has no model to correlate it's structure.   A highly organized but
encrypted message is "random" if you don't have the key to decode it.
Cosmic radiation knocking holes in your genome is "random" for all practical
purposes, even if it is highly correlated with solar and magnetosphere
activity.   



  Unpredictable?  Clearly false.  We know quite a lot, I think, about where
DNA is vulnerable, and where mutations are likely to occur.  

A "random" selection can still have a statistical distribution.   When
rolling pairs of dice, there is only one way to get a value of 2, (both dies
== 1), 2 ways to get a value of 3 (1,2 and 2,1) and 3 ways to get a value of
4 (1,3 and 3,1 and 2,2), etc.   this distribution is defined by simple
combinatorics, but any given sample is still "random".   Referencing above,
in principle every specific set of dice are less than perfect and every
dice-thrower might have some "handedness" which *might* lend a tiny bias to
the distribution (e.g. LOADED dice).   The resulting sequences are still
random, just biased in an unexpected way.   Flipping a coin is the same
(unless it is two-headed of course!).

I don't think that the DNA (or intermediate RNA?) is more vulnerable in some
regions (or among some sequences) than others to say, "cosmic radiation" but
I will accept that perhaps when the many potential causes of mutation and
the various mechanism for detection/repair are taken into account, some
parts of the sequence are more susceptible to "effective" mutation?   And of
course, at the phenotypic level, what is "effective" is what the natural
selection component is all about.

I will pause beating this horse for a moment but will try to respond to the
remainder of your response separately (perhaps even completing the thought
you thought I failed to complete?)

- Steve

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://redfish.com/pipermail/friam_redfish.com/attachments/20170811/fa6d85bd/attachment.html>