Reflectorites
On Mon, 16 Oct 2000 13:25:31 +0100, Richard Wein wrote:
[...]
>CC>Thanks, Steve.
My pleasure!
>CC>I'll look it over. I see a general problem with it already, in that many
>>of the items listed should *evolve* from others, and should not be included
>>as a basic part of the simulation.
What is this word "evolve"? The 47 things I have listed AFAIK are
*essential* to all known life. Chris can of course *imagine* that life could
start without them, but it is just his imagination, not biological reality.
CC>For example, a really good simulation of
>>evolution would start "at the beginning," with nothing more than a
>>simulation of suitable physics and chemistry, and the rest would all be
>>evolved from that starting point.
This begs the whole question that it *was* "the beginning".
But I would be happy for anyone to show by a simulation that starting
"with nothing more than a simulation of suitable physics and chemistry"
that "the rest would all be evolved from that starting point"!
CC>Nevertheless, I think the list is
>>valuable and probably could be extended.
Maybe. But these 47 will do for now!
>CC>However, much about evolution can be studied *without* many of
these
>>features.
RW>Yes. It's in the nature of a simulation that it's based on an abstracted
>model of reality. For total accuracy, we would have to simulate every
>sub-atomic particle, which is clearly impossible.
Not really. In the case of the origin of life, starting with amino and nucleic
acids will do fine.
RW>The point of a simulation
>is to model the significant features of the system in question while
>abstracting out the insignificant details. That always leaves the
>possibility that we have omitted some feature that we don't know about or
>consider to be insignificant, but which in reality *is* significant. But
>that's life. Science is not perfect.
I have listed 47 things that are essential for life as we know it. Any
simulation that does not faithfully represent those, is not biologically
realistic.
>CC>For example, one claim is that random processes, even repeated
>>and cumulative, cannot produce complexity.
The claim is *specified* complexity. Specified complexity is like
meaningful words on a page. A test of specified complexity is whether a
computer program to produce the string of characters would be shorter
than the string. In the case of (say) the play Hamlet, the program to
produce it would be longer that the play itself.
CC>This is not a specific claim
>>about DNA-based evolution, but much more general claim. It can therefore be
>>studied with simulations that do not represent DNA at all, but merely a
>>process that demonstrates whether complexity *can* be accumulated by this
>>kind of process. Similar considerations apply to selection.
I would be interested in any simulations that can generate specified
complexity, without: 1) starting with specified complexity; or 2) smuggling
in specified complexity during the course of the simulation.
And *please* don't give Dawkins' METHINKS IT IS LIKE A WEASEL
delusion! On second thoughts, please *do*! [Stop Press: I since read
Brian's post on this alleged creationist strawman and am not changing
a word. I have some interesting reactions of *evolutionists* at the time
in the pages of New Scientist where Dawkins first floated the idea.]
BTW, there is a good test of this. If unintelligent natural processes could
generate specified complexity then supercomputers could write books.
Dawkins' METHINKS IT IS LIKE A WEASEL program could write
Hamlet if the entire play was encoded in its memory, (just like it could
write a line of Hamlet if the line was encoded in its memory).
[...]
Steve
--------------------------------------------------------------------------
"Can we separate 'pattern' from 'process'? Every taxonomist since Darwin
has interpreted life as a vast tree in which all living creatures are the tips of
the branches, and fossils are the remains of ancestral branches. So the
pattern of nature-the forms that exist now and in the past-has been
interpreted in terms of the process of nature, the theory of branching
evolution through time. But has this assumption clouded our vision of
nature? Can we be certain that a particular fossil, which may appear to be
intermediate between other creatures, is really an ancestor? With the
growing sophistication of taxonomy there is a feeling that many of the neo-
Darwinian assumptions about fossils and ancestry may be scientifically
unfounded, and should be dropped. This realization, that the theory may be
incapable of helping taxonomy, and may even be a hindrance to it, has led
to a rejection of Darwinian ideas among some taxonomists who feel that
we should be finding out more about the pattern before we become
dogmatic about the process which is supposed to explain it." (Leith B.,
"The Descent of Darwin: A Handbook of Doubts about Darwinism,"
Collins: London, 1982, p.23)
Stephen E. Jones | Ph. +61 8 9448 7439 | http://www.iinet.net.au/~sejones
--------------------------------------------------------------------------
This archive was generated by hypermail 2b29 : Thu Oct 19 2000 - 20:37:18 EDT