NetNews Usenet Archive 1992 #16/NN_1992_16.iso/spool/sci/space/10960

Internet Message Format | 1992-07-29 | 7.2 KB

Path: sparky!uunet!mcsun!uknet!ieunet!tcdcs!unix1.tcd.ie!rwallace
From: [e-mail] (russell wallace)
Newsgroups: sci.space
Subject: Re: ETs and Radio
Message-ID: <[e-mail]>
Date: 29 Jul 92 20:33:55 GMT
References: <[e-mail]>
Sender: [e-mail] (NN required at ashe.cs.tcd.ie)
Distribution: sci
Organization: Trinity College, Dublin
Lines: 115
Nntp-Posting-Host: unix1.tcd.ie

In <[e-mail]> [e-mail] writes:
>The first part deals with an article in the May 1992 Scientific American, by
>John Horgan on page 30. The article discusses the work of Julius Rebek, Jr.,
>a chemist at MIT. Rebek has been developing molecules which self-replicate,
>much the same way that DNA does. He uses a simple system which duplicates
>some of the features of living systems. The system is so simple, that after
>a few steps, it ceases to demonstrate new qualities, but it points the way
>for developing evolutionary models based on non-living molecules, which
>could point the way to a better understanding of living systems.

>Here's how it works, in it's present version: three amines and an ester, all
>synthetic, are mixed in a chloroform solution. Each amine combines with an
>ester to produce an amide, each of the three different, which will self-
>replicate. That is, the amide will serve as a template for other amines and
>esters to cling to, and make another amide. Thermal jostling separates
>the two, and each goes on to produce more amides.

>The three slightly different amides replicate at roughly the same rate, but
>when irradiated with UV light, one amide mutates into a variant which
>reproduces much faster than the others.
>In another recent experiment, two esters and two amines were mixed to create
>four different amides. Two are duplicates of the earlier experiment, one is
>an even better replicant than the mutant, and one is sterile, that is, it
>cannot support replication at all.
>Unusually, each amide replicator can also serve as a template for the other
>amides, somewhat like the "hopeful monster" idea where new species would
>occasionally spring full-blown into being. However, due to the simplicity of
>the system, analogies to biological theory are still pretty weak.

As a matter of fact, they are nonexistent. What you have here is
*trivial* self-reproduction akin to the growth of salt crystals, not
non-trivial self-reproduction as required for life. This is because the
amide molecules do not store information about their own construction in
a blueprint which can be altered to create a different amide molecule,
and therefore they cannot serve as raw material for evolution. The
critical point is to get by random chance to a stage where evolution can
then take over, and the amide experiment does not do anything to fill
the gap.

>It has been mentioned already that a strict definition of life is still
>somewhat forthcoming, and that there is still a fuzzy border between the
>living and the nonliving.

No, there is not. A living entity is one which is capable of non-trivial
self-reproduction, and a non-living entity is one which is not. (Well,
there are fuzzy example, such as mules (which cannot reproduce
themselves, but which we regard as alive anyway), but these occur as the
products of systems which are fully alive. There is no fuzziness in the
boundary between a planet which supports life and one which does not.)

>Demonstrating that life-like structures can arise
>under primitive-earth conditions and that they can exhibit behavior similar
>to modern cells, to me, takes most of the wind out of the sails of those
>who contend that abiogenesis is improbable to the point of being impossible.
>We can demonstrate that structures can form which resemble fossil structures,
>we can demonstrate that these structures can perform many of the duties
>which modern cells need to perform, and we can demonstrate that many of the
>natural processes which produce these structures will produce similar
>structures under many conditions, meaning that although many types of processes
>may initiate life, those which go the farthest may all produce roughly the
>same product, leading towards a "molecular determinism" which standardized
>the form of the first life on earth, however many times it arose, and in
>however many places.

So what? However many globs of amino acids, lipids, RNA and whatever you
can produce in a test tube, you cannot produce a life form. Pointing to
globs of amino acids in a test tube and saying that this is close to
the spontaneous assembly of a life form is like pointing to a box of
10^6 transistors, jumbled up at random, and saying that this is very
close to the spontaneous assembly of a working computer. The point is
not the components, but their assembly into a working system.

>Proteinoid microspheres are near-proteins produced by polymerizing amino
>acids. A main contribution is from something called trifunctional amino acids
>which are found in such biologic-free samples as lunar rock, meteorites and
>bacteria-free terrestrial lava. The amino acids have informational and
>functional value. They order the formation of proteinoids and direct the
>structure and functions of the proteinoids themselves. The ordering, or non-
>randomness of the proteinoids is a function of the chemical structure of
>the amino acids, and does not require external ordering, direction through
>divine agencies, or extraterrestrial visitors. This has been demonstrated
>in the laboratory, and is the best argument against the "probabilistic"
>claims that life is a decendant of random processes.

[biochemical data deleted]

I'm not a biochemist, so I can't comment in detail on this information.
However, it would appear that nothing other than certain types of
chemical reactions have been demonstrated for proteinoid microspheres
(correct me if I'm wrong). This is irrelevant. The point is not that the
mere chemical components of living organisms are hard to form, but that
the system, composed of all the components arranged in the correct way,
is hard to form, because there are so many possible wrong ways for the
components to be arranged. (Similarly, because electricity and silicon
occur in nature does not imply that electronic computers do.)

>This implies that we are much closer to understanding the steps to cellular
>life than is commonly implied in popular accounts of abiogenic research, and
>demonstrates that the probabilistic models are missing a few assumptions. The
>main key here is that proteinoids should assemble themselves, and then
>conduct "molecular evolution" of structure and function, until at some point,
>they can be considered to be living cells.

Evolution before the existence of a life form is a contradiction in
terms. For Darwinian evolution to occur, there must be non-trivially
self-replicating entities, which store blueprints of themselves in an
information storage system, which can be mutated at random to create a
slightly different organism. A non-living entity does not have this, and
therefore by definition cannot undergo evolution (in the Darwinian sense
of a cumulative, directed series of changes).

--
"To summarize the summary of the summary: people are a problem"
Russell Wallace, Trinity College, Dublin
[e-mail]