genetic information (was Re: Computation and Human Experience(RRE)

Brett Knowlton brettk at unica-usa.com
Tue Jun 20 15:23:00 PDT 2000


Ted,


>>But organisms can be engineered. I read something very recently about how
>>tampering with sheep genes can lead them to produce spider silk in the
>>sheep's milk. This might not be engineering the entire organism, but it is
>>a kind of engineering. Modify the genes of an organism to obtain a desired
>>characteristic or ability. Isn't that engineering?
>
>Not if you obtain the result through a method which is entirely alien to
>engineering.
>
>Engineering is essentially applied mathematics. It involves the
>manipulation of passive components which have very predictable patterns of
>chemical "behavior" over time. It's all very precise and orderly. Genetic
>manipulation, however, is mostly intuitive and trial-and-error. Not
>surprisingly, it's more like gardening or herding or even sex. When you're
>dealing with something that's alive-- that's got its own will-- you're not
>going to have a lot of success with equations and slide rules. It's a
>different kind of "chemistry" altogether. If people commonly spoke of
>"engineering" orgasms, then I would accept the notion of engineering
>organisms.

I disagree with your characterization of engineering. There is plenty of trial-and-error and intuition in the engineering field. Lots are problems are solved this way rather than getting out the calculator and solving a bunch of differential equations (or some other precise methodology).

You simply have to rely on the fact that, once you get something to work once, it will continue to work that way. This is the reliance on predictability you mention above. And organisms are predictable. A fertilized human egg implanted into the womb will always yield a human baby (never a puppy or a chicken or a gold coin). You might not understand exactly why it works out that way, but it is clear that it does.

Same thing with the sheep's milk yielding spider silk (I think it might actually be goats, not sheep, but I'm not sure. I'll have to look up the article). You have a process that, if followed, grows sheep/goats which will always produce spider silk in their milk. Keep in mind, this particular result wasn't achieved willy nilly. Spider genes were somehow put into the sheep/goat genome, so the idea that spider genes are responsible for the spider's production of silk seems to be spectacularly verified. So, the researchers had enough knowledge to achieve the desired result, even if they don't have full knowledge of every last detail.

Now, you might not be able to determine in advance whether the sheep/goats will have an ill temper or not (or maybe we will be able to determine this some day, I don't know). As you say, living things aren't necessarily entirely deterministic. But that doesn't mean everything is out of bounds either. Just because you can't create an organism from scratch doesn't mean you can't tinker with one and change certain characteristics repeatably and reliably using the partial understanding you do have.


>>Organisms _are_ machines. They may not be mass produced on an assembly
>>line (at least not yet), but I don't see why you can't describe organisms
>>as machines. To take humans as an example, you can even identify the
>>various pieces which keep us working - we have pumps (the heart), filters
>>(the kidneys), actuators (muscles), etc.
>
>Of course organisms utilize mechanisms, both chemical and physical.
>Organisms will take advantage of anything that presents itself in their
>internal and external environments. But this doesn't mean they are
>essentially machines. If they were machines, we should be able to see how
>they work from top to bottom. We ought to be able to see how it is that an
>embryo is converted into an adult. Where's the chain of mechanisms that
>transforms cells into bodies? What tells a rabbit embryo to become a rabbit
>and a chicken embryo to become a chicken? We don't know. It used to be
>thought that genes encode the relevant information, but this idea turns out
>to be implausible. Now it's thought that the process is determined through
>the interaction of genes and master proteins. According to Enrico Coen, a
>mainstream molecular biologist, this process is inherently unmechanistic.
>He thinks the relevant macromolecules are literally creative-- in the same
>sense that humans are creative! That this could be posited by a serious
>researcher just goes to show how desperate the situation is becoming. The
>obvious solution-- though unacceptable to materialists-- is that
>species-memory, which guides the developing organism, is not "encoded" in
>matter. (That is, mind is not a function of matter.)

Just because you don't know how a machine works in every respect doesn't make it any less a machine. Perhaps we're dealing with semantics here, but if you transported an internal combustion engine back to 10,000 BC, I'm sure the people there would be baffled by it. Would that mean it would cease to be a machine?

As far as embryonic development is concerned, I'm sure it is true that the entire trajectory from fertilized egg to realized organism is not encoded in the DNA. But that doesn't mean that there is some extra-material force involved. A fertilized rabbit egg won't grow into a rabbit if you put it in a petry dish. The embryo needs to be in a particular environment, and it is quite possible that the interplay between the environment and the genes are required to realize the infant. But that leaves development open to a materialist explanation.


>Where is the information storage in the brain to account for personal
>memory? Where is the image-machine to account for vision? Where are the
>calculations being carried out which enable us to move about in our
>environments? Is the brain so stupid that-- like a goddamn computer-- it
>has to perform thousands of incredibly complex equations just to enable you
>to walk down a hallway without bumping into a wall? The idea that organisms
>are machines is merely assumed. It's assumed that there cannot be anything
>real which isn't reducible to matter. But is this really a sound
>assumption?

The reason for the materialist assumption is that you simply can't study anything without making that assumption. Saying that a phenomenon can't be explained by materialism means it can never be explained, so there is no point in even studying the subject.

Let's assume there is some other non-materialist explanation for how organisms develop, and for how the mind works. What might that be? Your guess is as good as mine, and since it is extra-material, we'll never be able to answer the question. It will forever be a mystery.


>Matter is weird.

And so weird phenomenon can be explained via materialism. I see no reason why we should give up on a materialist explanation of the mind or species-memory just yet.

Brett



More information about the lbo-talk mailing list