Why I want to write a book on Xenobiology.

[swestrup]

Its because there are idiots like the author of this article, who think they know something about the subject.

At first I was thinking I would write a long diatribe here describing in great detail all that the man is not aware of, but I realized that discussion of convergent evolution would fill an entire chapter of the book. So, I'm just going to make some bullet points:

Convergent evolution can happen, but its a far muddier issue than many think because:

• Evolution conserves DNA. Sometimes 'convergence' is just reexpression.
• Whenever you have a diverse group of anything, you can find subgroups that more closely resemble each other than they do the other members. This proves nothing.
• As was so strongly pointed out recently, the ability to evolve has evolved! This has implications on what body plans are easy to generate given the control structures that have evolved. If you assume different control structures (and they are essentially random) then different body plans are easier to evolve towards.
• Related to the last point is that the evolution of evolutionary strategies means that the historic evolutionary path of an organism matters in deciding what is easy or hard for it to adapt to. In a similar way, if you take a dozen engineers with widely varying backgrounds and ask them to design an object that meets certain constraints, they will invent wildly varying objects. Give the same task to a dozen engineers with very similar backgrounds, and they will produce much more similar objects. This is one of the main reasons that biodiversity turns out to be so important for ecosystem adaptability.
• The range of environments on Earth are constrained, not only by the fact that it is here, and not somewhere else, but because life modifies its environment. The result is a world more benign and friendly (for some values of 'benign' and 'friendly') than it would be if it were lifeless. It also means living conditions are clustered around a set-point chosen during a random walk. Some convergence under these conditions is expected. Change the set-point though, and the convergences happen else where in the design space. On other planets, they WILL have the set-points elsewhere.

Comments

[sps.livejournal.com]

"What if the only solutions to the problem of producing an organism capable of interstellar communication were in the DNA combinations resulting in bilateral symmetry, four limbs, opposable thumbs and brains more proficient than is required for the health and reproduction of the species?" he writes- and, by the same reasoning, what if gravity is caused by monkeys flying out of each other's butts too fast for time to notice?

The single most amazing thing about this article, I think, is that he gives the dolphin and the octopus as two of his examples of convergent evolution (dolphin convergent with shark, octopus with human(!)), but somehow misses the immediately adjacent point that our closest intellectual rivals on this planet look like a frickin' dolphin and a bloody octopus - and these start out using the same construction kit as us, not just the DNA and the amino acids (and what do they have to do with life-not-as-we-know-it?) but the same physiological and neural toolkit - and in the case of the dolphin, the same metabolic toolkit, reproductive mechanism, ancestral bodyplan, you name it.

What more evidence do you need that intelligence isn't bound to hands, feet and faces? Whatever's out there, if it's as similar to us as an octopus, that establishes a presumption of common origin. Because if the ape is your cousin, the octopus is your mother in law.

I wish you hadn't wasted our time with the annoying link ;).

[sps.livejournal.com]

PS Nancy points out that all is futile and there's no point in doing anything.