Book Review: What is Life?

[pic link]

This is not Schrodinger's What is Life? from before the DNA structure was solved. (The phrase is also used all over the 'net from the diversity of a quick Google search.) It is Lynn Margulis and Dorion Sagan's What is Life? from the mid-90's. (Yes, it's that Margulis, the famous one who proposed the endosymbiotic theory of mitochondria, but no, it's not that Sagan, it's his son!)
.

It's also an example of the blessed serendipity of the semi-annual Library Book Sale. This year I went to the book sale and found several books on origins (of life and/or the universe) that I was looking for, for a dollar each: the book by Martin Rees I reviewed earlier and this one topped the list. This is a beautiful edition, big pages with full-color pictures of microbes and rocks. It was deliberately formatted after the "old science books" of earlier centuries, and even has some full-page hand-drawn biological specimens like those old books had. It feels like an old book, and I love that about it.

.

The topic of the book is a history of life told by Margulis (I gather she's the driving editing force, what with the book's emphasis on symbiosis and reproduction). What I like about it is that the thermodynamics of entropy and energy dissipation is pretty much there, and the vignettes about certain bizarre microbes are wonderful. Well worth it just for those. It complements some of the other stories, like the RJP Williams one I keep going on and on about because so few people can find it!

.

Where it goes off the rails in my opinion (and since this is a book review it's all my opinion but let's just re-emphasize) is the final chapters in which man is discussed. The authors argue that Darwinian reductionism and directionlessness is wrong. They bring up Samuel Butler's ideas and expand on them to suggest that there is a direction to evolution and that direction is guided by "choice" -- a property that everything living (able to respond to its environment, I would reduce that to) shares. So this is the "microbiological existentialist's" version of creation. I'm still trying to figure out exactly what this "choice" is and where it comes from. I can't get away from the worry that at its heart it's just randomness magnified, whether a certain environmental element is sensed or not, that organism survives, and goes on to reproduce. That's the "choice" they're talking about, and is it any different from a computer's "choice" if a subroutine that ranks options and takes the best one is included, maybe with a random Monte-Carlo element built in? It's just a question of how much randomness happens. I'll have to think about it more but I can't find any way out of that mess, and I don't see how it's fundamentally different from the position they say is wrong. The central question is: Where does the "choice" come from?

.

I see a parallel here with Signature of the Cell by Meyer (although this is definitely debatable and incomplete): in both cases we have a fuzzy concept that's given a quantifiable backdrop, but the core of the concept is not correctly quantified (not for me at least!). Meyer gives probabilistic calculations of "design" that calculate these huge astronomical numbers, but if the starting assumptions are off the numbers don't matter. "Choice" in What is Life? has a similar role, although the authors stay away from quantifying what "choice" is, they argue as if it were quantified or a clearly defined part of the system. But the best I can gather for its definition -- if it is environmental sensing and response -- is that the only "surprising" part of choice would be random error in either sensing or response. That could even improve the system, sure, but it's not really any less "random" than the Darwinism you're arguing against in the first place. Meyer's calculations assume/quantify too much, and Margulis and Sagan's don't seem to assume enough (or they assume their "choice" is different from selection when it seems to me to be random noise).

.

Of course, this is off the cuff and I have to think about it more to tamp it down, but I'm not satisfied with either "design" calculations or "choice" philosophies.

.

The big difference with starting with chemistry (the approach that appeals to me) is that we start with quantifiable terms: energy, entropy, and enthalpy. (The complexity is in the APPLICATION of those numbers to the system!) The sun provides an excess of energy to the system that the system converts. On the individual level there's a lot of variation. But the way energy moves gives a direction -- perhaps contingent but a direction nonetheless -- to the system, suggesting the universe is "set up" to do this, to have this result.

.

Again, off the cuff. I'm probably going to regret the shortcuts in some of these statements, but that's kind of the point. I slap this online as a rough draft and you, dear readers, help me find the holes in it. There will be holes, oh yes.

.

So the concept of "choice" that the authors put forward, I "choose" not to integrate it into my thinking.

.

And this makes some of the more outlandish statements in the final chapters -- that science is creating a story that will replace the old time religion/creation myths, that man's capacity for self-deception is what sets us apart from animals, etc. -- it makes them ring hollow because I don't know what this "choice" thing is. An emergent property? How is that different from randomness? If self-deception is the key to humanity then why are we here in such a perverse universe? And why is the universe knowable at all if self-deception is our foundation? (Seems like cutting off the branch you're sitting on ... )

.

Regardless, this book is well worth it for its beautiful pictures and stories. Just make the "choice" to complete your own synthesis of those and take the final chapters with a big grain of salt.