« Putin 1, Obama 0 | Main | »

December 16, 2008

Comments

Nato

"while the physicist learns many more things in the same fashion, his general approach need never depart from common sense"

I don't think this is true, based on my various attempts to explain how physicists "know" various things they generally regard themselves as "knowing." There's a great deal of trying to find a mathematical story that accords with all the facts, some of which are not, in fact, directly observable in an unproblematic sense. This can go for chemistry as well, really. Basically, it can take a great deal of training to grasp the deductive steps the field takes to say that some fact "must" be the case; this interferometric pattern has to mean that DNA is a double-helix, that background radiation has to mean that the early Universe had a certain density and shape. There are constant announcements about having "found" black holes, despite having never created them in labs.

What's different about "Darwinism" is that it's a simple hypothesis and a full theory at the same time. That is, it is a general project of trying to find a Darwinian explanation for every form of life (or thing substantially *like* life) at the same time as it's a body of (Darwinian) explanations of specific historical data points. The former is in a sense unfalsifiable, since some recursion of Darwinian explanation is always available in any circumstance (so we have conclusive evidence that aliens made the squirrel? Well, perhaps the *aliens* evolved!), while specific Darwinian explanations are quite falsifiable (the squirrel can't have evolved naturally from other rodents - we found alien genes in them!)

There are, of course, plenty of Darwinian biologists who want to insist that all living things evolved from nonliving things through natural processes because they want to reject suggestions that some historical events are scientifically insoluble. They don't, in general, have the humility and philosophical background necessary to moderate their claims thusly: "Yes, other events could account for the evidence we have now and will discover in the future, but as a scientist I'm limiting myself to the kinds of explanation that science could elucidate."

Nathan Smith

Well, first, the lack of "humility and philosophical background" could probably be remedied in biology PhD programs pretty easily if it weren't encouraged by people like Daniel Dennett and by a bias in the education system enforced by judicial rulings against the will of the people, taboos in polite society on evolution-skepticism, and so on. Second, the identification of evolutionary explanations with "the kinds of explanation that science could elucidate" is not valid: future science-- or even philosophy!-- might elucidate forms of causation and explanation of phenomena not currently conceived of, and which do not lend themselves to classification alongside the theory of evolution. Also, my impression is that particular stories within evolutionary biology tend to be vitiated by presupposing the truth of evolutionary theory as a whole. Thus, if we observe that Species X has feature B in 1 million B.C., while (what is somehow inferred to be) its descendant Species Y has feature C in 800,000 B.C., a biologist will think up just-so stories about how feature B evolved into feature C, and pick the most plausible, "knowing" that *some* evolutionary account must be the right one. But the deeper problem is that we really have very little idea just how powerful evolutionary mechanisms really are, or are not, in doing Design work over time. We can't establish it by experiment except perhaps in very rare cases, maybe bacteria that evolve super-fast. Archeology cannot come close to showing all the links in the causal chain; indeed, the interpretation of the archeological record tends to be heavily dependent on the prior assumption of the truth of evolution, and without this assumption what we could deduce might be rather limited. Our best bet might be computational simulations, which can illuminate evolution to some extent, but those come nowhere close to matching the complexity of the evolution of life right now, and probably never will. But I think much particular research is dependent on the macro assumptions of evolution, such that, if in a computer simulation one could demonstrate that dynamic change can be driven by transient "genes" that spread rapidly through a population and then destroy themselves with no trace, having transformed the system in the meantime, many of the conclusions of workaday evolutionary biology would have to be abandoned.

It's true that physics soon becomes highly mathematicized, and, like economics, I suppose it sometimes becomes so abstracted from its subject matter that it might seem more like pure mathematics than a study of anything real. Whether that's a departure from common sense, I'm not sure: there is in common sense a certain amount of arithmetic, as well as pattern-recognition that is implicitly statistical. I don't think evolution departs from common sense through similar mathematicization. Rather, a vast labor of abstract dynamic systems analysis, much exceeding the complexity of mathematical physics, would need to be executed before we could properly assume that the possibility of all life on earth emerging through evolution is not prohibitively improbable. Even if this were done it would do nothing to explain how evolution could, say, give rise to consciousness.

This is not necessary in practice for evolutionary theorists to make their theory convincing, for the same reason that no proper evidence for the book of Genesis was necessary to make generations of Christians believe that story. The reason is that human beings do not like to have gaps in their knowledge and so are credulous of creation-myths. But we rationalistic freethinkers try not to indulge in this weakness.

Nato Powell

"a vast labor of abstract dynamic systems analysis, much exceeding the complexity of mathematical physics, would need to be executed before we could properly assume that the possibility of all life on earth emerging through evolution is not prohibitively improbable"

So the default assumption should be that all life on earth emerging through evolution is prohibitively improbable? As opposed to what?

Tom

Actually, probability theory informs us that any particular sequence of outcomes is equally unlikely. In other words, the sequences of events where Humans do evolve from lifeless matter are just as unlikely as the sequences of events where Humans don't evolve from lifeless matter. You are as likely to get heads 10,000 times in a row on an unbiased coin-flip as you are to get 9,999 heads and then a tail, or 9,998 heads and then a tail and a head. Every permutation of heads-tails sequences is equally (un)likely. As the number of events 'n' approaches infinity, the probability of any particular sequence of events goes to zero (but the sum of all of the probabilities equals 1). So the probability of our universe, given no other information, is zero. On the other hand, the probability of our universe, given basically any information (ie, empirical observation), is one.

Nato

I think the probability of things eventuating as they have in the most global, (multi) Universal sense is probably one. That's not a very interesting kind of probability, though.

Nathan Smith

The probability of the universe being the way it is, given that it is the way it is, is one. But there's a lot we don't know about how the universe is. We have to guess and theorize, and collect and interpret evidence. When deciding between guesses, or deciding how much confidence to have in any guess, *antecedent* probability is an important factor.

Tom

"I think the probability of things eventuating as they have in the most global, (multi) Universal sense is probably one. That's not a very interesting kind of probability, though."

It's the identity axiom, "A = A". True, it's not very interesting; it's a logical statement without much content. But one could still say that the probability that the universe could have been other than what it is or has become is zero. If that's not the case, I think there would be severe problems for whatever epistemology one subscribes to.

The comments to this entry are closed.

My Photo

Only use a payday cash advance as a last resort.

Categories

Blog powered by Typepad