Monday, February 16, 2009

Wells on Charles

Dr. Jon Wells, author of Icons of Evolution, has an incisive commentary on Darwinism for Darwin's 200th birthday. The main point is that Darwinism (not an ancient earth or evolution as differential population changes within species over time) is more materialist philosophy than verifiable science. The more thoughtful Darwinists even admit this, as he notes.

1 comment:

Jeremy said...

First, Wells's piece may place too much blame on Darwin. It is just not the case that Darwin alone shifted science to naturalism. For example, we see a denial of our ability to discover final causes in Descartes' Meditations (cf. either the 5th or 6th Meditation). Descartes then suggested that science forget teleology.

That Darwin was already situated in an environment friendly to naturalism is shown by Well's quote from Neal C. Gillespie: "it is sometimes said that Darwin converted the scientific world to evolution by showing them the process by which it had occurred," but "it was more Darwin's insistence on totally natural explanations than on natural selection that won their adherence." Darwin gave the scientific community what it wanted; he did not force them into naturalism.

Second, Wells presents a caricature of Darwin's "one long argument." Wells writes:

"Darwin called 'The Origin of Species' 'one long argument,' and it took the following form: The features of living things are "inexplicable on the theory of creation" but fully explicable as products of unguided natural forces. Darwin lacked sufficient evidence for the latter, however, so he ruled out the former by simply declaring that only natural explanations are 'scientific.'"

This is not at all accurate. The Origin certainly is one long argument, but it is divided into two parts. The first part is an argument from analogy: artificial selection drives variation among animal populations bred by humans. This artificial selection is quick given its dependency on agency. However, nature will select for traits that yield adaptive benefits much like a shepherd would select dogs to breed that had desirable sheep-dog traits. Natural selection is slow because it is not driven by agency. But regardless of how slow the process is, the same amount of variation is possible in both cases--all you need is heritability and selective pressure. Given enough time mere variation becomes speciation. And while a single common ancestor is not necessarily entailed by evolutionary theory, it is no far explanatory leap to posit universal common descent.

So, the first part, the argument from analogy establishes the *mechanism* by which speciation takes place. No one denies that it is a naturalistic explanation for speciation (it is not at all an explanation of the origin of life). Even so, there is no reason to suppose that evolutionary theory should be thrown out of court because it is a naturalistic explanation. But it shouldn't therefore be accepted on the grounds that it is naturalistic either. Rather, it should be evaluated in terms of explanatory success (think virtues like explanatory power, simplicity, plausibility, unification, lack of ad hoc premises, etc.). And it should be evaluated in comparison with any other known rival theory. This is where the second part comes in. Darwin systematically works through some of the issues that were most troubling to 19th century scientists, and compares how his theory explains/predicts these issues versus how the rival hypothesis that God specially created each species explains/predicts these issues. Darwin suggested that his explanation was superior.

But there is nothing problematic here. We all use explanatory reasoning, and we use it more than straight deduction. (As far as straight induction is concerned, G. Harman's first article on inference to the best explanation shows how induction is a special case of IBE). Anytime we have two possible explanations (e.g., the pudding on the wall is either explained by a mild earthquake shaking the refrigerator open just as the pudding cup spontaneously explodes, shooting pudding all over, or it is explained by my son, who was just eating pudding out of a pudding cup, got bored and decided to "paint.") we compare the stories and pick the best. To the extent that Darwin characterized the rival theory correctly (which may be an interesting question in its own right), Darwin did not beg any questions against the creationist. Darwin never made the move that biologists make today (the move Wells accuses Darwin of making)--non-natural explanations are just not science, and so any explanation that entails agency is ruled out a priori! Therefore, I am inclined to think Wells's account of Darwinism is wrong.

Now, I'm no friend of Darwin. But the ID program has enough problems without making things worse by mischaracterizing rival theories. Responsible exposition is a necessary condition for serious consideration by the scientific community. Of course, it is not sufficient. We've got to deliver the empirical goods too. I suggest that this is where ID researchers ought to focus. We have to make clear exposition of our rivals and be staunch advocates of explanatory reasoning. But, the data ID appeals to as evidence that ID is probably correct simply underdetermines ID and evolutionary theory. We must be bold enough to make predictions that could be implications of only ID (or the prediction could induce ad hoc changes to evolutionary theory). We can't be content to let the debate rest at the level of philosophy of science, arguing over explanatory virtues (although that's REALLY important).