What's sauce for the goose is sauce for the gander (or something like that...)
In The Panda's Thumb, Stephen Jay Gould worked hard to show that less-than-ideal design in nature is an argument against a divine designer and an argument for evolution (Dawkins' "blind watchmaker"). I think his argument is horribly circular, and so is quite invalid, though it is often rhetorically effective. My claim, on the other hand, is that less than ideal design is just as much an argument against evolution. Here's why:
Dembski, in arguing against the need for larger brains for human intelligence in The Design of Life, made note of the evolutionists' answer to the problem of anomalies in relation to brain size - that the brain must contain tremendous amounts of redundancy. He then made the following counter-argument:
If the brain is redundant, then why didn't we evolve the same cognitive abilities without developing larger brains? Redundancy carries hidden costs. Big brains make it difficult for humans babies to pass through the birth canal, which, historically, has resulted in heavy casualties - many mothers and babies have died during delivery. Why should the selective advantage of bigger brains with lots of redundancy outweigh the selective advantage of easier births due to smaller brains that, nonetheless, exercise the same cognitive functions, though with lowered redundancy?(p. 12)
The point is a good one, and not only in response to the specific argument in the text. This "design" is less than ideal enough to argue against the likelihood of its evolutionary origin. Natural selection would have weeded out such a considerable cause of high infant mortality long before it became the norm for our species.
There are many examples touted by the mainstream bioscience community as evidence of the "tinkering" of evolution rather than design by immeasurable intelligence. One problem seems to be that there are examples of poor and harmful design that seem statistically incompatible with the proposed evolutionary process. Proponents of evolution sometimes even use these (ironically) as evidence against a creator. In fact, I've heard the "big head vs. small birth canal" used in just this way before.
An adaptive, "tinkering" approach to design as an answer to such things as, say, Gould's panda thumb, seems a lot more rational to me than holding the statistical demands of evolution in tension with obvious counterexamples.
Am I on to something here, or am I missing something?