Wednesday, May 14, 2008
Can't design science
From the RoundTable blog
Read the latest entries
John Gay
Gay, of Blacksburg, is a philosophy student at the College of William and Mary.
Linda Whitlock's May 9 commentary, "God, science not exclusive," suggests that science should be an orderly and rational investigation of the natural world, but that in practice it is an effort to explain all natural phenomena by natural causes only, which she says is an arbitrary presupposition that is foundational to the exclusion of intelligent design from scientific practice and instruction. While science's investigation of nature through natural causes (philosophers of science call this naturalism) is worth much discussion, it is not the only reason that intelligent design is branded unscientific.
The central thesis of modern intelligent design is this: There are methodological approaches which enable the detection of design in nature. This statement must be valid if intelligent design is to regard itself as a science -- an ID proponent's claim that some aspect of an organism is designed must be subjectable to rigorous tests that generate results that (with some semblance of clarity) either verify or falsify it. ID proponents know that science makes predictions that can be tested. Popular works on intelligent design like "Darwin's Black Box" propose certain design-detecting tests (irreducible complexity, specified complexity, etc.) and then apply these tests to the real world, theoretically yielding elements of nature that are designed.
In order to make one of these tests, we need a sound and clear definition of design. Of course, design stands in opposition to naturalistic origin, entailing a level of complexity or order high enough to preclude everything except supernatural intervention.
Remembering that science requires appeal to clear, testable claims, we realize that there are two sorts of tests we need to make in order to detect design. First, we must be able to detect order within the chaos of nature. Second, having found examples of order, we need to be able to clearly distinguish between orders of natural and unnatural origin. Both of these propositions require strains of science.
Order is a vague concept -- it seems to be not just the appearance of a statistically improbable pattern but an assemblage of such patterns into a meaningful higher pattern (for example, one of the components of a cell might be taken as a pattern in comparison to a bunch of random molecules, and then these components add up to form cells, which add up to form organisms, which add up into species, societies, etc.).
It is of no help that the human mind often naturally assumes order in an ambiguous situation. At an elementary level, the gambler at the roulette wheel, seeing the ball land on red several times in succession, may put a bet for the next spin on black, since it is due, when in fact the pattern of reds was nothing but chance. Scientists are not immune to this by any means: For example, the astronomers Kepler (a giant in the field) and Bode saw extensive and complicated mathematical orders in the relationships between the sizes of planetary orbits, but the discoveries of new planets in places that didn't fit their pattern would later show them to be seeing patterns where none existed.
Respected, innovative scientists like Stuart Kauffman have done extensive work involving order, and there are plausible mathematical definitions of order, but these work much better in the simplicity of computer models than in the jumbled real world.
If order itself is difficult to define (outside those cut-and-dry computer models), and claims of order often turn out to be erroneous, claims of unambiguous detections of a very specific kind of order (design) seem almost silly. This feeling is magnified when we consider that many ID proponents accept some form of natural selection, which itself can generate order well above the level of chance, and that they think their method can clearly spot instances where selected order is not present.
Even if we were, in the face of all this, to accept the claim that design is detectable, the current models promoted by design proponents like Michael Behe and William Dembski generate examples of design (like the bacterial flagellum) that in fact show substantial evidence of natural origin, i.e. nondesign.
The central thesis of the intelligent design movement is thus apparently quite false: Not only do the proposed tests for design yield false predictions, it is extremely difficult (if not impossible) to create a rigorous test for design in the first place. Intelligent design is thus exposed as unscientific without any appeal to naturalism.
